Microsoft Word - SDPD Vehicle Stops (Nov. 22)

Traffic'enforcement'in'San'Diego,'California'

An'analysis'of'SDPD'vehicle'stops'in'2014'and'2015'

Joshua'Chanin,'Meg an'Welsh,'Dana'Nurge,'an d'Stu art'H e nry ''''''''''''''''''''''''''''''''''''

San'Diego'State'University''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''

November'2016'

TABLE!OF!CONTENTS!

EXECUTIVE!SUMMARY!..................................................................................................................!ii!

ACKNOWLEDGEMENTS!...............................................................................................................!vii!

LIST!OF!TABLES!...........................................................................................................................!viii!

LIST!OF!FIGURES!...........................................................................................................................!xi!

CHAPTERS!

1.!!Introduction!.................................................................................................................!1!

2.!!Policing!in!San!Diego!....................................................................................................!3!

3.!!Description!of!the!Data!.............................................................................................!14!

4.!!The!Decision!to!Initiate!a!Traffic!Stop!........................................................................!26!

5.!!Evaluating!Post-Stop!Outcomes!................................................................................!48!

6.!!Summary!and!Recommendations!................................................................ ............. !68!

APPENDICES!!

1.!!Detailed!data!on!SDPD!staffing!and!crime!in!San!Diego!............................................!93!

2.!!The!San!Diego!Police!Department!Vehicle!Stop!Data!Card!.......................................!95!

3.!!SDPD!Officer!Survey!..................................................................................................!96!

4.!!Limiting!the!veil!of!darkness!analysis!to!stops!involving!moving!violations!............!104!

5.!!Limiting!the!veil!of!darkness!analysis!to!stops!involving!male!drivers!....................!108!

6.!!Division-level!traffic!stop!patterns,!by!year!.............................................................!112!

7.!!Using!logistic!regression!to!model!post-stop!outcomes!..........................................!118!

8.!!Describing!matched!and!unmatched!drivers!...........................................................! 121!

9.!!Modeling!driver!hit!rates!after!dropping!missing!contraband!cases!.......................!127!

10.!!Modeling!driver!hit!rates!after!dropping!missing!contraband!cases!..................... !128!

11.!!SDPD!officer!training!.............................................................................................!129!

! !

ii!

EXECUTIVE!SUMMARY!

This! Report! presents! the! results! of! an! independent! analysis! of! records! generated! following!

259,569!traffic!stops!initiated!by! S an!Diego!Police!Department!(SDPD)!officers!between!January!

1,!2014!and!December!31,!2015.!This!review!focused!on!the!extent!to!which!these!data!reveal!

Department-! and! division-level! racial/ethnic! disparities! in! (1)! the! decision! to! initiate! a! traffic!

stop;!(2)!the!decision!to!issue!a!citation;!(3)!the!decision!to!conduct!a!field!interview;!(4)!the!

decision!to!initiate!a!search;!(5)!the!disco very!of!contraband;!and!(6)!the!decision!to!make!an!

arrest.!Our!findings!can!be!summarized!as!follows:!

• Citywide,!disparities!between!Black!and!White!drivers!were!evident!in!vehicle!stop!data!

from!2014,!but!not!2015!or!the!combined!2014/2015!dataset,!while!no!such!disparities!

were!found!between!Whites!and!either!Hispanic!or!Asian/Pacific!Islander!(API)!drivers!in!

2014!or!2015;!

• Data!from!both!2014!and!2015!revealed!distinct!and!divergent!stop!patterns!by!driver!

race/ethnicity!in!police!divisions!located!above!and!below!Interstate!8;!

• Citywide! and! across! 2014! and!2015,! Black!and! Hispanic! drivers! were! more!likely! than!

White!drivers!to! be!searched!following!a!traffic!stop,!and ! despite! facing!higher!search!

rates,!were!less!likely!to!be!found!with!contraband;!!

• Black,!Hispanic,!and!API!drivers!were!subject!to!field!interviews!at!greater!rates!than!

White!drivers;!!

• No!meaningful!difference!existed!in!the!rate!at!which!drivers!from!each!racial/ethnic!

group!were!arrested;!!

• Black!drivers!were!less!likely!to!receive!a!citation!than!White!drivers!stopped!under!

similar!circumstances,!while!matched!Hispanic,!White,!and!API!drivers!were!cited!at!

similar!rates;!!

• Records! of! traffic! stops! conducted! in! 2014! and! 2015! were! often! incomplete,! raising!

questions!as!to!whether!data!generated!by!the!SDPD’s!traffic!stop!data!card!system!are!

a!reliable!measure!of!actual!traffic!stops!conducted;!and!!

• City!residents!who!participated!in!our!focus!groups!and!SDPD!officers!who!participated!

in! an! electronic! survey! and! follow-up! interviews! recognized! a! tension! between! the!

Department!and!minority!community!members.!

The! remainder! of! this! executive! summary! provides! an! overview! of! the! data! and! analytic!

methods!used!to!examine!traffic!stops!and!post-stop!outcomes,!a!more!detailed!review!of!our!

findings,!and!a!brief!description!of!our!recommendations!to!the!SDPD!to!address!the!identified!

racial/ethnic!disparities.!!

iii!

Traffic!stops!

To!examine!the!effect!that!driver!race/ethnicity!has!on!the!likelihood!that!an!individual!will!be!

stopped! by!the!police,!we!draw!o n!what!has!become!known! as!the!‘veil!of!darkness’!techni qu e.!

This!approach!is!premised!on!the!assumption!that!if!officers!are!relyin g!on!driver!race/ethnicity!

to! guide! stop! decisions,! then! such! bias! will ! be! more! apparent! in! daylight! stops,! when! a!

motorist’s! race/ethnicity! is! more! likely! to! be! visible,! than! stops! conducted! after! dark,! when!

physical!appearance!is!harder!to!detect.!

The!veil!of!darkness!technique,!which!thus!far!has!been!used!by!police!scholars!to!study!traffic!

stops! in! six! other! U.S.! locations,! allows! researchers! to! avoid! the! difficulty! of! identifying! and!

applying!a!benchmark!against!which!to!compare!traffic!stop!data.!This!is !the!central!challenge!

in!the! anal ysis!of!traffic!stops,!as!the!driving!population!in!a!given!area!may!lo ok!quite!different!

from!the!residential!population.!!

To!account!for!the!possibility!that!the!composition!of!daytime!drivers!may!differ!from!those!on!

the!road!at!night,!we!limited!the!analysis!to!what!is!known!as!the!‘inter-twilight!period,’!or!the!

time!period!between!the!earliest!end!of!civil!twilight!(approximately!5:09!pm!on!Nov.!27)!and!

the!latest!(approximately!8:29!pm!on!Jun.!27).!Focusing!on!this!period!allowed!us!to!capitalize!

on! a! natural! experiment! produced! by! seasonal! changes.! Because! the! sun! goes! down! much!

earlier!in!San!Diego!during!winter!months!than!it!does!in!the!summer,!people!on!the!road!at!

6:00! pm!in!January!would!experience!darkness,!but!in!July!th e!same!drive!would!occur!in!broad!

daylight.!Thus,!we!are!able!to!compare!the!l ikelih ood !that!drivers!on!the!road!during!this!3-hour!

and!20-minute!window!were!stopped!in!daylight!versus!darkness,!and!to!be!confident!that!any!

differences!found!are!due!to!race/ethnicity!rather!than!other!factors.!!

We! omitted! from! the! analysis! stops! that! occurred! as! a! result! of! a! suspect! description,! code!

enforcement!effort,!or!other!type!of!call!for!service.!By!limiting!our!samp le!to!only!those!stops!

that!involve!an!equipment!(e.g.,!a!broken!tail!light)! or!moving!violation!(e.g.,!an!illegal!left!turn),!

we! are! able!to! focus! on! discretionary! decisions,!where! an!officer’s! use! of!race/ethnicity! may!

indicate!disparate!treatment.!

Our!analysis!produced!a!series!of!mixed!results.!In!2014,!Black!drivers!were!more!likely!to!be!

stopped!during!d aylight!hours!than!after!dark,!compared!to!White!drivers.!We!found!n o !such!

disparity!in!2015!or!in!the!combined!2014/2015!dataset.!

Our!review!of! citywide! stops!involving!Hispanic!and! API! drivers!revealed! no!d isp arities! in!the!

day-night! stop! patterns! of! either! group! compared! to! White! drivers! in! 2014,! 2015,! or! the!

combined!total.!Put! another!way,!the!odds ! of!an!Hispanic!or!API! driver!being!stopped!during!

iv!

daylight!hours!are!statistically!similar!to!the!odds!o f!a!stop!involving!an!Hispanic!or!API!driver!

occurring!after!dark,!compared!to!the!day-night!stop!patterns!of!White!drivers.!!

To! complement! our! citywide! analysis,! we! also! examined! division-l evel! stop! patterns! in! 2014!

and! 2015.! Our! review! of! aggregate! data! from! th e! five! divisions! located! above! Interstate! 8!

revealed! no! statistically! significant! disparities! in! the! day-night! stop! patterns! of! either! Black,!

Hispanic,!or!API!drivers!as!compared!to!White!drivers.!Narrowing!the!focus!to!the!division!level,!

we! foun d ! evidence! of! disparities! in! the! day-night! stop! patterns! of! both! Black! and! Hispanic!

drivers!stopped!in!the!Northeastern!division,!as!compared!to!Whites.!No!such!disparities!were!

found!between!API!and!White!drivers,!or!in!any!of!the!other!four!divisions!located!above!I-8.!

Data!on!stops!conducted!below!Interstate!8!in !2014!and!2015!revealed!a!much!different!set!of!

results.!We!find!evidence!to!suggest!that!in!the!aggregate,!Black!and!Hispanic!drivers!were!less!

likely!be!stopped!during!daylight!hours!than!they!were!after!dark,!as!compared!White!drivers.!

In!other!words,!when!officers!on!patrol!below!I-8!were!able!to!see!a!driver’s!race,!they!were!

more!likely!to! stop! a!White!driver!than!either!a! Black! or!Hispanic!(but!not!API)!driver.!At! the!

division!level,!this! type! of! dispari ty! was!evident! i n! stops!occurring! in!the! Central!division! and!

exclusively!among!Hispanic!drivers!stopped!in!the!Mid-City!division.!!

Post-stop!outcomes!

The!Report!also!includes!a!detailed!analysis!of!the!extent!to!which!key!post-stop!outcomes!vary!

by!driver!race.!In!an!effort!to!eliminate!other!possible!explanations!for!racial/ethnic!disparities!

in!the!decision!to!initiate!a!search,!issue!a!citation,!conduct!a!field!interview,!or!effectuate!an !

arrest,! we! matched! API,! Black,! and! Hispanic! drivers! with! White! drivers! across! a! set! of!

demographic!and! stop-based!characteristics!using!a!statistical!technique!known!as!propensity!

score!matching.!Analysis!of!the!post-stop!outcomes!between!matched!pairs!shows!distinct!and!

sizable! differences! in! the! experiences!of! Black!and! Hispanic! drivers! and! their!matched! White!

counterparts.! No! statisti call y! significant! differences! were! evident! in! our! analysis! of! the! API-

White!pairing.!

Specifically,! the! data! show! that! SDPD! officers! were! more! likely! to! search! Black! and! Hispanic!

drivers! than! White! drivers! stopped! under! similar! circumstances.! These! results! were! largely!

consistent!across!all!search!types,!including!high!discretion!searches,!l ike!consent!searches,!and!

low! discretion! searches,! like! inventory! searches.! Across! 2014! and! 2015,! White! drivers! were!

searched!at!a!greater!rate!than!API!drivers.!!

Analysis! of! ‘hit! rates,’! or! the! percentage! of! searches! that! led! to! the! discovery! of! illegal!

contraband,! revealed! Black! and! Hispanic! drivers! were! either! less! likely! to! be! found! with!

contraband!or!found!with!contraband!at!similar!rates!than!matched!White!drivers,!depending!

on!the!nature!of!the!search.!We!found!no!meaningful!differences!in!the!hit!rates!of!matched!

API!and!White!drivers.!

We!also! used! the!propensity!score! matching! technique!to!evaluate! how! driver!race/ethnici ty!

influenced! arrest! and! field ! interview! rates,! as! well! as! the! decision! to! issue! a! citation.! Our!

analysis!showed!no!statistical!difference!in!the!arrest!rates!of!matched!Black!and!White!drivers,!

while! Hispanic! drivers! were! arrested! slightly! more! often! than! matched! Whites.! Matched! API!

drivers!were!arrested!less!frequently!than!their!matched!White!counterparts.!!

Black! drivers! were! subjected! to! field! interviews! more! than! twice! as! often! as! their! matched!

White! peers,!while!there!was!a!much!smaller!though!statistically!significant!difference!between!

both! Hispani c! and! API!drivers! as!compared! to!matched! White!drivers.! Finally,!we! found!that!

Black!drivers!received!citations!less!often!than!matched!Whites,!while!matched!Hispanic,!API,!

and!White!drivers!were!all!cited!at!nearly!identical!rates.!!

Recommendations!

Analysis! of! the! 2014! an d! 2015! traffic! stop! card! data,! as! well ! as! the! contextual! insights! we!

gained!from!several!focus!groups!with!San!Diego!community!members,!interviews!with!dozens!

of! SDPD! officers,! and! an! electronic! survey! of! SDPD! officers! suggest! three! broad,! thematic!

results.! First,! data! on! the! SDPD’s! stop! and! post-stop! enforcement! patterns! show! meaningful!

differences! in! the! treatment! of! Black! and! Hispanic! drivers,! as! compared! to! Whites.! Second,!

these! disparities,! which! match! the! perceptions! of! some! members! of! San! Diego’s! minority!

communities,! contribute! to! a! recognized! tension! between! these! communities! and! the! SDPD.!

Third,! SDPD’s! existing! system! for! collecting! and! managing! traffic! stop! data! is! fundamentally!

flawed.!!

Our!recommendations!to!the!Department!are!designed!to!address!these!broad!findings.!!

Systemic$disparities$

1. Acknowledge! the! existence! of! racial/ethnic! disparities! and! make! combatting! such!

disparities!a!priority;!

2. Continue!to!enhance!training!and!supervision!around!issues!of!racial/ethnic!disparities;!

3. Make!traffic!stop!practices!more!transparent;!and!

4. Make!traffic!stop!practices!more!systematic!and!data-driven.!!

Police-community$relations$

5. Make!community!engagement!a!core!departmental!value;!and!

vi!

6. Work!to!improve!communication!and!transparency!regarding!police!practices.!!

Data$collection$and$management$

7. Revise!the!current!data!collection!system;!

8. Coordinate!existing!data!collection!efforts;!

9. Collect!additional!data;!!

10. Strengthen!accountability!and!oversight!of!data!collection!and!management.!

We!submit!this!Report!during!a!challenging!time!for!police!departments!and!individual!officers!

across!the!country.!Public!scrutiny!of!the!role!of!police!in!our!society!and!tension!between!law!

enforcement!and!communities!of!color! has!seldom!been!more!acute!than!it!is!today.!Analysis!of!

2014!and!2015!traffic!stop!data!shows!that!perceptions!of!differential!treatment!are!supported!

by! data,! and! highlight! several! substantive! issues! th at,! in! our! view,! should! be! given! the!

Department’s!full! attenti on. !Insights!from!both!community!members!and!SDPD!officers!suggest!

that!these!are!not!insurmountable!challenges.!Rather,!the!goal!of!a!fair!and!transparent!police!

force!defined!by!a!strong!bond!with!City!residents!is!one!that!all!involved!care!deeply!about.!!!

! !

vii!

ACKNOWLEDGEMENTS!

City!Councilwoman! Marti! Emerald!commissioned!this! research,! with!funding!from! the! City!of!

San!Diego.!We!are!appreciative!of!all!the!individuals!who!made!this!research!possible.!

We!thank!Marisa!Berumen,!Deputy!Chief!of!Staff!for!Councilwoman!Emerald,!and!Ronald!Villa,!

Deputy!Chief!Operating!Officer!of!the!City!of!San!Diego,!for!facilitating!this!research.!!

Chris!Haley,!Information!Services!Program!Manager!for!the!San!Diego!Police!Department,!was!

instrumental! in! sharing! data! with! us.! Both! Ms.! Haley! and! Almis! Udrys,! Director! o f! the!

Performance! and! Analytics! Department! of! the! City! of! San! Diego,! provided! thorough! and!

thoughtful!feedback!on!early!drafts!of!the!Report.!!

We!thank!Kristina!Peralta,!Director,!Purchasing!and!Contracting!for!the!City!of!San!Diego,!and!

Sandra! Nordahl! and!Tannaz!Niknejadi! of!the!San! Diego! State!University! Research! Foundation!

for!diligently!making!sure!our!contract!to!conduct!the!research!was!executed.!!

We! than k! the! San! Diego! Police! Department! officers! who! were! generous! with! their! time! in!

participating!in!an!electronic!survey!as!well!as!follow-up!interviews!with!us.!!

We!thank!our!colleagues!at!Harder+Company!Community!Research,!particularly!Nicole!Bracy,!

Amy!Ramos,!Anna!Cruz,!Laura!Frutos,!and!Ana!Ramundo,!for!generously!lending!their!time!and!

expertise!in!planning!an d!conducting!the!community!focus!groups.!We!are!also!grateful!to!th e!

participants!of!those!groups!for!sharing!their!thoughts!and!experiences.!

We!are!grateful!to!Julie!O’Connor!of!the!School!of!Public!Affairs!at!San!Diego!State!University!

for!her!assistance!with!our!budgeting!for!this!project.!

Joyce!Gattas!and!Christianne!Penunuri,!both!of!the!College!of!Professional!Studies!and!Fine!Arts!

at! San! Diego! State,! provided! crucial! support! and! feedback! throughout! this! project.! We! also!

thank! John ! Petreikis! for! design! and! formatting! support.! Elliott! Alvarado,! Stacey! Davis,! Alexa!

Evans,!and!Anthony!Triola!provided!research!assistance!on!various!parts!of!this!research.!

Lastly,! this! Repo rt! has! been! greatly! enhanced! through! the! thoughtful! feedback! of! several!

individuals:! Hank! Fradella,! Ed! Maguire,! and! Mike! White! of! Arizona! State! University;! Preeti!

Chauhan!of!John!Jay!College! of! Criminal! Justice;!Gary!Cordner!of!Kutztown!University;!Carroll !

Seron!of!the!University! o f!Califo rnia!Irvine;!and!Rulette!Armstead,!Shawn!Flanigan,!Paul!Kaplan,!

and!Lanie!Lockwood,!all!of!San!Diego!State.! !

viii!

LIST!OF!TABLES!

2.1! Demographic!profile!of!sworn!SDPD!officers,!by!race/ethnicity,!gender,!and!year!

2.2! Racial/ethnic!composition!of!SDPD!patrol!division!residents,!ages!15!and!above!

2.3! SDPD!traffic!stop!card!data!from!2000!and!2001!

2.4! SDPD!search!rates!in!2000!and!2001,!by!driver!race/ethnicity!

2.5! Hit!rates!in!2000!and!2001,!by!driver!race/ethnicity!

3.1.!! Information!missing!from!the!2014!and!2015!datasets!

3.2.!! Incomplete!stop!cards!submitted!in!2014!and!2015,!by!police!division!!

3.3.! Incomplete!stop!cards!submitted!in!2014!and!2015,!by!driver!race/ethnicity!!

3.4.!! Comparing!judicial!citation!records!with!stop!card!citation!records!!

3.5.!! Focus!groups!and!participants!

3.6! Descriptive!statistics!for!police!officer!survey!respondents

4.1.!! Previous!research!employing!the!veil!of!darkness!analytical!approach!

4.2.!! Describing!data!generated!by!traffic!stops!conducted!by!SDPD!officers!in!2014!and!2015,!

by!stop!type!!!

4.3.!! Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!!

citywide!for!either!a!moving!violation!or!an!equipment!violation!

4.4.!! Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!!

citywide!for!either!a!moving!violation!or!an!equipment!violation!during!the!DST!period!

4.5.!! SDPD!vehicle!stops,!by!patrol!division,!2014!and!2015!combined!

4.6.!! Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!for!either!

a! moving! violation! or! an! equipment! violation! in ! 2014! and! 2015! combined,! by! stop!

location!

4.7.!! Modeling! the! effects! of! daylight! on! the! odds! that! Hispanic! drivers! will! be! stopped!

citywide!for!either!a!moving!violation!or!an!equipment!violation!

4.8.!! Modeling! the! effects! of! daylight! on! the! odds! that! Hispanic! drivers! will! be! stopped!

citywide!for!either!a!moving!violation!or!an!equipment!violation!during!the!DST!period!

4.9.!! Modeling!the! effects! of!daylight!on! the! odds!that!Hispanic! drivers! will!be!stopped! for!

either!a!moving!violation!or!an!equipment!violation!in !2014!and!2015!combined,!by!stop!

location!

4.10.! Modeling!the!effects!of!daylight!on! the!odds!that!Asian/Pacific!Islander!drivers!will!be!

stopped!citywide!for!either!a!moving!violation!or!an!equipment!violation!

4.11.! Modeling!the!effects!of!daylight!on!the!odds!that!Asian/Pacific!Islander!drivers!will!be!

stopped!citywide!for!either!a!moving!violation!or!an!equipment!violation!during!the!DST!

period!

ix!

4.12.! Modeling!the!effects!of!daylight!on!the!odds!that!Asian/Pacific!Islander!drivers!will!be!

stopped!for!either!a!moving!violation!or!an!equipment!violation!in!2014!and!2015!

combined,!by!stop!location!

4.13.!! The!demographic!profile!of!drivers!stopped!in!2014!and!2015!!

5.1.!! Traffic!stops!and!post-stop!outcomes!in!2014!and!2015,!by!SDPD!patrol!division!!

5.2.!!! Traffic!stops!and!post-stop!outcomes,!by!stop!time!!

5.3.!!! Traffic!stops!and!post-stop!outcomes,!by!driver!race/ethnicity!

5.4.!!! Comparing!search!rates!among!matched!Black!and!White!drivers!!

5.5.!!! Comparing!search!rates!among!matched!Hispanic!and!White!drivers!!

5.6! ! Comparing!search!rates!among!matched!Asian/Pacific!Islander!and!White!drivers!

5.7.!!! Raw!data!on!the!discovery!of!contraband!

5.8.!!! Comparing!hit!rates!among!matched!Black!and!White!drivers!

5.9.!! Comparing!hit!rates!among!matched!Hispanic!and!White!drivers!!

5.10! Comparing!hit!rates!among!matched!Asian/Pacific!Islander!and!White!drivers!

5.11.!! Comparing!arrest!rates!for!matched!Black!and!White!drivers!!

5.12.!! Comparing!arrest!rates!for!matched!Hispanic!and!White!drivers!!

5.13.! Comparing!arrest!rates!for!matched!Asian/Pacific!Islander!and!White!drivers!

5.14.!! Comparing!field!interview!rates!for!matched!Black!and!White!drivers!!

5.15.!! Comparing!field!interview!rates!for!matched!Hispanic!and!White!drivers!

5.16.! Comparing!field!interview!rates!for!matched!Asian/Pacific!Islander!and!White!drivers!

5.17.!! Comparing!citation!rates!for!matched!Black!and!White!drivers!!

5.18.!! Comparing!citation!rates!for!matched!Hispanic!and!White!drivers!

5.19.!! Comparing!citation!rates!for!matched!Asian/Pacific!Islander!and!White!drivers!

A1.1.!! SDPD!Patrol!Staffing,!by!division,!watch,!and!year!

A1.2.!! Crime!in!San!Diego,!CA,!by!crime!type,!location,!and!year!

A4.1.!! Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!citywide!

for!a!moving!violation!!

A4.2.!! Modeling! the! effects! of! daylight! on! the! odds! that! Hispanic! drivers! will! be! stopped!

citywide!for!a!moving!violation!!

A4.3.!! Modeling! the! effects! of! daylight! on! the! odds! that! Black! drivers! will ! be! stopped! for! a!

moving!violation,!above!and!below!Interstate!8!

A4.4.!! Modeling!th e!effects!of!daylight!on!the!odds!that!Hispanic!drivers!will !be!stopped!for!a!

moving!violation,!above!and!below!Interstate!8!

A5.1.!! Modeling! the! effects! of! daylight! on! the! odds! th at! Black! male! drivers! will! be! stopped!

citywide!for!either!a!moving!violation!or!equipment!violation!

A5.2.!! Modeling!the!effects!of!daylight!on!the!odds!that!Black!male!drivers!will!be!stopped!for!

either!a!moving!violation!or!equipment!violation,!above!and!below!Interstate!8!

A5.3.!! Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!male!drivers!will !be!stopped!

citywide!for!either!a!moving!violation!or!an!equipment!violation!

A5.4.!! Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!male!drivers!will !be!stopp ed!

for!either!a!moving!violation!or!equipment!violation,!above!and!below!Interstate!8!

A6.1.! Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!for!either!

a!moving!violation!or!an!equipment!violation!in!2014,!by!stop!location!!!

A6.2.! Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!for!either!

a!moving!violation!or!an!equipment!violation!in!2015,!by!stop!location!!!

A6.3.! Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!drivers!will!be!stopped!for!

either!a!moving!violation!or!an!equipment!violation!in!2014,!by!stop!location!!!

A6.4.! Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!drivers!will!be!stopped!for!

either!a!moving!violation!or!an!equipment!violation!in!2015,!by!stop!location!!!

A6.5.! Modeling!the!effects!of!daylight!on!the!odds!that!Asian/Pacific!Islander!drivers!will!be!

stopped!for!either!a!moving!violation!or!an!equipment!violation!in!2014,!by!stop!

location!!!

A6.6.! Modeling!the!effects!of!daylight!on!the!odds!that!Asian/Pacific!Islander!drivers!will!be!

stopped!for!either!a!moving!violation!or!an!equipment!violation!in!2015,!by!stop!

location!!!

A7.1.!! Using! l ogisti c! regression! to! model! the! likelihood! that! SDPD! officers! will! search! Black!

drivers!

A7.2.!! Using!logistic!regression!to!model!the!likelihood!that!SDPD!officers!will!search!Hi span ic!

drivers!

A7.3.!! Using!logistic!regression!to!model!post-stop!outcomes!for!Black!drivers!

A7.4.!! Using!logistic!regression!to!model!post-stop!outcomes!for!Hispanic!drivers!!

A8.1.!! Describing!matched!and!unmatched!Black!and!White!drivers!!

A8.2.!! Describing!matched!and!unmatched!Hispanic!and!White!drivers!!

A9.1.!! Comparing!hit!rates!among!matched!Black!and! White!drivers!after!dropping!missing!and!

null!cases!!!

A9.2.!! Comparing!hi t!rates!among!matched!Hispanic!and!White!drivers!after!dropping!missing!

and!null!cases!!!

A10.1.!!Comparing! citation! rates! for! matched! Black! and! White! drivers! after! dropping! missing!

and!null!cases!!!

A10.2.!!Comparing!citation!rates!for!matched!Hispanic!and!White!drivers!after!dropping!missing!

and!null!cases!!!

! !

xi!

LIST!OF!FIGURES!

2.1.!! Comparing!violent!crime!rates!across!five!major!California!cities!

2.2.!! Comparing!property!crime!rates!across!five!major!California!cities!

2.3.!! San!Diego!Police!Department!neighborhood!divisions!

2.4.!! Violent!and!property!crime!rate,!by!SDPD!neighborhood!division!

2.5.!! The!relationship!between!division!crime!rates!and!the!allocation!of!SDPD!patrol!officers!

3.1.!! Tracking!missing!data,!by!month!

3.2.!! Comparing!monthly!traffic!stop!volume,!by!year!

3.3.!! Monthly!traffic!stop!percentages,!by!driver!race/ethnicity!!

4.1.!! Comparing!driver!stop!rates!in!2014!and!2015!with!San!Diego’s!racial/ethnic!

composition!

4.2.!! Scatterplot!of!traffic!stops!included!in!the!veil!of!darkness!analysis!

4.3.!! Scatterplot!of!traffic!stops!included!in!the!Daylight!Saving!Time!veil!of!darkness!!

! ! analysis!

4.4.!! Examining!the!relationship!between!vehicle!stop!rates!and!crime,!by!SDPD!police!!

! ! division!

5.1.!! The!average!percentage!difference!between!matched!and!unmatched!Black!and!!

! ! White!drivers!across!eight!variables!used!to!complete!matching!process!

5.2.!! The!average!percentage!difference!between!matched!and!unmatched!Hispanic!and!

White!drivers!across!eight!variables!used!to!complete!matching!process!

CHAPTER!1:!INTRODUCTION!

In!February!2015!the!City!of!San!Diego!contracted!with!the!San!Diego!State!Uni versity!School!of!

Public!Affairs!to!analyze!the!San!Diego!Police!Department’s!(SDPD)!enforcement!of!local!traffic!

law.! Thi s! Report! encompasses! our! analysis! of! the! 259,569! traffic! stops! conducted! between!

January!1,!2014!and!December!31,!2015.

!Four!questions!drove!our!inquiry:!!

1. To!what!extent!is!there!a!department-level!pattern!of!racial/ethnic!disparity!in!the!

initiation!of!traffic!stops?!!

2. To!what!extent!are!racial/ethnic!disparities!in!the!in iti atio n!of!traffic!stops!evident!at!

the!patrol!division!level?!!

3. To!what!extent!is!there!a!department-level!pattern!of!racial/ethnic!disparity!in!the!

outcome!of!traffic!stops?!!

4. How!does!the!SDPD’s!traffic!enforcement!regime!affect!police-community!relations!

in!San!Diego?!!

The! Report! is! organized! as! follows.! In! Chapter! 2! we! contextualize! our! analysis! by! discussing!

policing! in! San! Diego.! We! begin! by! describing! the! organization! and! operation! of! the!

Department!and!summarizing!citywide!crime!trends.!We!then!review!the!Department’s!recent!

history,!which!has!included!efforts! to!address!allegations!of!officer!misconduct! and !tension!with!

communities!of!color.

!Finally,!we!discuss!in!some!detail!findings!f rom!a!previous!independent!

analysis!of!SDPD!traffic!stop!data!conducted!in!2000!and!2001.

In!Chapter!3!we!describe!the!data!u sed!to!complete!our!analysis.!We!review!the!mechanism! f or!

recording! information! about! traffic! stops,! the! ‘vehicle! stop! card,’! and! discuss! observable!

patterns!in!the!volume! and! quality! of!the! dataset.!We! also! describe!the! process! of!gathering!

contextual! information! about! traffic! stops! through! conducting! focus! groups! with! San! Diego!

community!members!and!surveying!and!interviewing!SDPD!officers.!!!

In!Chapter!4!we!examine!traffic!stop!patterns!at!the!Department!level,!at!the!individual!patrol!

division! level,! and! compare! stop! patterns! above! Interstate! 8! with! those! occurring! below! I-8.!

After! discussing! the! analytical! challenges! presented! by! this! issue,! we! describe! in! detail! the!

statistical!method!used!to!address!the!extent!to!which !racial/ethnic!disparities!exist.!The!‘veil!of!

!The!raw!data!files!w e!received!from!the!SDPD!contained!a!total!of!259,586!records.!17!records!were!corrupted!

and!thus!dropped!from!the!analysis.!

!Police!Executive!Research!Forum!(PERF).!(2015).!Critical$response$technical$assessment$review:$Police$

accountability$-$findings$and$na tion al$im plica tion s$of$a n$as sessment$of$the$San$Dieg o$P olice $dep artm e nt.!

Washington,!DC:!Office!of!Community!Oriented!Policing!Services,!U.S.!Department!of!Justice.!!

!Cordner,!G.,!William s,!B.,!&!Zu niga ,!M .!(200 1).!San$Diego$Police$Department$vehicle$stop$study:$Year-end$report.!

San!Diego,!CA.!!!

darkness’! technique,! our! chosen! approach,! allows! the! researcher! to! isolate! the! effect! of!

race/ethnicity!from!other!factors!by!comparing!the!distribution!of!stops!made!during!daylight!

hours,!when!th e!race/ethnicity!of!the!driver!is!more!apparent,!to!those!made!after!sundown,!

when! driver! race/ethnicity! is! obscured! by! darkness.! We! complete! the! Chapter! by! comparing!

day-night!stop!patterns!experienced!by!Asian/Pacific!Islander!(API),!Black,!Hispanic,!and!White!

drivers.!

In! Chapter! 5!we!present! our!analysis!of! post-stop!outcomes,!with! a!focus! on! examining!how!

race/ethnicity!affects!the!likelihood!that!a!driver!will!have!their!person!or!vehicle!searched!and!

whether! that! search! will! lead! to! the! d iscovery! of! contraband.! We! also! examine! how! driver!

race/ethnicity! influences! the! odds ! that! a! stopped! driver! receives! a! citation! or! is! given! a!

warning,! is! subject! to! a! field! interview,! and! whether! the! driver! is! ultimately! arrested.! The!

Chapter! begins! with! a! detailed! discussion! of! the! analytical! approach! driving! our! analysis.!

Propensity!score!matching!is!a!technique!that!allows!the!researcher!to!match!drivers!based!on!

a!set!of!demographic!and!stop-related!characteristics!so!as!to!isolate!the!effect!of!race.!From!

there!we!present!a!detailed!analysis!of!data!on!several!post-stop!outcomes,!including!searches,!

‘hit! rates,’! or! the! percentage! of! searches! that! lead! to! the! discovery! of! illegal! contraband,!

arrests,!field!interviews,!and!the!issuance!of!citations!and!warnings.!!

We! conclude! the! Report! in! Chapter! 6! with! a! brief! summary! of! our! findings! and! a! series! of!

recommendations.!

! !

CHAPTER!2:!POLICING!IN!SAN!DIEGO!

Introduction!

San! Diego,! California! is! the! eighth! l argest! city! in! the! United! States! and! one! of! the! country’s!

most!diverse!places! to! live.

!It!is! also! one! of! the!safest.! As!Figures! 2.1!and! 2.2!indicate,! both!

violent! and! property! crime! in! San! Diego! are! relatively! rare! occurrences,! compared! to!

California’s! other! major!cities.! Further,! in!2014,! the!City! of! San!Diego! had! the!second! lowest!

violent! crime! rate! (3.81! per! 1,000! residents)! and! property! crime! rate! (19.59! per! 1,000!

residents)! among! the! country’s! 32! cities! with! populations! greater! than! 500,000.

! Even! with!

slight! increases! in! 2015,! the! rates! of! both! violent! crime! (up! 5.3! percent! from! 2014)! and!

property!crime!(up!7.0!percent)!in!San!Diego!remain!at!historically!low!levels.

Despite! these! optimal! circumstances,! the! recent! history! of! the! San! Diego! Police! Department!

(SDPD)!has!been!challenged!by!hiring!and!retention!difficulties,!all egations!of!misconduct,!and!

public! criticism.

! In! this! Chapter,! we! discuss! the! context! of! policing! in! San! Diego! and! briefly!

review!the!issues!that!precipitated!this!Report.!!!!

! !

!United!States!Census!Bureau.!(2015,!May).!Annual!estimates!of!the!resident!population!for!incorporated!places!of!

50,000!or!more,!ranked!by!July!1,!2014!population:!April!1,!2010!to!July!1,!2014.!Retrieved!Aug.!24,!2016,!from!

http://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk;!Cima,!R.!(2015,!A u g u s t!

11).!The!most!and!least!diverse!cities!in!America.!Retrieved!Aug.!24,!2016,!from!http://priceonomics.com/the-

most-and-least-diverse-cities-in-america/.!!

!Burke,!C.!(2016,!Apr.).!Thirty-six!years!of!crime!in!the!San!Diego!region:!1980-2015.!SANDAG,!Criminal!Justice!

Research!Division.!Retrieved!Jul.!19,!2016,!from!

http://www.sandag.org/uploads/publicationid/publicationid_2020_20533.pdf.!

!Burke,!C.!(2016,!Apr.).!Thirty-six!years!of!crime!in!the!San!Diego!region:!1980-2015.!SANDAG,!Criminal!Justice!

Research!Division.!Retrieved!Jul.!19,!2016,!from!

http://www.sandag.org/uploads/publicationid/publicationid_2020_20533.pdf.!

!e.g.,!Dillon,!L.!(2014 ,!Dec.!23).!Misconduct!issues!will!fo llo w !S D P D !in to !2 01 5 .!Voice$of$San$Diego.!Retrieved!Au g.!

22,!2016,!from!http://www.voiceofsandiego.org/topics/government/misconduct-issues-will-follow-sdpd -in t o -

2015/;!Garske,!M.,!&!Stickney,!R.!(2014,!Sept.!24).!$5.9M!paid!to!settle!ex-cop!Anthony!Arevalos!civil!lawsuit.!NBC$

&$San$Diego.!Retriev ed !N ov .!8 ,!20 16 ,!fro m !http://www.nbcsandiego.com/news/local/Anthony-Arevalos-Jane-Doe-

Settlement-Details-SDPD-Sex-Crimes-277069491.html!;!$Kucher,!K.,!Davis,!K.,!& !Repard,!P.!(2015,!Mar.!17).!Audit:!

SDPD!flaws!led!to!misconduct.!The$San$Diego$Union$Tribune.!Retrieved,!Nov .!8,!2016,!from!

http://www.sandiegouniontribune.com/sdut-police-misconduct-review-ju stic e - 2015m ar17-htmlstory.html.!

Figure!2.1.!

Comparing!violent!crime!rates!across!five!major!California!cities!

Source:!Federal!Bureau!of!Investigation!(2012)!

Figure!2.2.!

Comparing!property!crime!rates!across!five!major!California!cities!

Source:!Federal!Bureau!of!Investigation!(2012)!

500

1,000

1,500

2,000

2,500

3,000

1985 1988 1991 1994 1997 2000 2003 2006 2009 2012

Fresno Los'Angeles San'Diego San'Francisco San'Jose

2,000

4,000

6,000

8,000

10,000

12,000

1985 1988 1991 1994 1997 2000 2003 2006 2009 2012

Fresno Los'Angeles San'Diego San'Francisco San'Jose

The!San!Diego!Police!Department!

As!of!October!3,!2016,!the!San!Diego!Police!Department!(SDPD)!employs!1,869!sworn!officers,!

or!about!1.4!sworn!officers!per!1,000!residents.

!This!ratio!is!notably!lower!than!the!average!

rate!of!police!departments!in!other!similarly!sized!American!cities.

!The!department’s!ongoing!

struggle! to! hire! and! retain! qualified! officers! has! been! well-publicized,

! as! have! been! the!

corresponding!public!safety!and!departmental!morale!concerns.

Table!2.1.!!

Demographic!profile!of!sworn!SDPD!officers,!by!race/ethnicity,!gender,!and!year!

!Officer!Race!

Male!

Female!

Total!

Citywide!

demographic!

profile!

2014!

!!!!!Asian/Pacific!Islander!

145!(7.7%)!

23!(1.2%)!

168!(9.0%)!

20.2%!

!!!!!Black!!

108!(5.8)!

10!(0.5)!

118!(6.3)!

5.5!

!!!!!Hispanic!

319!(17.0)!

65!(3.5)!

384!(20.5)!

27.0!

!!!!!White!

1,011!(54.0)!

193!(10.3)!

1,204!(64.2)!

47.2!

$$$$$2014$Total$

1,583$(84.5)$

291$(15.5)$

1,874$(100.0)$

100.0$

2015!

!!!!!Asian/Pacific!Islander!

142!(7.6%)!

28!(1.5%)!

170!(9.1%)!

20.2%!

!!!!!Black!

105!(5.6)!

12!(0.6)!

117!(6.3)!

5.5!

!!!!!Hispanic!

325!(17.4)!

70!(3.7)!

395!(21.2)!

27.0!

!!!!!White!

997!(53.4)!

188!(10.1)!

1,185!(63.5)!

47.2!

$$$$$2015$Total$

1,569$(84.0)$

298$(16.0)$

1,867$(100.0)$

100.0$

Note:!Native!American!and!‘Other’!dr iv e rs !in c lu d ed !in !t h e!A sia n /Pacific!Islander!category.!Discrepancies!in!the!

percentage!totals!are!owed!to!rounding!error.!

!City!of!San!D iego,!Report!to!the!City!Council,!Public!Safety!&!Livable!Neighborhoods!Committee.!(2016,!O ctober!

26).!San!Diego!Police!Department!Swo rn,!Civilian!and!Communication!Staffing!Update.!Retrieved!Oct.!30,!2016,!

from!http://docs.sandiego.gov/councilcomm_agendas_attach/2016/psln_161026_2.pdf.!!

!Reaves,!B.!(2015,!May).!Local!police!departments,!2013:!Personnel,!policies,!and!practices.!U.S.$Department$of$

Justice,$Office$of$Justice $Pro gra m s ,$Bu rea u$o f$Jus tice$S tatistic s.!Retrie ve d!A u g.!2 4 ,!20 1 6,!fro m !

http://www.bjs.gov/content/pub/pdf/lpd13ppp.pdf.!

!e.g.,!Keats,!A.!(2016,!Apr.!4).!SD!police!hoping!to!rehire!retirees!—!and!it!could!save!the!chief’s!job!too.!Voice$of$

San$Diego.!Retrieved!Jul.!1 9 ,!20 16 ,!fro m !http://www.voiceofsandiego.org/topics/government/sd-police-hoping-to-

rehire-retirees-sa ve-th e-chiefs-job-too /;!Re p a rd ,!P .!(2 0 1 6 ,!Mar.!11).!M o re !S D P D !o ffic e rs !leaving!despite!better!pay.!

The$San$Diego$Union-Tribune.!Retrieved!Jul.!19,!2016 ,!fro m !

http://www.sandiegouniontribune.com/news/2016/mar/11/sdpd-police-retention-hiring/!

!e.g.,!Monroy,!M.!(2014,!Sept.!20).!SDPD’s!staffing!problems!are!‘hazardous!to!your!health.’!Voice$of$San$Diego.!

Retrieved!Jul.!19,!2016,!from!http://www.voiceofsandiego.org/2014/09/20/sdpds-staffing-problems-are-

hazardous-to-your-health/.!

Per! Table! 2.1,! despite! efforts! to! diversify! the! force,

! the! demographic! profile! of! the! SDPD’s!

sworn! officers! is! disproportionately! male! and! l ess! racially! and! ethnically! diverse! than! the!

citywide!population.

!The!SDPD!is!not!unique!in!its!relative!homogeneity.!In!fact,!according!to!a!

recent!New$York$Times!analysis!of!2007!FBI!data,!the!“race/ethnicity!gap”!between!the!police!

and!residents!in!other!major!ci ties,!including!Los!Angeles,!San!Francisco,!and!many!others,!is!far!

greater!than!in!San!Diego.

!We!also!note!that!as!of!this!writing!SDPD’s!force!is!comprised!of!16!

percent! female! officers,! slightly! below! the! 17! percent! average! among! departments! serving!

cities!with!populations!of!250,000!or!more.

Figure!2.3.!!

San!Diego!Police!Department!neighborhood!divisions!

!Tragaser,!C.!(2015,!Aug.!21).!San!Diego!Police!Department!academy!class!sees!increased!diversity.!KPBS.org.!

Retrieved!July!28,!2016,!from!http://www.kpbs.org/news/2015/aug/21/san-diego-police-department-academy-

class-sees-inc/.!

!United!States!Census!Bureau.!(2015,!August!12).!State!&!County!QuickFacts,!San!Diego!(city),!California.!

Retrieved!Aug.!24,!2016,!from!http://quickfacts.census.gov/qfd/states/06/0666000.html.!

!Ashkenas,!J.,!&!Park,!H.!(2015,!April!8).!The!race!gap!in!America’s!police!d ep artm e nts .!The$New$York$Times.!

Retrieved!from!Aug.!11,!2016,!from!http://www.nytimes.com/interactive/2014/09/03/us/the-race-gap-in-

americas-police-departments.html?_r=0.!

!Reaves,!B.!(2015,!May).!Local!police!departments,!2013:!Personnel,!policies,!and!practices.!U.S.!Department!of!

Justice,!Office!of!Jus tice!P rog ram s ,!Bu rea u!o f!Jus tice!S tatistic s.!Re trieve d !Aug.,!24,!2016,!from!

http://www.bjs.gov/content/pub/pdf/lpd13ppp.pdf.!

JOLLA

OTAY MESA

Pacific Ocean

SOUTHERN

CENTRAL

SOUTHEASTERN

EASTERN

NORTHEASTERN

RAMONA

ESCONDIDO

LAKESIDE

POWAY

CARLSBAD

SAN DIEGUITO

SWEETWATER

MIRAMAR

EL CAJON

SANTEE

CHULA VISTA

SPRING VALLEY

LA MESA

NATIONAL

CITY

NORTHERN

NORTHWESTERN

WESTERN

MID-CITY

San Diego Police Department District Boundaries

Legend

District Boundaries

Beats

0 52.5 Miles

Datum: D_North_American_1983

Projection: Lambert_Conformal_Conic

Source: San Diego Police Department, U.S. Census Bureau (2010)

KEARNY

MESA

RAMONA

Scripps Ranch

The!Department!divides!patrol!activities!across!nine!geographic!divisions,!visible!in!Figure!2.3.!

These! divisions! vary! greatly! across! several! relevant! categories,! including! residents’! racial! and!

ethnic! composition,! their! socio-economic! status,! as! well! as! the! presence! of! both! crime! and!

police.!!

Table!2.2.!!

Racial/ethnic!composition!of!SDPD!patrol!division!residents,!ages!15!and!above!

Asian/PI!

Black!

Hispanic!

White!

Total!

Above!Interstate!8!

Northern!

37,473!(19.0%)!

3,440!(1.7%)!

25,673!(13.0%)!

130,299!(66.2%)!

196,885!(100.0%)!

Northeastern!

63,499!(35.6)!

5,184!(2.9)!

18,239!(10.2)!

91,654!(51.3)!

178,576!(100.0)!

Eastern!

17,685!(14.9)!

6,162!(5.2)!

18,201!(15.3)!

76,539!(64.5)!

118,587!(100.0)!

Western!

13,232!(11.5)!

4,136!(3.6)!

20,014!(17.4)!

77,629!(67.5)!

115,011!(100.0)!

Northwestern!

15,380!(27.1)!

510!(0.9)!

3,908!(6.9)!

36,889!(65.1)!

56,687!(100.0)!

Sub-total$

147,269$(22.1)$

19,432$(2.9)$

86,035$(12.9)$

413,010$(62.0)$

665,746$(100.0)$

Below!Interstate!8!

Central!

6,605!(8.2%)!

6,213!(7.7%)!

32,844!(40.9%)!

34,728!(43.2%)!

80,390!(100.0%)!

Southeastern!

32,904!(25.8)!

22,024!(17.3)!

59,397!(46.5)!

13,344!(10.5)!

127,669!(100.0)!

Southern!

10,524!(13.0)!

2,999!(3.7)!

58,859!(72.6)!

8,701!(10.7)!

81,083!(100.0)!

Mid-City!

20,364!(15.5)!

12,751!(9.7)!

51,516!(39.2)!

46,800!(35.6)!

131,431!(100.0)!

Sub-total$

70,397$(16.7)$

43,987$(10.5)$

202,616$(48.2)$

103,573$(24.6)$

420,573$(100.0)$

Citywide!total!

217,666!(20.0)!

63,419!(5.8)!

288,651!(26.6)!

516,583!(47.6)!

1,086,319!(100.0)!

Source:!The!City!of!San!Diego.

!Note:!Percentage!d iscre pa n c ies !reflect !ro un d ing!erro r.!

Table!2.2!displays! the! racial! and! ethnic!breakdown! of! the!Department’s! nine! police!division s.!

The! highest! concentrations! of! Black! residents! are! found! in! the! S outh eastern! and! Mid-City!

divisions,! where! White! and! Asian/PI! populations! are! among! their! l owest.! Similarly,! Hispanic!

residents!tend!to!reside!in!the!Southern,!Southeastern,!and!Mid-City!divisions.!Poverty!is!also!

concentrated!in!these!neighborhoods.!In!fact,!census!tracts!in! th ese!divisions!are!home!to!many!

of!the!San!Diego’s!poorest!residents.

!Conversely,!neighborhoods!located!above!Interstate!8,

!The!City!of!San!Diego,!Public!Safety!&!Livable!Neighborhoods!Committee!(2015,!Feb.!13).!Report!to!the!City!

Council!(Report!No.15-016).!Vehicle!Stop!Data!Cards:!January!through!December!2014.!Retrieved!Aug.!27,!2016,!

from!http://docs.sandiego.gov/councilcomm_agendas_attach/2015/psln_150225_3.pdf.!

!Kyle,!K.!(2012,!A ugust!6).!Where!San!Diego’s!poorest!live:!Map.!The$Voice$of$San$Diego.!Retrieved!Aug.!24,!201 6 ,!

from!http://www.voiceofsandiego.org/community/where-san-diegos-poorest-live-map/.!

including!those!in!the!Northern,!Northeastern,!Northwestern,!Eastern,!and!Western!divisions,!

where!income!levels!tend!to!be!higher,!are!also!home!to!greater!percentages!of!White!and!API!

residents.!!

Figure!2.4.!!

Violent!and!property!crime!rate,!by!SDPD!neighborhood!division!

Source:!The!City!of!San!Diego.

Note:!Crime!rates!are!calculated!per!1,000!patrol!division!reside nts !and !reflect!data!from!2014!and!2015.!!

Figure!2.4!highlights!the!relationship!between!property!crime!an d!violent!crime!across!the!nine!

divisions.

!In!2014!and!2015,!the!highest!rate!of!violent!crime!occurred!in!the!Central!division!

(11.0! incidents! per! 1,000! residents),

! followed! by! the! Mid-City! (6.0)! and! Western! (5.6)!

!We!use!Interstate!8!here!and!throughout!the!remainder!of!the!Report!as!a!rough!point!of!demarcation!for!

divisions!and!neighborhoods!in!the!northern!portion!of!the!City!and!those!in!the!southern!p o r tio n !o f!th e !C ity.!T h e !

distinction!between!locations!‘Above!Interstate!8’!and!‘Below!Interstate!8’!is!not!exact,!as!two!patrol!divisions!that!

we!consider!‘Above!I-8’!include!small!parcels!of!land!located!below!I-8.!!!!

!See!The!City!of!San!Diego,!Actual!Crimes!by!Neighborhood,!2014!and!2015,!Crime$Statistics$and$Maps:$

Automated$Regional$Justice$Information$System$(ARJIS).!Retrieved!Oct.!14 ,!20 1 6 ,!from!

https://www.sandiego.gov/police/ se rvic es /sta tis tics.!

!See!Appendix!1!for!a!detailed!description!of!property!and!violent!crime!across!the!SDPD’s!nine!patrol!divisions!in!

2014!and!2015.!

!According!to!the!he!SDPD,!“Crime!rates!per!1,000!population!are!commonly!used!to!com pare!crime!in!different!

areas,!and!work!well!fo r!areas!that!have!a!significant!residential!population.!!Caution!is!advised!when!comparing!

Property!crime!rate

Violent!crime!rate

Violent!crime!rate Property!crime!rate

divisions.!The!highest!rate!of!property!crime!occurred!in!the!Western!(33.7!per!1,000!residents),!

Central!(33.2),!and!Eastern!divisions!(24.4).

!On!average,!in!2014!and!2015,!violent!crime!was!

more!likely!to!occu r!below!Interstate!8!(6.2!incidents!per!1,000!people)!than!in!divisions!to !the!

north!of!the!highway!(2.6),!while!the!property!crime!rates!were!similar!in!each!location!(21.6!

below!Interstate!8!compared!to!20.6!above!Interstate!8).!!

Figure! 2.5! shows! the! relationship! between! a! division’s! crime! rate! and! the! allocation! of! non-

traffic!patrol!officers.

!!!!

Figure!2.5!

The!relationship!between!division!crime!rates!and!the!allocation!of!SDPD!patrol!officers!

Source:!San!Diego!Police!Department,!City!of!San!Diego.!

Note:!Crime!data!reflect! averages! from! 2014!and! 2015!per! 1,000! residents.!Officer!rates,!which!also!reflect!the!

average!between!2014!and!2015,!are!listed!per!100,000!residents.!!

crime!rates!in!areas!with!few!residents,!especially!areas!with!significa nt!daytime!popu lation!due!to!large!

recreational!and/o r!comm ercial!areas,!since!crime!rates!use!residential!population!figures.!Higher!crime!rates!can!

be!expected!in!areas!such!as!downtown,!where!the!large!daytime!working!population!and!nighttime!

entertainment!district!crowds!are!not!included!in!the!area’s!residen tial!po p ula tion .”!

!The!correlation!coefficient!(Pearson’s!r)!between!violent!and!property!crime!is!0.719,!indicating!a!m oderately!

positive!relationship!between!violent!and!property!crime.!

!The!two!variables!are!strongly!correlated!(Pearson’s!r!=!0.8725 ),!w hic h!means!that!hig h !crim e !ra tes !are !

associated!with!high!patrol!officer!presence.!

100

120

Patrol !officer!rate

Crime!rate

Crime!rate Patrol!officer!rate

10!

The! highest! concentration! of! non-traffic! patrol! officers! occurs! in! those! divisi ons! with! the!

highest!crime!rates,!including!the!Central!(99.5!officers!per!100,000!residents),!Western!(69.8),!

and!Mid-City!(63.3)!divisions.!(A!full!documentation!of!officer!allocation!by!division!is!found!in!

Appendix! 1.)! The! SDPD! did! not! provide! us! with! data! on! the! geographic! allocation! of! traffic-

specific!officers,!who!are!not!assigned!to!a!particular!division!and!thus!may!patrol!anywhere!in!

the!City’s!jurisdiction.!

To! summarize,! Black! and! Hispanic! San! Diego! city! residents! tend! to! live! in! different!

neighborhoods!than!their!White!and!Asian /PI!counterparts.!Neighborhoods!south!of!Interstate!

8,! including! those! in! the! Central,! Mid-Ci ty,! Southern,! and! Southeastern! Divisions,! are! more!

racially!and!ethnically!diverse!than!those!located!north!of!Interstate!8,!and!some!–!but!not!all!–!

of! th ese!divisions!tend!to!face!higher!th an!average!crime!rates.!Police!presence!is!also!higher!in!

those!predominantly!non-White!Divisions.!

Police-Community!Relations!

In! this! section,! we! review! the! recent! history! of! the! Department! with! the! hope! of! providing!

context!for!our!analysis!of!the!2014!and!2015!traffic!stop!data.!!

In!early!2014,!following!several!high!profile!incidents!of!officer!misconduct,!former!SDPD!Chief!

William! Lansdowne! sought! assistan ce! from! the! U.S.! Department! of! Justice’s! (DOJ)! Office! of!

Community! Oriented! Policing! Services! (COPS! Office)! in! reviewing! the! Department’s!

management! of! offi cer! misconduct! cases,! their! approach! to! recruitment! and! background!

screening,!and!the!operation!of!the!SDPD!internal!affairs!unit.!The!COPS!Office!hired!the!Police!

Executive!Research!Forum!(PERF)!to!conduct!the!assessment.!!

The!2015!PERF!Report,

!which!detailed!the!findings!of!the!yearlong!audit,!identified!a!series!of !

organizational,! policy,! and! personnel! weaknesses! that! contributed! to! the! Department’s!

misconduct!problems.!The!report!set!a!comprehensive!reform!agenda!designed!to!strengthen!

the! SDPD’s! ability! to! prevent! misconduct! and! respond! effectively! to! incidents! that! do! occur.!

PERF! also! made! clear! that! the! miscond uct! scandals! had! undermined! the! Department! in! the!

eyes! of! San! Diego! City! residents,! particularly! among! communities! of! colo r.! The! authors!

repeatedly!underscored! the!importance!of!Department!attention!to!issues!of!racial/ethnic!bias,!

at!one!point!noting!that,!

the! most! common! suggestions! heard! from! community! members! regarding! how! to!

!Police!Executive!Research!Forum!(PERF).!(2015).!Critical$response$technical$assessment$review:$Police$

accountability$-$findings$and$na tion al$im plica tion s$of$a n$assessment$of$the$San$Diego$Police$department.!

Washington,!DC:!Office!of!Community!Oriented!Policing!Services,!U.S.!Department!of!Justice.!

11!

improve!policing!in!San!Diego!were!to!increase!police-community!engagement!through!

proactive! and! positive! interactions! and! to! add ress! issues! of! perceived! bias,! especially!

racial!bias.

This!was!not!the!first!time!the!Department!had ! been! accused! of! racial/ethnic! bias.! In! fact,! in!

2000,!a!very!similar!set!of!issues!motivated!SDPD!leadership!to!request!an!independent!review!

of!traffic!stop!data!nearly!identical!to!the!one!we!have!undertaken!here.!!

Revisiting!the!2000!and!2001!data!

In! January! 2000,! in! response! to! “concern…! expressed! by! some! community! members! about!

whether! they! [were]! being! treated! fairly! in! contacts! with! law! enforcement,”

! SDPD! officers!

began!capturing!information!about!every!traffic!stop!conducted!in!San!Diego.! Dr.!Gary!Cordner,!

a!criminologist!at!Eastern!Kentucky!University!at!the!time,!analyzed!these!data!in!an!effort!to!

address! the! extent! to! which! officer! stop! and! post-stop! decision-making! reflected! race-based!

disparities.!

Table!2.3.!!

SDPD!traffic!stop!card!data!from!2000!and!2001!

2000!

2001!

Vehicle!Stops!

168,901!

121,013!

Citation!rate!(%)!

66.1!

68.8!

Search!rate!(%)!

6.4!

7.1!

Hit!rate!(%)!

8.9!

8.4!

Arrest!rate!(%)!

1.9!

High-level! descriptive! data! from! traffic! stop! cards! gathered! in! 2000! and! 2001! are! shown! in!

Table!2.3.!Officers!co mpleted!significantly!fewer!stop!cards!in!2001!than!in!2000,!yet!remained!

fairly! consistent! from! year! to! year! in! terms! of! post-stop! activity,! including! the! rate! at! which!

stopped!drivers!were!given!citations,!searched,!and!arrested.!!

!Police!Executive!Research!Forum!(PERF).!(2015).!Critical$response$technical$assessment$review:$Police$

accountability$-$findings$and$na tion al$im plica tion s$of$a n$as sessment$of$the$San$Dieg o$P olice $dep artm e nt.!

Washington,!DC:!Office!of!Community!Oriented!Policing!Services,!U.S.!Department!of!Justice,!p.!22!

!Cordner,!G.,!W illiams,!B.,!&!Zuniga,!M.!(2001).!San$Diego$Police$Department$vehicle$stop$study:$Year$end$report.!

San!Diego,!CA,!p.!ii.!

12!

The!28.4!percent!decline!from!2000!to!2001!led!Cordner!and!his!colleagues!to!openly!question!

the!accuracy!of!the! 2001! data.! The! authors!argued! that!the! “ very! substantial!decrease! raises!

serious!questions!about!the!validity!of!the!vehicle!s top!data.!One!question!is!whether!officers!

always!filled!out!the!vehicle!stop!forms!–!the!answer!to!this!is!clearly!no.”

!They!went!on!to!

assert!th at!the!officers’!non-compliance!i n!completing!traffic!stop!cards!“was!a!bigger!problem!

in! more! ethn ical ly-di verse! and! less-affluent! divisions,! possibly! skewing! the! data.”

! The!

researchers!were!unabl e!to!interpret!how!the!missi ng!data!may!have!affected!the!rate!of!post-

stop! activity,! or! draw! conclusions! about! whether! unrecorded! post-stop! activity! may! have!

disproportionately! affected! certain! racial/ ethni c! groups.! As! such,! they! urge! caution! in! the!

interpretation!of!data!gathered!in!2001.!!

Table!2.4.!!

SDPD!search!rates!in!2000!and!2001,!by!driver!race/ethnicity!

2000!

2001!

Asian/Pacific!Islander!!!

3.2%!

3.3%!

Black!!

10.1!

11.1!

Hispanic!!$

11.4!

12.7!

White!!!

3.2!

4.1!

Source:!Cordner!et!al.!(2001;!2002)!

Note:!These!data!reflect!what!Cordner!et!al.!term!“chances!of!being!searched”!and!are!based!on!a!raw!comparison!

of!search!rates!across!all!stop!and!search!types.!!

As!will!be!discussed!in!detail!in!Chapter!4,!isolating!the!influence!of!driver!race/ethnicity!on!an!

officer’s! decision! to! stop! a! driver! is! a! complicated! task.! The! central! chal lenge,! noted! by! the!

Cordner-led!team!and!many!others,

!is!identifying!the!appropriate!benchmark!against!which!to!

compare!race-based!stop!patterns.!After!acknowledging!the!absence!of!a!“reliable!method!of!

determining!the!actual!ethnic!composition!of!the!driving!population,”!th e!Cordner!et!al.!study!

proceeded! to! compare! the! racial/ethnic! composition! of! drivers! stopped! to! the! City’s!

demographic!profile!according!to!the!U.S.!Census.!In!2000,!“Hispanics!represent!20.2%!of!the!

city’s!driving-age! po pul atio n! but!29.0%! of! vehicle!stops;! the! comparable!numbers!for! African!

!Cordner,!G.,!W illiams,!B.,!&!Velasco,!A.!(2002).!San$Diego$Police$Department$vehicle$stops$in$San$Diego:$2001.!

San!Diego,!CA,!p.!1.!

!Cordner,!G.,!W illiams,!B.,!&!Velasco,!A.!(2002).!San$Diego$Police$Department$vehicle$stops$in$San$Diego:$2001.!

San!Diego,!CA,!p.!2!

!Engel,!R.S.,!&!Calnon,!J.M.!(2004).!Comparing!benchmark!methodologies!for!police-citizen!contacts:!T raffic!stop!

data!collection!for!the!Pennsylvania!State!Police.!Police$Quarterly,!7(1),!97-125;!Fridell,!L.A.!(2004).!By$the$numbers:$

A$guide$for$analyzing$race$data$from$Vehicle$Stops.!Washington,!D .C .:!Po lice !Ex ec ut ive !Re se arc h!F o rum;!Ridgeway ,!

G.!&!MacDonald,!J.!(2010).!Methods!for!assessing!racially!biased!policing.!In!S.K.!Rice!&!M.D.!White!(Eds.)!Race,$

ethnicity,$and$policing:$New$and$essential$readings!(pp.!180-204).!New!York:!New!York!University!Press.!

13!

Americans! are! 8.0%! and! 11.7%,! respectively.”

! The! 2001! data! showed! similar! disparities! for!

both!Black!and!Hispanic!drivers.

Cordner! and! colleagues! also! examined! the! influence! of! driver! race/ethnicity! on! officers’!

decision!to!cond uct!a!search!of!the!driver,!passenger,!or!vehicle.!Unlike!with!traffic!stop!data,!

researchers!are!not!reliant!upon!benchmarks!to!assess!the!influence!of!race/ethnicity!on!post-

stop!outcomes,!like!citation!and!search!rates.!As!Table!2.4!shows,!in!2000!and!2001,!Black!and!

Hispanic!drivers!were!searched!at!higher!rates!than!either!White!or!Asian/PI!drivers.!!

Table!2.5.!!

Hit!rates!in!2000!and!2001,!by!driver!race/ethnicity!

2000!

2001!

Asian/Pacific!Islander!!

9.2%!

10.1%!

Black!!

13.9!

12.4!

Hispanic!!

5.1!

5.0!

White!!

13.1!

11.7!

Note:!These!data!reflect!a!raw!comparison!of!hit!rates!across!all!stop!and!search!types.!!

Table! 2.5! shows! the! ‘hit! rate,’! or! the! percentage! of! searches! that! led! to! the! discovery! of!

contraband,!achieved!by!SDPD! offi cers!in!2000!and!2001.!Hit!rates!varied!considerably! by!driver!

race/ethnicity!while!remaining!fairly!consistent!from!year!to! year.!Black!drivers!were!most!l ikely!

to!be!found!with!contraband,!followed!closely!by!Whi tes.!Hispanic!drivers!were!more!likely!to!

be! searched!than! any! other! racial/ethnic! group,!yet! searches! involving! Hispanic! drivers! were!

substantially!less!likely!to!uncover!possession!of!contraband.!

For! several! reasons,! most! saliently! the! low! quality! of! the! 2001! data,! we! agree! with! Dr.!

Cordner’s! recommended! cautious! interpretation! of! these! results.! With! that! said,! Cordner’s!

analysis! of! data! from! stop! cards! completed! in! 2000! and! 2001! appear! to! show! race-based!

disparities! in! SDPD! officers’! decision! to! initiate! a! traffic! stop! and! various! post-stop! actions,!

including!the!decision!to!search.!However,!without!evidence!to!show!that!post-stop!outcomes!

were! the! result! of! race-based! decisions,! we! cannot! assume! this! causal! link.! As! we! discuss! in!

Chapter!4,!this! is!why!the!veil!of!darkness! technique!is!so!important!as!it!controls! for!factors!

other!than!race/ethnicity!in!the!decision!to!make!a!stop.! !

!Cordner,!G.,!W illiams,!B.,!&!Zuniga,!M.!(2001).!San$Diego$Police$Department$vehicle$stop$study:$Yea r$end$report.!

San!Diego,!CA,!p.!v ii.!

!Cordner,!G.,!W illiams,!B.,!&!Velasco,!A.!(2002).!San$Diego$Police$Department$vehicle$stops$in$San$Diego:$2001.!

San!Diego,!CA.!

14!

CHAPTER!3:!DESCRIPTION!OF!THE!DATA!

In! Chapter! 3,! we! describe! the! data! used! for! this! Report,! beginning! with! the! administrative!

records!generated!by!the!SDPD!following!traffic!stops!conducted!between!January!1,!2014!and!

December!31,!2015.!From!there!we!go!on!to!detail!the!process!used!to!gather!the!perspectives!

of!SDPD!staff!and!members!of!the!community.!

Traffic!Stop!Data!

When!an!SDPD!officer!completes!a!traffic!stop,!they!are!required!under!Department!poli cy!to!

submit!what!is!known!as!a!‘vehicle!stop!card’!(see!Appendix!2!for!a!copy!of!the!card).!Officers!

use! the! stop! card! to! record! basic! demographic! information! about! the! driver,! including! their!

race,! gender,! age,! and! San! Diego! City! residency,! along! with! the! date,! time,! location! (at! the!

division!level),!an d!reason!for!the!stop.!There!are!also!fields!for!tracking!what!we!term!‘post-

stop!outcomes,’!including!whether!the!interaction!resulted!in:!!

• the!issuance!of!a!citation!or!a!warning;!

• the!initiation!of!a!field!interview;!

• a!search!of!the!driver,!passenger(s),!and/or!vehicle;!!

• the!seizure!of!property;!

• discovery!of!contraband;!and/or!

• an!arrest.!!

Lastly,!the!stop!card!gives!officers!space!to!provide!a!qualitative!descripti on !of!the!encounter.!

When!included,!these!data!tend! to !explain!why!a!particular!action!was!taken!or!to!describe!the!

type!of!search!conducted!or!contraband!discovered.!!

Compared!to!other!cities,

!the!vehicle!stop!card!is!a!solid!tool!for!tracking!officer!activity!and!

for!identifying!trends!in!the!enforcement!of!existing!traffic!law.!As!we!will!discuss!in!Chapter!6,!

however,! there! is! substantial! room! to! improve! the! SDPD’s! current! data! collection! efforts.!

Regardless! of! what! this! system! looks! like,! the! Department! should! consider! including! several!

data!points!recommended!by!the!U.S.!Department!of ! Justice.

!The!most!important!potential!

additions!include:!!

• race/ethnicity!and!gender!of!the!officer!involved;!!

• specific!geo-location!of!the!stop/search;!!

!See,!for!example,!Engel,!R.S.,!Tillyer,!R.,!Cherkauskas,!J.C.,!& !Frank,!J.!(2001,!Nov.!1).!Traffic$Stop$Data$Analysis$

Study:$Year$1$Final$Report.!Cincinnati,!OH:!University !o f!Cin cin n at i!Po licin g!In stit ute .!R etrie v ed !Se pt .!5,!2 01 6 ,!from!

http://www.azdps.gov/about/reports/docs/Traffic_Stop_Data_Report_2007.pdf.!!

!McMahon,!J.,!&!Kraus,!A.!(2005).!A$suggested$approach$to$analyzing$racial$profiling:$Sample$templates$for$

analyzing$car-stop$data.!Washington,!D C :!Off ice !of!C o m m u n ity !O rien te d!P o licin g!S erv ice s,!U .S .!De p art m e nt !of!

Justice.!Retrieved !Au g.!12 ,!201 6!fro m !http://ric-zai-inc.com / Pu b lic a tio n s/ co p s-p071-pub.pdf.!

15!

• make,!model,!and!vehicle!condition;!and!!

• driver/passenger!demeanor.!

While! our! analysis! was! l imited! by! the! absence! of! this! information,! the! incomplete! and!

inconsistent! quality! of! the! data,! which! we! discuss! i n! the! following! section,! was! a! more!

substantial!challenge.!!

Missing!and!inconsistent!data!

Of!the!several!chal lenges!we!faced!in!converting!the!raw!files!we!received!from!the!SDPD!into!a!

reliable!dataset,!missing!data!was!the!most!significant:!19.0!percent!of!the!combined!259,569!

stop! records! submitted! in! 2014! and! 2015! were! missing! at! least! one! piece!of! information.! As!

Table!3.1!shows,!the!data!were!comprehensive!on!driver!race/ethnicity!and !gender,!as!well!as!

the!date,!time,!location,!and!reason!for!the!stop,!but!were!less!so!in!documenting!the!driver’s!

age!and!residency!status.!!

Several!post-stop!variables!also!contained!high!levels!of!missing!data,!including!information!on!

whether! a! citation! was! issued! (10.6! percent),! and! whether! the! driver! was! subject! to! a! field!

interview!(7.9!percent)!or!a!search!(4.4!percent).!There!was!also!an!exceedingly!high!number!–!

93!percent!–!of!missing!cases!associ ated! with!the!discovery!of!contraband! and!the!seizure!of!

property,! raising! questions! about! the! reliability! of! these! data.! This! may! be! reflective! of! the!

database!management!rather!than!either!officer!carelessness!or!non-compliance.!For!example,!

an!officer!simply! may! not! h ave! filled!out! a!response! for!contraband,! which!would ! have!been!

irrelevant!if!a!search!did!not!occur!during!a!stop.!!

! !

16!

Table!3.1.!

Information!missing!from!the!2014!and!2015!datasets!

Stop!Feature!

2014!

2015!

Demographic/stop!description!

Driver!race!

222!(0.2%)!

2!(<0.1%)!

Driver!age!

8,655!(6.0)!

0!(0.0)!

Driver!gender!

213!(0.2)!

232!(0.2)!

Residency!status!

4,622!(3.2)!

11,372!(9.9)!

Stop!location!

3,160!(2.3)!

3,315!(2.9)!

Reason!for!stop!

212!(0.2)!

0!(0.0)!

Stop!time!

482!(0.3)!

408!(0.4)!

Stop!date!

0!(0.0)!

Post-stop!outcomes!

Citation!issued!

11,126!(7.7)!

16,352!(14.2)!

Field!interview!conducted!

4,045!(2.8)!

16,352!(14.2)!

Search!conducted!

2,044!(1.4)!

9,447!(8.2)!

Contraband!discovery!

132,782!(92.1)!

109,420!(94.8)!

Property!seized!

132,806!(92.1)!

109,459!(94.8)!

Arrest!

1,872!(1.3)!

8,845!(7.7)!

2014:!N!=!144,164;!2015:!N!=!115,405!

Analyzing!patterns!of!missing!data!can!help!explain!how!and!why!the!omissions!occurred!and!

provide!some!insight!into!what!they!mean!for!the!reliability!of!the!dataset!and!its!effect!on!the!

broader!analysis.!!

Figure!3.1!tracks!changes!in!the!volume!of!missing!demographi c!and!post-stop!data!over!time.!

Of! all! stop! cards! submitted! in! 2014,! 17.4! percent! were! missing! at! least! one! p iece! of!

information.

! Nine! percent! were! missing! demographic! data,! 6.1! percent! were! missing! only!

post-stop!data,!and!2.3!percent!were!missing!some!of!both.!!

! !

!This!figure!does!no t!in c lu d e!d a ta !fro m!either!the!‘c o n tra b a n d!d is co v e ry’!o r!‘p ro p e rty !se ize d ’!va ria b les .!

17!

Figure!3.1.!

Tracking!missing!data,!by!month!

Note:!Figure!3.1! does!not! include!figures! for! data! missing! from! either! the! ‘contraban d! discovered’! o r! ‘p ro perty!

seized’!variables.!!

In!2015,!21.1!percent!of!stop!cards!were!missing!at!least!one!pi ece!of!information,!with!nearly!

half!of!those!missing!both!demographic!and!post-stop!information.!A!significant!s pike!of!stop!

cards! missing! both! field! interview! and! citation! data! occurred! between! March! and! August! of!

that! year,! raising!questions! about! the!quality! of!these! data! during!that! period. ! We!also! note!

that!the! vol ume!of!missing!data!increased!as!monthly!stop!totals!reached!their!lowest!levels.!In!

other!words,!the!quality!of!the!stop!card!data!declined!across!the!year!along!with!the!number!

of!both!recorded!stops!and!searches.!!

Table!3.2!lists!missing!data!by!patrol!division.!The!high est!percentage!of!incomplete!stop!cards!

were!filed!in!the! Southeastern! division! (24.1!percent),! followed! by!the! Central! (21.1!percent)!

and!Southern! d ivi sion s! (20.0!percent).! These! findings,!together!with! the! data!shown!in! Table!

3.3,!which!lists!missing!records!by!driver!race,!suggest! th at!this!dataset!does!not!provide!the!full!

picture!of!traffic!stops!in!San!Diego,!particularly!of!those!involving!minority!drivers!and!drivers!

stopped!in!divisions!located!below!Interstate!8.!As!we!noted!previously,!this!is!the!exact!pattern!

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

10%

15%

20%

25%

30%

35%

Jan!2014

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Jan!2015

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Stop!card!volume

Missing!data!(%)

Demographic!data Post-stop!data Some!of!both Stop!cards

18!

that! prompted! Gary! Cordner! and! his! colleagues! to! question! the! validity! and! reliability! of! the!

2000!and!2001!data.

Table!3.2.!

Incomplete!stop!cards!submitted!in!2014!and!2015,!by!police!div ision!!

Stop!cards!

submitted!

Missing!

demographic!data!

Missing!post-

stop!data!

Missing!both!

types!of!data!

Total!

incomp le t e !

Above!Interstate!8!

!!!!!Northern!

37,203!

1,872!(5.0%)!

3,567!(9.6%)$

965!(2.6%)$

17.2%$

!!!!!Eastern!

31,788!

1,505!(4.7)!

2,217!(7.0)$

1,467!(4.6)$

16.3!

!!!!!Northwestern!

16,306!

903!(5.5)!

802!(4.9)$

784!(4.8)$

15.3$

!!!!!Western!

30,078!

1,247!(4.1)!

2,242!(7.5)$

784!(2.6)$

14.2$

!!!!!Northeastern!

31,692!

950!(3.0)!

1,242!(3.9)$

1,020!(3.2)$

10.1$

!!!!!Sub-total!

147,067!

6,477$(4.4)!

10,070$(6.8)$

5,020$(3.4)$

14.7$

Below!Interstate!8!

!!!!!Southeastern$

19,292$

1,773!(9.2%)$

1,866!(9.7%)$

1,002!(5.2%)$

24.1%$

!!!!!Central!

29,692!

1,429!(4.8)!

3,070!(10.3)$

1,756!(5.9)$

21.1$

!!!!!Southern!

29,351!

705!(2.4)!

1,362!(4.6)$

3,791!(12.9)$

20.0$

!!!!!Mid-City!

27,692!

1,309!(4.7)!

2,304!(8.3)$

1,034!(3.7)$

16.8$

$$$$$Sub-total!

106,027!

5,216$(4.9)!

8,602$(8.1)$

7,583$(7.2)$

20.2$

City-wide$total!

253,094!

11,693$(4.6)!

18,672$(7.4)$

12,603$(5.0)$

17.0$

Note!1:!Missing!data!do!not!include!variables!indicating!the!discovery!of!contraband!or!property!seizure.!

Note!2:!Tab le!3.2!does!not!include!the!6,475!stop!records!submitted!without!stop!location!information,!which!

explains!the!discrepancy!between!the!city-wide!totals!listed!h ere!an d !tho se!refren ced!elsewhere!in!the!Report.!!!

The!frequent!incidence!of!missing!data!reduced!the!quality!of!our!analysis!and!raises!concerns!

over! whether! the! stop! card! records! provide!a! complete!picture! of!traffic! stops! in! San! Diego.!

These! concerns! are! co mpoun ded! by! the!unexplained! changes!in! monthly! traffic! stop! volume!

during!the!time!period!we!analyzed.!

Many!of!the!questions!raised!about!the!q uali ty!of!the!data!used!in!the!2000!and!2001!analysis!

were!driven!by!a!substantial!decrease!–!28.4!percent!–!in!the!number!of!data!cards!s ubmi tted!

between!the!first!and!second!year!of!the!Cordner!team’s!analysis.!We!find!a!si mil ar!pattern!in!

!Cordner,!G.,!W illiams,!B.,!&!Velasco,!A.!(2002).!San$Diego$Police$Department$vehicle$stops$in$San$Diego:$2001.!

San!Diego,!CA.!

19!

the!2014!and!2015!data,!as!is!shown!in!Figure!3.2.!In!2015,!SDPD!officers!compl eted!115,405!

stop!cards,!nearly!20!percent!fewer!than!the!144,164!completed!in!2014.!!

Table!3.3.!

Incomplete!stop!cards!submitted!in!2014!and!2015,!by!driver !race/ethnicity!!

Stop!cards!

submitted!

Missing!

demographic!

data!

Missing!post-

stop!data!

Missing!both!

types!of!data!

Total!

incomp le t e !

Asian/Pacific!Islander!

41,021!

2,625!(6.4%)!

2,429!(6.4%)!

1,922!(4.7%)!

17.5%!

Black!

28,535!

2,136!(7.5)!

2,577!(7.5)!

1,302!(4.6)!

19.6!

Hispanic!

77,934!

5,258!(6.7)!

5,584!(6.7)!

5,563!(7.1)!

20.0!

White!

111,855!

7,051!(6.3)!

8,082!(6.3)!

4,690!(4.2)!

17.7!

Total!

259,345!

17,070!(6.6)!

18,672!(7.2)!

13,477!(5.2)!

19.0!

Note:!These!data!do!not!include!the!224!stop!records!submitted!without!driver!race/ethnicity.!

Data!from!2000!and!2014,!the!first!years!of!each!study,!show!steep!declines!over!the!course!of!

the!year,!whil e!the!volume!in!2001!and!2015!is!substantially!lower,!and!comparatively!flat!from!

month!to!month. !In!January!2000,!SDPD!officers!recorded!20,487!stops,!nearly!twice!the!annual!

low!of!11,094,!from!December!of!that!year.!In!2014,!there!was!a!39!percent!drop!from!14,745!

stops!recorded!in!February,!that!year’s!busiest!month,!to!the!8,988!submitted!in! December,!the!

slowest.! Contrast! that! with! 2001!and! 2015,! where! the! high-to-low! monthly! differences! were!

28.0!percent!and!18.9!percent,!respectively.!!!

Figure!3.3!indicates!that!despite!changes!in!the!volume!of!stop!cards!and!in!the!rate!of!missing!

data! reported,! the! proportion! of! stops! by! race/ethnicity! remained! relatively! stable.! These!

figures! h elp! to! ad dress! some! concerns! that! the! decline! in ! stops! recorded,! and! the! overall!

quality! of! the! data! produced,! may! have! di sprop ortion ately! affected! one! or! more! groups! of!

drivers,!or!that!the!downward!trends!indicate!overt!race-driven!data!manipulation.!!

In! sum,! the! volume! of! stop! cards! submitted! by! SDPD! officers! has! steadily! decli ned! between!

January! 2014! and! December! 2015.! Over! that! same! period,! the! number! of! incomplete! cards!

increased,!with!a!disproportionate!number!involving!traffic!stops!occurring!in!higher-minority!

divisions!located!below!Interstate!8.!We!do!not!know!whether!these!trends!reflect!a!change!in!

SDPD!policy!and/or!leadership,!a!natural!seasonal!shift!in!driving!patterns,!or!some!other!factor.!!

Finally,!we!note!what!appears!to!be!substantial!under-reporting!of!traffic!stops.!On!August!9,!

2016,!we!received!complete!judicial!records!of! ci tatio ns!issued!in!San!Diego!between!January!1,!

2014!and! December! 31,! 2015.!Th ese! records!are!drawn!from! the! p hysical ! citations!issued!by!

20!

SDPD!officers!and!are!wholly!distinct!from!the!vehicle!stop!card!records!that!form!the!basi s!of!

our!broader!analysis.!And!because!traffic!citations!are!subject!to!judicial!oversight,!they!are!a!

more!accurate!reflection!of!officer!activity!than!are!the!stop!card!records,!which!are!not!subject!

to!external!verification.!!!!

Figure!3.2.!

Comparing!monthly!traffic!stop!volume,!by!year!

! !

According! to! these!data,! the! SDPD! issued! 183,402!citations! over! this! two-year! period,!a! sum!

26.1!percent!greater!than!the!145,490!citations!lo gged!by!officers!via!the!traffic!stop!data!card.!

As! is! shown! in! Table! 3.4,! we! used! stop! card! citation! rates! for! each! racial/ethnic! group! to!

generate! rough! estimates! of! unreported! traffic! stops.! All! told,! we! estimate! that! the! SDPD!

conducted! somewhere! between! 60,000! and! 70,000! traffic! stops! for! which! no! stop! card!

information!was!sub mitted.

!We!do! note!that!the!racial/ethnic!composition!of!the! stop! card!

citation!records!largely!reflects!the!composition!of! the!actual!citations!issued,!which! suggests!

that!the!under-reporting!was!not!race-determinative.!!

!These!calculations!reflect!at!least!one!major!assumption.!We!are!forced!to!assume!that!the!SDPD!underreported!

citation!stops!at!the!same!rate!as!non-citation!stops.!Because!we!do!not!have!records!of!warnings!given,!there!is!

no!way!to!confirm!this!one!way!or!another.!We!also!highlight!the!possibility!that!the!discrepancy!b etw ee n!sto p!

card!records!of!citations!issued!and!judicial!records!of!citations!issued!may!reflect!missing!data.!In!fact,!27,478!stop!

cards!issued!in!2014!and!2015!were!missing!information!about!the!issuance!of!a!citation.!!

5,000

10,000

15,000

20,000

25,000

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

2000 2001 2014 2015

21!

Figure!3.3.!

Monthly!traffic!stop!percentages,!by!driver!race/ethnicity!!

Taken! together,! the! missing! and! underreported! data! affect! the! reliability! of! the! stop! card!

dataset.!In!our!recommendatio ns!(Chapter!6),!we!discuss!several!ways!in!which!the!SDPD!might!

enhance! its! data! collection! activities! to! ensure! a! full! and! accurate! record! of! its! traffic!

enforcement!regime.!!

Table!3.4.!

Comparing!judicial!citation!records!with!stop!card!citation!records!!

Stop!cards!

issued!

Stop!card!

citation!records!

Citation!

rate*!

Judicial!citation!

records!

Projected!traffic!

stops!

Asian/Pacific!Islander!

41,021!

23,483!(16.1%)!

57.2%!

33,919!(18.5%)!

59,251!

Black!

28,535!

13,160!(9.1)!

46.1!

17,040!(9.3)!

36,948!

Hispanic!

77,934!

44,165!(30.3)!

56.7!

55,674!(30.4)!

98,243!

White!

111,855!

64,682!(44.5)!

57.8!

76,769!(41.9)!

132,757!

Total$

259,345$

145,490$(100.0)$

56.1$

183,402$(100.0)$

326,926$

*Based!on!2014!and!2015!stop!card !records.!

Note:!The!224!stop!records!submitted!without!driver!race/ethnicity!data!account!for!the!difference!between!the!

totals!listed!in!Table!3.4!and !those!referenced!throughout!the!Report.!

10%

15%

20%

25%

30%

35%

40%

45%

50%

Jan!2014 Apr!2014 July!2014 Oct!2014 Jan!2015 Apr!2015 July!2015 Oct!2015

White Black Hispanic Asian/Other

22!

Contextual!data!collection!

To!supplement!our!examination!of!the!stop!card!data,!we!collected!an!array!of!additional!data!

to!better!understand!what!transpires!during!traffic!stops!as!well!as!to!provide!context!around!

the!state!of!police-community!relations!in!San!Diego.!!

Community!focus!groups!

We!sought!to!capture!San!Diego!residents’!experiences!with!and!perceptions!of!policing!–!and!

of! traffic! sto ps! in! particular! –! through! community! focus! groups.! Focus! group! interviews! are!

useful! for! extracting!detailed!information! about! individuals'!and! groups'! feelings,!perceptions!

and! experiences,! and! are! typically! more! cost-! and! time-effective! than! conducting! individual!

interviews.!Because!focus!groups!can!help!facilitate!a!safe!space!where!participants!can!share!

their!ideas!with!others!of!similar!backgrounds,!the!group! context! can! be! especially! useful! for!

gleaning! information! from! participants! who! otherwise! might! be! reluctant! to! express!

themselves!openly!about!certain!topics.!

The! SDSU! research! team! collaborated! with! Harder+Company,! a! local! research! company! with!

expertise! in! facilitating! such! group! discussions.! We! held! focus! groups! in! four! SDPD! police!

divisions:!Central,!Mid-City,!Southern,!and!Southeastern.!We!selected!these!division s! because!

they! have! the! highest! levels! of! crime,! police! activity,! and! racial/ethnic! diversity.!

Harder+Company! assisted! SDSU! researchers! in! focus! group! recruitment,! staffing,! and!

transcription.!SDSU!researchers!attended!and!observed!focus!groups!and!undertook!qualitative!

analyses!of!the!interview!data.!!

Participants! were! recruited! through! announcements! placed! through! a! variety! of! channels,!

including:! Craigslist,! restaurants,! community! centers,! barber! shops,! libraries,! and! other! local!

businesses.!Selection!criteria!for!focus!group!participation!included!that!participants!must!be:!

• between!the!ages!18!and!55;!!

• comfortable!speaking!in!either!English!or!Spanish;!and!!

• a! current! resident! of! one! of! the! communities! served! by! the! four! identified! SDPD!

divisions.!

Additionally,!during! the! screening!process,!we! oversampled! for!young!adults! (ages! 18!to!30),!

Blacks,! Hispanics,! and! people! who! self-reported! as! regular! drivers.! These! oversampling!

decisions! were! made! based! on! empirical! literature! that! indicates! that! these! are! the!

demographic! groups! most! likely! to! be! stopped! while! driving.! Given! that! the! focus! group!

participants! were! not! randomly! selected! from! the! po pul atio n! of! City! (or! division)! residents,!

findings!from!our! discussions! are!therefore!not!necessarily!representative! of!all!residents’!(or!

23!

those! divisions’! residents’)! perceptions.! Although! our! sampling! technique! is! a! common! and!

appropriate! one! for! this! type! of! qualitative! research,! it! limits! our! ability! to! generalize! the!

findings!or!draw!inferences!to!the!larger!population.!!

During!the!Spring!and!Summer!of!2016,!we!held!10!community!focus!groups!with!a!total!of!50!

participants.!Table!3.5!summarizes!the!number!of!participants!by!police!division.!Due!to!having!

to! comply! with! Institutional! Review! Board! requirements! regarding! protection! of! ou r!

participants’! identities,! we! were! unable! to! capture! precise! demographics.! We! captured! this!

information!during!the!recruitment! and! screening!process,!but!in!order!to ! ensure! anonymity,!

we! were! unable! to! verify! participants’! identities.! However,! of! the! 55! people! who! expressed!

interest! in! participating! and! met! ou r! screening! criteria:! 21.8%! identified! as! Black! or! African-

American;!32.7%!identified! as! Hispanic! or!Latino;! 31%!identified! as!White! or!Caucasian;! 3.6%!

identified! as! Asian-American;! and! 11%! identified! as! another! race/ethnicity! not! otherwise!

captured.!

Focus!group! questions! sought!to!gather! information! about!community!residents’! perceptions!

of:!!

• community!safety;!

• the!visibility!and!presence!of!police;!!

• the!extent!to!which!residents!trust!the!police;!

• experiences!being!stopped!by!the!police!while!driving;!!

• how!race/ethnicity!shapes!interactions!with!the!police;!and!!

• what!improved!police-community!relationships!might!entail.!!

Focus!group!participants!were!provided!a!light!meal!and!a!$20!gift!card.!

Table!3.5.!

Focus!groups!and!participants!

Division!

Number!of!groups!

Participants!

Central!!

10!

Mid-City!

24!

Southern!

12!

Southeastern!

Total$

10$

50$

24!

Officer!survey!

From! May! to! June! 2016,! the! SDSU! research! team! conducted! a! department-wide,! electronic!

survey!of!all!1,867!active!SDPD!officers.!Table!3.6!lists!basic!descriptive!information!for!the!365!

respondents!(response!rate!=!19.5!percent).!Officers!were!asked!about!several!pertinent!issues,!

including:!!

• the!extent!to!which!they!believe!San!Diego!residents!trust!the!police;!

• whether!recent!events! involving! the!police! nationally!(e.g.,!Ferguson,!MO)! have! made!

their!jobs!more!difficult;!

• the!process!of!collecting!traffic!stop!data;!

• how!race/ethnicity!shapes!police!interactions!with !the!public–both!generally!and!in!the!

context!of!traffic!stops;!and!

• how!the!SDPD!handles!the!issue!of!racial/ethnic!bias,!both!in!training!its!officers!and!in!

handling!incidents!of!race-based!misconduct.!

Table!3.6!

Descriptive!statistics!for!police!officer!survey!respondents!!

Frequency!

Percent!

Race/ethnicity!

Asian$

11$

3.0$

Black$

2.5$

Hispanic$

51$

14.0$

White$

203$

55.6!

Other$

47$

12.9$

No!response$

44$

12.1$

Rank!

Police!Officer!(patrol)$

179$

49.0!

Sergeant!or!above$

141$

38.6!

Other$

1.9!

No!response!

38$

10.4$

Experience!(years)!

1!or!fewer!!$

1.1$

Between!2!and!5$

47$

12.9!

Between!6!and!10$

62$

17.0$

Between!11!and!20$

97$

26.6$

21!or!more!!$

120$

32.9!

No!response$

35$

9.6$

25!

A!full!copy!of!the!survey!is!found!in!Appendix!3.!

Officer!interviews!

Lastly,! during! June! 2016,! the! SDSU! research! team! also! conducted! in-depth,! one-on-one!

interviews!with!52!SDPD!officers!drawn!from!each!of!SDPD’s! n ine!patrol!divisions!as!well!as!the!

city-wide!traffic!division.!Most!interviews!lasted!between!30!and!60!minutes!and!were!intended!

to!delve!deeper!into!the!topics!covered!by!the!department-wide!survey.!We!also!asked!several!

of!the!same!questions!of!officers!as!we!did!of!community!residents!in!focus!groups!to!identify!

similar!and!divergent!perspectives!across!these!groups.!Particularly,!we!sought!to!hear!directly!

from!officers!about:!

• their!perceptions!of!community!safety!and!trust!in!the!police;!

• procedures!followed!during!traffic!stops,!including!how!stop!data!are!collected;!

• how! race/ethnicity! is! and! is! not! used! in! policing,! including! what! training! they! receive!

around!these!issues;!!

• difficulties!officers!encounter!in!doing!their!jobs;!and!

• what!can!and!should!be!done!to!improve!police-community!relations.!!

We!do!not!present!the!full!results!from!each!of!these!three!additional!so urces!of!data!in!this!

Report.! Rather,! in! Chapter! 6,! we! draw! on! our! findings! from! these! data! to! contextualize! and!

support!our!recommendations!to!the!Department.! !

26!

CHAPTER!4:!EXAMINING!THE!DECISION!TO!INITIATE!A!TRAFFIC!STOP!

Introduction!

Police!officers!in!the!United!States! d o!their!jobs!with!considerable!independence.!They!typically!

operate! outside! the! view! of! their! supervisors! and! are! often! the! only! source! of! information !

about!their!conduct.!Though!guided!by!federal,!state,!and!local!law,!as!well!as!organizational!

rules!and!norms,!they!alone!are!responsible!for!determinin g!which!drivers!to!stop,!how!best!to!

make! an! arrest,! and! when! to! call! for! backup,! among! countless! other! decisions.! This!

discretionary!authority!undergirds!the!American!criminal!justice!system;!it!fills!the!gaps!created!

by! a!society! with! insufficient! resources! to! support! full! enforcement!of! the! existing! corpus! of!

criminal!and!administrative!law.!!

The! discretionary! authority! granted! to! police! officers! also! forces! citizens! to! accept! a! certain!

degree!of!inequality.!Often,!one!driver!is!stopped!while!another!going!at!a!similar!speed!is!not!

stopped.!Most!rolling!stops!and!illegal!U-turns!are!done!outside!the!view!of!the!police,!and!thus!

go!un-enforced.!Those! who!are!stopped!and!ti cketed!for!such!infractions!are!the!exception,!and!

thus! may,! rightly! or! wrongly,! see! their! ticket! as! the! product! of! selective! enforcement! or!

prejudice.!Yet!only!the! of ficer!knows!for!sure!why!he!or!she!decided!to!stop!one!car!as!opposed!

to!another.!It!is !nearly!impossible!to!determine!why!these!decisions!are!made!in!the!way!that!

they!are.!!

For! this! reason,! rather! than! focusing! on! individual! stop! decisions,! we! analyze! the! entire!

population!of!individual!decisions!in!an!effort!to!identify!larger!trends.!It!i s!through!this!broader!

lens!that!we!attempt!to!determine!whether!stop!patterns!vary!by!race/ethnicity!and!whether!

such! variance! is! indicative! of! systemic! disparities! in! the! way! SDPD! officers! enforce! the! City’s!

traffic!laws.!!

In!February!2015,!SDPD!Police!Chief!Shelley!Zimmerman!presented!to!the!City!Council’s!Public!

Safety! and! Livable! Neighborhoods! Committee! a! report! that! addressed! the! SDPD’s! traffic!

enforcement!in! 2014.

!These! data!showed!disparities!between!actual!driver!stop!rates!and!the!

stop! rates! one! would! expect! given! the! City’s! racial/ethnic! composition:! Black! and! Hispanic!

drivers! were! stopped! more! than! their! demographic! profile! would! predict,! while! White! and!

Asian/Pacific! Islander! drivers! were! stopped! less.! As! is! shown! in! Figure! 4.1,! these! disparities!

carried!over!into!2015.!!

!City!of!San!D iego,!Report!to!the!City!Council,!Public!Safety!&!Livable!Neighborhoods!Committee.!(2015,!Feb.!13).!

Vehicle!Stop!Data!Cards:!January!through!December,!2014.!Report!No:!15-016.!Retrieved!Sept.!5,!2016,!from!

http://docs.sandiego.gov/councilcomm_agendas_attach/2015/psln_150225_3.pdf.!

27!

Figure!4.1.!!

Comparing!driver!stop!rates!in!2014!and!2015!with!San!Diego’s!racial/ethnic!composition!!

Yet! these! differences! provide! very! little! if! any! insight! into! whether! there! are! racial/ethnic!

disparities!in!how!traffic!stop!decisions!are!made!by! SDPD! officers.! Consider! that! in! 2014,! 65!

percent!of!d rivers!stopped!were!male,!despite!the!fact!that!males!comprise!only!51!percent!of!

the! City’s! population,! according! to! the! 2010! U.S.! Census.

! Perhaps! this! disparity! is! in! fact!

because!SDPD!offi cers!are!more!proactive!in!targeting!men!than! women.!It!may!also!reflect!the!

fact!that!more!men! than!women!drive!on!city!streets,!that!men!are!more! li kely!to!violate!traffic!

laws,!or!that!more!men!drive!in!areas!heavily!populated!by!law!enforcement,!and!are!thus!more!

likely!to!be!observed! viol atin g!the!law.

!In!other!words,!s ome!drivers!run!a!greater!risk!of!being!

stopped!than!o thers,!for!reasons!having!nothing!to!do!with!their!gender.!The!same!logic!should!

define!our!thinking!about!driver!race.

!Census!viewer:!San!Diego,!California!population:!Census!2010!and!2000!interactive!map,!demographics,!

statistics,!quick!facts.!Retriev ed !S ep t !28 ,!2 01 6 ,!fro m!http://censusviewer.com/city/CA/San!Diego.!

!See!Fridell,!L.A.!(2004).!By$the$numbers:$A$guide$for$analyzing$race$data$from$Vehicle$Stops.!Washing to n ,!D.C .:!

Police!Executive!Research!Forum;!Ridgeway,!G.,!(2009).!Cincinnati$Police$Department$traffic$stops:$Applying$RAND’s$

framework$to $ana lyze $racia l$dispa rities .!Sa n ta!Monica,!CA:!RA N D !C o rp ora tio n .! !

!Ridgeway,!G.!(2009).!Cincinnati$Police$Department$traffic$stops:$Applying$RAND’s$framework$to$analyze$racial$

disparities.!Santa!Monica,!CA:!RAN D !C o rp ora tio n .! !

10%

15%

20%

25%

30%

35%

40%

45%

50%

Asian/other Black! Hispanic White

2014 2015 Citywide!population

28!

As!a!result,!to!properly!assess!the!effect!that!a!driver’s!ra ce/ethni city!has! o n!the! l ikelih ood !that!

he!or!she!will!be!stopped,!researchers!must!develop!a!benchmark!that!enables!the!co mparison !

of!actual!stop!rates!with!a!driver’s!risk!of!being!stopped!in!the!absence!of!bias.

!An!appropriate!

benchmark! must! incorporate! the! various! legal! and! non-legal! factors! that! shape! stop! risk,!

including!when,!where,!and!how!often!they!drive,!the!make,!model,!and!condition!of!their!car,!

and!their!behavior!and!demeanor!while!driving.

The! most! common! approach! to! thi s! challenge! has! been! to! draw! on! U.S.! Census! fi gures! to!

capture!a!j urisd ictio n’s!demographic!profile!and!then!use!these!data!to!make!inferences!about!

the!city’s!driving!population.

!Though!inexpensive!and!relatively!easy!to!i mpl ement,!the!u se!of!

Census! data! has! come! under! heavy! criticism! for! its! inability! to! accurately! reflect! not! only! a!

jurisdiction’s! driving! population,! but! the! various! other! risk! factors! at! play.

! Other! statistical!

proxies,!including!drivers’!license!data

!and!no-fault!traffic!accident!figures,

!have!also!been!

used!to!address!these!limitations.!!

Other! researchers! have! made!efforts! to!observe! the! characteristics! of! the! driving! population!

first!hand.!Rather!than!relying!on!outside!information!as!the!benchmark,!some!have!attempted!

to! chart! the! demographic! profile! of! a! jurisdiction’s! drivers! at! various! locations! an d! times! of!

day.

!The!observational!approach!is!both!expensive!and!time-consuming,!and!not!without!its!

own!challenges.

!Tillyer,!R.,!Engel,!R.S.,!&!Cherkauskas,!J.C.!(2010).!Best!practices!in!vehicle!stop!data!collection!and!analysis.!

Policing:$An$International$Journal$o f$P olic e $Str at eg ies $& $Management,!33(1),!69-92.!

!Fridell,!L.A.!(2004).!By$the$numbers:$A$guide$for$analyzing$race$data$from$Vehicle$Stops.!Washington,!D.C.:!Po lice !

Executive!Research!Forum;!Ridgeway,!G.!&!MacDonald,!J.!(2010).!Methods!fo r!assessing!racially!biased!policing.!In!

S.K.!Rice!&!M.D.!White!(Ed s.)!Race,$ethnicity,$and$policing:$New$and$essential$readings!(pp.!180-204).!New!York:!

New!York!University!Press;!Tillyer,!R.,!Engel,!R.S.,!&!Cherkauskas,!J.C.!(2010).!Best!practices!in!vehicle!stop!data!

collection!and!analysis.!Policing:$An$International$Journal$of$Police$Strategies$&$Management,!33(1),!69-92;!and!

Walker,!S.!(2001).!Searching!for!the!denominator:!Problems!with!police!traffic!stop!data!and!an!early!warning!

system!solution.!Justice$Rese a rc h$a n d $Po lic y,!3,!63-95.!

!Cordner,!G.,!W illiams,!B.,!&!Zuniga,!M.!(2001).!San$Diego$Police$Department$vehicle$stop$study:$Yea r$end$report.!

San!Diego,!CA,!p.!ii;!Cordner,!G.,!Williams,!B.,!&!Velasco,!A.!(2002).!San$Diego$Police$Department$vehicle$stops$in$

San$Diego:$2001.!San!Diego,!CA .!

!Engel,!R.S.,!Frank,!J.,!Klahm ,!C.F.,!&!Tillyer,!R.!(2006,!Jul.).!Cleveland$Division$of$Police$Traffic$Stop$Data$Study:$

Final$Report.!Cincinnati,!OH:!Univ e rsity !of !Cin cin n ati!D iv isio n !of!C riminal!Justice.!!

!Fridell,!L.A.!(2004).!By$the$numbers:$A$guide$for$analyzing$race$data$from$Vehicle$Stops.!Washington,!D.C.:!Po lice !

Executive!Research!Forum.!

!Alpert,!G.P.,!Dunham,!R.G.,!&!Smith,!M.R.!(2007).!Investigating!racial!profiling!by!the!Miami-Dade!police!

department:!A!multimethod!approach.!Criminology$&$Public$Policy,!6,!25-56.!

!E.g.,!Lamberth,!J.C.!(2013,!Sept.).!Final$Report$for$the$City$of$Kalamazoo$Department$of$Public$Safety.!West!

Chester,!PA:!Lamberth!Consulting.!

!Engel,!R.S.,!&!Calnon,!J.M.!(2004).!Comparing!benchmark!methodologies!for!po lice-citizen!contacts:!Traffic!stop!

data!collection!for!the!Pennsylvania!State!Police.!Police$Quarterly,!7,!97-125.!

29!

We! address! the! problem! of! whether! race/ethnicity! impacts! police! decisions!to! initiate!traffic!

stops!by!employing!a!technique!known!as! the! “veil! of! darkness”! method.

!What!follows!is!a!

description!of!this!method!and!a!detailed!analysis!of!our!findings.!

The!Veil!of!Darkness!Technique!

The! veil! of! darkness! technique! allows! the!

researcher! to! compare! the! racial/ethnic!

distribution!of!traffic!stops!made!in!daylight!

with! that! of! stops! made! after! dark.

! The!

approach! rests! on! the! assumption! that! if!

driver! race/ethnicity! is! a! factor! in!

determining!who! will!be!stopped,! it! will!be!

more! apparent! among! stops! made! in!

daylight,! when! drivers’! physical! profile! is!

more! likely! to! be! detectable,! than! at! night!

when!these! characteristics! are!obscured!by!

darkness.

! We! do! not! suggest! that!

race/ethnicity! is! somehow! impossible! to!

discern! at! night! or! a! certainty! during! the!

day;!rather,!that!“ the!rate!of!police!knowing!driver!race/ethnicity!in!advance!of!the!stop!must!

be!smaller!at!night!than!during!daylight.”

The!strongest!argument!for!this!approach!comes!from!researchers!who!have!tried!to!measure!

driver!race/ethnicity!at!night.!According!to!a!2003!analysis!of!traffic!law!enforcement!i n!Santa!

!E.g.,!Grogger,!J.!&!Ridgeway,!G.!(2006).!Testing!for!racial!profiling!in!traffic!stops!from!behind!the!veil!of!

darkness.!Journal$of$the$American$Sta tistica l$As so cia tion,!101(47 5),!87 8-887.!Retrieved!Aug.!24,!2016,!from!

https://www.rand.org/content/dam/rand/pubs/reprints/2007/RAND_RP1253.pdf;!Ridge way,!G.,!(200 9 ).!Cincinnati$

Police$Department$traffic$stops:$Applying$RAND’s$framework$to$analyze$racial$disparities.!Santa!Monica,!CA:!RA N D !

Corporation;!W orden,!R.E.,!McLean,!S.J.,!&!Wheeler,!A.P.!(2012).!Testing!for!racial!profiling!with!the!veil-of-

darkness!method.!Po lice$Quarterly,!15,!92-111.!

!Ridgeway,!G.,!(2009).!Cincinnati$Police$Department$traffic$stops:$Applying$RAND’s$framework$to$analyze$racial$

disparities.!Santa!Monica,!CA:!RAND!Corporation.!!

!This!assumption!is!potentially!complicated!by!several!unknown!factors,!including!th e!presence!or!absence!of!

ambient!light,!glare,!shadowing,!heavily!tinted!windows,!and!so!on,!at!the!time!of!the!stop.!Interestingly,!the!one!

study!to!control!for!ambient!light!found!evidence!of!racial!disparity!when!the!effects!of!street!lights!were!

accounted!for!and!no!evidence!of!racial!disparity!when!no!such!controls!w ere!included!in!the!veil!of!d arkness!

analysis.!See!Horrace,!W.C.,!&!Rohlin,!S.M!(2016).!How!dark!Is!dark?!Bright!lights,!big!city:!Racial!profiling,!Review$

of$Economics$and$Statistics,!98,!226 -232.!Retrieved!Oct.!24,!2016,!from !

https://pdfs.semanticscholar.org/84ff/4695f264da05e69cbc4e3e5dbd794bf9e298.pdf.!

!Ridgeway,!G.,!(2009).!Cincinnati$Police$Department$traffic$stops:$Applying$RAND’s$framework$to$analyze$racial$

disparities.!Santa!Monica,!CA:!RAN D !C o rp ora tio n ,!p.!1 2. !

The! “veil! of! darkness”! technique! allows!

researchers! to! avoid! the! difficulty! of!

identifying! and! applying! a! benchmark! –! a!

point! of! reference,! such! as! Census! data! –!

against! which! to! compare! traffic! s top! data.!

This! is! the! central! challenge! in! the! review! of!

such!data,!as!the!driving!population!of!a!given!

area! may! look! quite! different! f rom! the!

residents! of! that! area,! as! counted! by! the!

Census.! Instead,! using! the! veil! technique,!

analysts! can! examine! the! likelih ood! that,! for!

example,! Black! drivers!will!be! stopped! during!

the! day! versus ! at! night,! and! compare! that!

likelihood!with!the!day-versus-night!likelihood!

of!White!drivers!being!stopped.!!

30!

Cruz,!California,!the!most!difficult!observational!conditio ns!occur!either!at!dawn!or!dusk!“or!in!

dark!areas!where!no!su ppl emental!lighting!is!provided.”

!As!a!result,!study!authors!relied!on!

the!use!of!su pp lemental!lighting!to!enhance!driver!visibility!during!these!periods.!That!the!use!

of! supplemental! lighting! has! become! commonplace! among! observational! researchers!

underscores!the!point.

!Others!report!having!to!eliminate!nighttime!observations!altogether,!

finding!“reliable!data!collection!on!the!race/ethnicity!of!the!driver…![to!be]!impossible”!at!dusk!

and!after!sundown.

Table!4.1.!!

Previous!research!employing!the!veil!of!darkness!analytical!approach!

Author(s)/Year!

Jurisdiction!

Time!Period!Analyzed!

Day-night!

Disparity!Found?!

Grogger!&!Ridgeway!(2006)!

Oakland,!CA!

Jun!2003!– !Dec!2003!

No!

Ridgeway!(2009)!

Cincinnati,!OH!

2003!-!2008!

No!

Worden!et!al.!(2012)!

Syracuse,!NY!

2006-2009!

No!

Ritter!(2013)

Minneapolis,!MN!

2002!

Yes!

Horrace/ethnicity!&!Rohlin!(2014)!

Syracuse,!NY!

2006-2009!

Yes!

Ross!et!al.!(2016)

State!of!CT!

Oct!2013!–!Sept!2014!

Yes!

Taniguchi!et!al.!(2016)

Durham,!NC!

Jan!2010!–!Oct!2015!

Yes!

The!challenge!of!accurately!categorizing!a!driver’s!race/ethnicity!at!night!is!also!con sistent!with!

research! o n! the! validity! of! eyewitness! testimony.!To! summarize! years! of! research,! witnesses!

!Rickabaugh,!C.A.!(2003,!Sept.).!A$study$to$analyze$traffic$stop$data$in$Santa$Cruz$County.!Chadds!For d,!P A :!

Lamberth!Consulting,!p.!30.!

!E.g.,!Lange,!J.E.,!Johnson,!M.B.,!&!Voas,!R.B.!(2005).!Testing!the!racial!profiling!hypothesis!for!seemingly!

disparate!traffic!stops!on!the!New !Jersey!turnpike.!Justice$Quarterly,$22,!1 9 3 -223;!Lamberth,!J.C.!(2013,!Sept.).!

Final$Report$for$the$City$of$Kalamazoo$Department$of$Public$Safety.!West!Chester,!PA :!La m b er th!C o ns u lting . !

!Alpert,!G.P.,!Dunham,!R.G.,!&!Smith,!M.R.!(2007).!Investigating!racial!profiling!by!the!Miami-Dade!police!

department:!A!multimethod!approach.!Criminology$&$Public$Policy,!6(1),!25-56,!p.!36.!

!Ritter,!J.A.!(2013).!Racial!bias!in!traffic!stops:!Tests!of!a!u nified!model!of!stops!and!searches.!University!of!

Minnesota!Population!Center,!Working!Paper!No.!2013-05.!Retrieved!Oct.!24,!2016,!fro m!

http://ageconsearch.umn.edu/bitstream/152496/2/WorkingPaper_RacialBias_June2013-1.pdf.!

!Ross,!M.B.,!Fazzalaro,!J.,!Barone,!K.,!&!Kalinowski.!(2016).!State!of!Connecticut!traffic!stop!data!analysis!and!

findings,!2014-15.!Connecticut!Racial!Profiling!Prohibition!Project.!Retrieved!Oct.!24,!2016,!from!

http://www.ctrp3.org/reports/.!

!Taniguchi,!T.,!Hendrix,!J.,!Aagaard,!B.,!Strom ,!K.,!Levin-Rector,!A.,!&!Zimmer,!S.!(2016).!Exploring$racial$

disproportionality$in$traffic$stops$conducted$by$the$D urham$Police$Department.!Research!Triangle!Park,!NC:!RTI!

Internatio n al. !R et rie ve d !O c t.!2 4 ,!2 0 16 ,!fr o m !

https://www.rti.org/sites/default/files/resources/VOD_Durham_FINAL.pdf.!

31!

are!much!better!at!describing!basic!features!of!criminal!suspects,!including!race/ethnicity!and!

gender,!when!observed!during!daylight!hours!rather!than!at!night.

The! veil! of! darkness!approach! was!first! utilized! by! Grogger! and! Ridgeway! for! their! review! of!

traffic! stops! in! Oakland,! California.

! Since! then,! scholars! have! relied! on! this! technique! to!

examine! data! from! five! other! jurisdictions. ! With! minor! exceptions,! each! of! the! replications!

listed!in!Table! 4.1!followed!Grogger!and!Ridgeway’s!original!method!and!analytical!approach.!

We!follow!suit.!!

To!measure!possible!day-night!disparities,!we!take!advantage!of!a!natural!experiment!produced!

by! seasonal! changes! throu ghout! the! calendar! year.! In! San! Diego,! the! sun! goes! down! earlier!

during!winter!months!than!it!does!in!the!summer.!Someone! driving! home! from! work! at! 6:00!

pm! in! January! woul d! experience! darkness,! but! in! July! the! driver’s! commute! would! occur! in!

broad!daylight.!!

The!analysis!is!confined!to!the!“inter-twilight!period,”!or!the!period!between!the!earliest!end!of!

civil!twilight!(5:09!pm!on!Nov.!27)!and!the!latest!(8:29!pm!on!Jun.!27),!as!defined!by!the!U.S.!

Naval!Observatory,!in!order!to!control!for!changes!in!the!driving!population!du ring!the!course!

of!the!day.

!The!veil!of!darkness!technique!allows!the!analyst!to!assess!di fferences! between!

daylight! and! darkness!stop! p atterns! within! this! window!of! time.! Furthermore,!because! these!

comparisons!occur!within!the!same!segment!of!the!driving!population !(i.e.,!d rivers!on!the!road!

between!5:09!and!8:29!pm!during!darkness!with!drivers!on!the!road!between!5:09!and! 8:29!pm!

during!daylight),!there!is!no!need!for!an!external!benchmark.!

We!excluded! from!the!analysis!those!stops!that!occurred!between!sundown!(also!as!defined!by!

the! U.S.! Naval! Observatory)! and! the! start! of! ci vil ! twilight! (n=3,349),! as! there! was! no! clear!

strategy!for!determining!whether!these!stops!occurred!in!‘daylight’!or!‘darkness.’

!We!further!

limit! our! sample! by! including! onl y! those! stops! that! occurred! as! a! result! of! either! equipment!

Loftus,!G.!R.!(1985).!Picture!perception:!Effects!luminance!on!available!information!and!information!extraction!

rate.!Journal$of$Exp erimental$Psycho log y:$G en era l,!114,!34 2–356;!!

Meissner,!C.A.,!Sporer,!S.L.,!&!Schooler,!J.W.!(2007).!Person!descriptions!of!eyewitness!evidence.!In!R.C.L.!Lindsay,!

D.F.!Ross,!J.D.!Read,!&!M.P.!Toglia!(Eds.)!The$handbook$of$eyewitness$psychology,$Vol.$II!(pp.!1!–!34).!New!York:!

Psychology!Press;!Yarmey,!A.!D.!(1986).!Verbal,!visual,!and!voice!id entification!of!a!rape!suspect!under!different!

levels!of!illumination.!Jou rna l$of$A p plied $Ps ych o log y,!71,!36 3 –370.!

!Grogger,!J.!&!Ridgeway,!G.!(2006).!Testing!for!racial!profiling!in!traffic!stops!from!behind!the!veil!of!darkness.!

Journal$of$the$A m erica n $Sta tistica l$Ass oc iatio n,!101(47 5),!878 - 887.!Retrieved!Aug.!24,!2016,!from!

https://www.rand.org/content/dam/rand/pubs/reprints/2007/RAND_RP1253.pdf.!

!The!full!schedule!can!b e!found!here:!http://aa.usno.navy.mil/data/docs/RS_OneYear.php. !

!Worden,!R.E.,!McLean,!S.J.,!&!Wheeler,!A.P.!(2012).!Testing!for!racial!profiling!with!the!veil-of-darkness!method.!

Police$Quarterly,!15,!92-111.!

32!

(e.g.,!a!broken!tail!li ght)! or!moving!violations!(e.g.,!an!illegal!left!turn).

!As!is!shown!in ! Table!

4.2,! these! types! of! stops,! which! are! the! product! of! a! highly! discretionary! decision-making!

process,!comprise! the! vast! majority!o f! traffic!stops!in!San! Diego.! Stops!made!as!a! result! of! a!

suspect!description,!an!informant’s!tip,!or!pre-existing!officer!knowledge!are!excluded,!as!they!

involve! a! much! lower! level! of! discretionary! auth ority! and! may! lawfully! include! a! driver’s!

race/ethnicity!as!part!of!the!justification!for!stop.!!

Table!4.2.!

Describing!data!generated!by!traffic!stops!conducted!by!SDPD!officers!in!2014!and!2015,!by!

stop!type!!!

!Stop!type!

2014!

2015!

Total!

High!discretion!!

!!!!!Moving!violation!

103,491!(71.9%)!

86,387!(74.9%)!

189,878!(73.2%)!

!!!!!Equipment!violation!

38,426!(26.7)!

27,453!(23.8)!

65,879!(25.4)!

$$$$$Sub-total$

141,917$(98.6)$

113,840$(98.6)$

255,757$(98.6)$

Low!discretion!!

!!!!!Radio!call!

763!(0.5%)!

497!(0.4%)!

1,260!(0.5%)!

!!!!!Code!violation !

752!(0.5)!

366!(0.3)!

1,118!(0.4)!

!!!!!Prior!knowledge!of!suspect!

277!(0.2)!

263!(0.2)!

540!(0.2)!

!!!!!Suspect!information!

211!(0.2)!

161!(0.1)!

372!(0.1)!

!!!!!Other!

32!(<0.1)!

278!(0.2)!

310!(0.1)!

$$$$$Sub-total$

2,035$(1.4)$

1,565$(1.4)$

3,600$(1.4)$

Total$

143,952$(100)$

115,405$(100)$

259,357$(100)$

Note:!Totals!do!not!include!stop!records!subm itted!without!data!on!stop!type.!Discrepancies!in!the!percentage!totals!are!owed!

to!rounding!error.!

Figure!4.2!is! a!scatterplot!of!the!date!an d! times! of!all!stops!included!in!the!full! sample.! Note!

that!black!markers!represent!those!stops!that!occurred!after!the!end!of!ci vil !twilight,!which!we!

classify! as! occurring! during! darkness.! Grey! markers! represent! daylight! stops,! which! occurred!

prior!to!sunset.!!

!We!note!that!some!have!argued!that!because!some!equipment!violations!(a!broken!tail!light,!for!example)!are!

easier!to!identify!after!dark,!they!should!be!excluded!from!a!veil!of!darkness!analysis!(Worden,!R.E.,!McLean,!S.J.,!&!

Wheeler,!A.P.!(2012).!Testing!for!racial!profiling!with!the!veil!of!darkness!method.!Police$Quarterly,!15,!92-111.).!To!

account!for!this!possib ility,!we!replicated!both!the!citywide!and!location-based!analysis!using!just!m oving!

violations.!The!results,!shown!in!Appendix!4,!showed!no!meaningfully!difference!from!the!analysis!described!

herein.!!!

33!

Our!statistical!analysis!aggregates!an d!averages!all!stops!made!during!the!inter-twilight!period!

in!an!attempt!to!evaluate!day-night!disparities!between!several!driver!categories,!including:

• Black!vs.!White!drivers!

• Young!Black!vs.!Young!White!(25!and!under)!

• Hispanic!vs.!White!

• Young!Hispanic!vs.!Young!White!(25!and!under)!

• Asian/Pacific!Islander!v.!White!

• Young!Asian/Pacific!Islander!vs.!Young!White!(25!and!under)!

Figure!4.2.!!

Scatterplot!of!traffic!stops!included!in!the!veil!of!darkness!analysis!

We!distinguish!drivers!25!and!under!in!light!of!the!consistent!evidence!that!younger!drivers!are!

!As!the!relevant!dependent!variable!is!dichotomous!(w hether!the!stop!occurred!during!daylight!or!after!dark),!we!

rely!on!logistic!regression!m o d els!to!p erfo rm !the!a na lysis.!

End of twilight

Sunset

Jan 1 Summer solsce

(Jun 21)

Standard me

starts (Nov 2)

Standard me

ends (Mar 9)

Dec 31

Clock me

5:00 pm

6:00 pm

7:00 pm

8:00 pm

9:00 pm

Figure 1. Date and Time of Intertwighlight Trac Stops

34!

often!l ess!willing!to!comply!traffic!laws,

!and !tend!to!be!more! reckless!drivers!in!general.

!The!

research! is! also! very! clear! th at! young! people! are! also! more! susceptible! to! criminological!

behavior!than!are!adults!further!into!their!life!course.

67,68

To! account! for! potential! changes! to! the! driving! population! over! time,! our! models! include!

dichotomous!variables!for! each! 15-minute! interval!in! the!3-hour! and!20-minute! inter-twilight!

period.!This!allows!us!to!control!for!the!likelihood!that!the!racial/ethnic!composition!of!drivers!

varies!by!time!of!day.!!

The! driving! population! may! also! change! based! on! the! day! of! the! week! (for! example,! thos e!

people! on! the! road! at! 7:30! pm! on! Friday! evening! may! look! and! act! differently! than! those!

driving!at!7:30!on!a!Tuesday),!so! we!also!include!dichotomous!variables!for!the!day!of!the!week.!

These!adjustments!al lo w!us!to!hold!the!day!of!the!week! con stant,!further!isolating!the!effect!of!

daylight.!Similarly,!to!account!for!seasonal!differences!in!the!driving!population,!we!control!for!

the!effects!of!stop!month!and!stop!location.!

! !

!Yagil,!D.!(1998).!Gender!and!age-related!differe nce s!in!attitu de s!tow a rd!traffic!law s !and !traffic!vio lation s.!

Transportation$Research$Part$F:$Traffic$Psychology$and$Behaviour,!1,!1 2 3 -135;!McCartt,!A.T.,!&!Northrup,!V.S.!

(2004).!Factors!rela ted !to!sea t!be lt!use!a m o ng !fatally!in ju re d !te en a g e!d riv er s.!Jo urn a l$of$Sa fety $Re sea rch ,!35,!29 -38.!

!Lawton,!R.,!Parker,!D.,!Stradling,!S.!G.,!&!Manstead,!A.!S.!R.!(1997).!Self-reported!attitude!towa rds!sp ee din g!an d!

its!possible!c o n se q u en c es !in !fiv e!d iff ere n t!ro a d !co n t ex ts.! Jo u rna l$of$C o m m unity$and$Applied$Social$Psychology,!7,!

153-165;!Lawton,!R.,!Parker,!D.!Manstead,!S.!G.,!&!Strad ling,!A.!S.!R.!(1997).!The!role!of!affect!in!predicting!social!

behaviors:!The!case!of!road!traffic!violations.!Journal$of$Applied$So cia l$Psy ch o log y,!27,!1258-1276.!

!Farrington,!D.P.!(1986).!Age!and!crime.!Crime$and$Justice,!7,!189-250;!Jennings,!W.G.,!&!Reingle,!J.M.!(2012).!On!

the!number!an d !shap e!o f!dev elop m en tal/life-course!violence,!aggression,!and!delinquency!trajectories:!A!state-of-

the-art!review.!Journal$of$Crimina l$Justice,!40,!4 7 2 -489;!Sampson,!R.J.,!&!Laub,!J.H.!(1993).!Crime$in$the$Making.!

Cambridge:!Harvard!University!Press.!

!There!is!also!a!well-established!body!of!research!showing!that!males!are!more!likely!to!engage!in!both!reckless!

(see,!for!example,!Keane,!C.,!Maxim,!P.S.,!&!Teevan,!J.!J.![1993].!Drinking!and!driving,!self-control,!and!gender:!

Testing!a!general!theory!of!crime.!Journal$o f$Res ea rch $in$C rim e$a nd $D elinq u en cy,!30,!30-46)!and!crim ina l!beh av ior!

(Synder,!H.N.![201 2].!Arre st!in!th e!U nited !Sta tes,!19 90 - 2010.!U.S.!Department!of!Justice,!O ffice!of!Justice!

Programs,!Bureau!of!Justice!Statistics.!Retrieved!Sept.!29,!2016,!from!

http://www.bjs.gov/content/pub/pdf/aus9010.pdf).!!To!account!for!the!possib ility!tha t!SD PD !offic ers!m a y!as!a !

result!police!males!d ifferen tly!tha n!th ey!d o!fem a les,!w e!an alyze d!d ay-night!disparities!using!a!sample!of!male!only!

drivers.!The!results,!which!showed!no!meaningful!difference!from!the!mixed!gender!analysis,!are!listed!in!Appendix!

5.!

35!

Results!

Before!presenting!the!results!of!our!traffic!

stop! analysis,! it! may! be! h elpfu l! to! review!

the!metrics!used!to!interpret!the!data.!The!

findings!will!be!presented!in!terms!of!odds!

ratios,! which! indicate! the! odds! (or!

likelihood)!of!daylight!affecting!traffic!stop!

patterns.! An! odds! ratio! of! 1.0! indicates!

that! time! of! day! does! not! influence! the!

odds! of! Black! drivers! being! stopped;! in!

that! case,! they! are! no! more! and! no! less!

likely! to! be! stopped! after! dark! than! they!

are!during!daylight,!compared!to! the!stop!

pattern! of! White! drivers.! A! p ositi ve! odds!

ratio!(>1.0)!suggests!that!Black!drivers!are!

more! likely! to! be! stopped! during! the! day!

than! at! night,! and! thus! may! indicate!

racial/ethnic! disparity.! A! negative! odds!

ratio!(<1.0)!i nd icates!that!Black!drivers!are!

more! likely! to! be! stopped! at! night! than!

during! the! day! (or,! put! anoth er! way,! that!

White! drivers! are! more! likely! to! be!

stopped!in!daylight!than!after!dark).!

Black!Drivers!

Table!4.3!displays!the!results!of!our!analysis!of!discretionary!traffic!stops!conducted!in!the!City!

of! San! Diego! between! January! 1,! 2014! and! December! 31,! 2015! involving! Black! and! White!

drivers.!The!data!show!that!in!2014,!when!driver!race/ethnicity!was!visible,!Black!drivers!were!

nearly!20!percent!more!likely!to!be!the!subject!of!a!discretionary!traffic!stop!than!were!White!

drivers.!When! confined! to!drivers!aged! 25! and!under,!young! Black! drivers!in!2014!were!43.8!

percent!more!likely!to! be!stopped!in!daylight!than !after!dark,!compared!to! youn g!Whites.!These!

findings!are!statistically!significant!at!the!0.01!level!and!thus!indicate!racial/ethnic!disparity!in!

the!distribution!of!traffic!stops.!!

A! p-value! is! commonly! used! measure! of!

statistical! significance.! The! smaller! the! p-

value,!the!more!confidence!we!have!that!the!

results! would! not! occur! under! the! null!

hypothesis! (e.g.,! that! no! relationship! exis ts !

between! an! officer’s! decision! to! stop! a!

particular!driver!and!that!driver’s!race).!!

For!example,!a!p-value!of!0.01!means!that!we!

are!99%!confident! that!our! result!is!not! due!

to!chance.!Following!common!practice!in!the!

social!sciences,!we!report!p-values!of!.05!and!

lower,! which! correspond! to! a! level! of!

confidence! of! 95%! or! higher,! as! statistically!

significant:!

!!!!!!!!!!!!p-value!!!!!!!!!!!Level!of!confidence!!

!!! !!0.001!! !!!!99.9%!!

!!0.01! !!! !!!!99%!!

!!!! !!0.05!! ! !!!!95%!!

36!

Table!4.3.!

Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!citywide!for!

either!a!moving!violation!or!an!equipment!violation!!

Odds!

Ratio!

p-value!

Standard!

error!

95%!

Confidence!

Interval!

Number!

of!Stops!

2014!

!!!!!Black!v.!White!

1.196!

0.005!

0.077!

1.055,!1.356!

8,332!

!!!!!Young!Black!v.!Young!White!

1.438!

0.003!

0.177!

1.129,!1.832!

2,189!

2015!

!!!!!Black!v.!White!

0.800!

0.118!

0.114!

0.605,!1.058!

6,216!

!!!!!Young!Black!v.!Young!White!

0.783!

0.068!

0.105!

0.602,!1.018!

1,631!

Combined!

!!!!!Black!v.!White!

1.052!

0.293!

0.051!

0.957,!1.156!

14,548!

!!!!!Young!Black!v.!Young!White!

1.098!

0.309!

0.101!

0.917,!1.316!

3,820!

These!same!disparities!were!not!present!in!the!2015!data.!When!the!2015!sample!is!limited!to!

stops! involving! drivers! aged! 25! and! younger,! there! is! evidence,! albeit! of! relatively! weak!

statistical! power,! that! Black! drivers! were! less$ likely! to! be! stopped! during! the! day! than! after!

dark.!When!the!2014!and!2015!data!are!combined,!we!find!no!meaningful!statistical!distinction!

between!Blacks!and!Whites.!!

To! further! control! for! potential! seasonal! differences! among! the! drivin g! population,! we! also!

conduct!an!analysis!limited!to!inter-twilight!stops!occurring!30!days!before!and!after!Daylight!

Saving! Time! (DST)! clock! changes,! which! in! 2014! occurred! at! 2:00! am! on! March! 9th! and!

November!2nd.!In!2015,!California!moved!clocks!ahead!on!March!8!and!back!on!November!1.!

Figure!4.3!is!a!scatterplot!of!those!data!included!in!the!2014!DST-only!analysis,!reflecting!traffic!

stops!occurring!d urin g!60-day!periods!in!the!Spring!(Feb.!7th!–!Apr.! 9th) !and!the!Fall!(Oct.!3rd!–!

Dec.!2nd).! The! 2015!DST! period!includes!stops! recorded! between!February!6th!and! April! 8th!

and!between!October!2nd!and!December!1st.!

37!

Figure!4.3.!!

Scatterplot!of!traffic!stops!included!in!the!Daylight!Saving!Time!veil!of!darkness!analysis!

Delimiting!the!analysis!is!a!way!to!evaluate!the!robustness!of!the!findings!discussed!above!and!

to!provide!more!thorough!protection!against!the!influence!of!seasonal!changes!to!the!driving!

population.!The!primary!trade-off!of! this! more! conservative! approach! is! the!loss! of! statistical !

power.!As! Ridgeway! notes,!the!smaller! sample! si zes! required!are!still! large! enough!to!reflect!

significant!day-night!disparities,!but!smaller!differences!may!not!be!as!readily!apparent.

As!is!shown!in!Table!4.4,!our!estimates!shift!somewhat!under!these!more!restrictive!co ndi tio ns,!

with!changes!most!apparent!in!the!2014!data.!When!the!analysis!is!confined!to!stops!occurring!

during!the!DST-only!period,!disparities!between!Black!and!White!drivers!are!no!longer!evident.!

Results! generated! by! analysis! of! the! 2015! and! combined! datasets! remain! substantively!

unchanged:!no!statistical!difference!exists!in!the!likelihood!that!Black!drivers!are!more!likely!to!

!Ridgeway,!G.,!(2009).!Cincinnati$Police$Department$traffic$stops:$Applying$RAND’s$framework$to$analyze$racial$

disparities.!Santa!Monica,!CA:!RAN D ! Corporation.!

Jan 1 Summer solsce

(Jun 21)

Standard me

starts (Nov 2)

Standard me

ends (Mar 9)

Dec 31

Clock me

5:00 pm

6:00 pm

7:00 pm

8:00 pm

9:00 pm

Figure 2. DST Trac Stops

End of twilight

Sunset

38!

be!stopped!by!police!during!daylight!hours!than!they!were!after!dark!when!compared!to!White!

drivers.!

Table!4.4.!!

Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!citywide!for!

either!a!moving!violation!or!an!equipment!violation!during!the!DST!period!

Odds!

ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!

of!stops!

2014!

!!!!!Black!v.!White!

1.109!

0.480!

0.163!

0.831,!1.479!

2,564!

!!!!!Young!Black!v.!Young!White!

1.175!

0.573!

0.336!

0.670,!2.059!

671!

2015!

!!!!!Black!v.!White!

1.184!

0.337!

0.208!

0.839,!1.671!

1,994!

!!!!!Young!Black!v.!Young!White!

0.720!

0.343!

0.249!

0.365,!1.419!

547!

Combined!

!!!!!Black!v.!White!

1.143!

0.233!

0.128!

0.918,!1.423!

4,558!

!!!!!Young!Black!v.!Young!White!

0.951!

0.816!

0.206!

0.621,!1.455!

1,218!

Though! we! include! controls! for! stop! location! in! the! citywide! models,! for! several! reasons! we!

believe! there! is! value! in! taking! a! closer! look! at! division-level! differences! in! the! treatment! of!

Black!and!White!drivers.!First,!as!shown!in!Figure!4.4,!th ere!appears!to!be!a!loose!relationship!

between! division-level! stop! rates! and! the! localized! crime! rates! (Pearson’s! r! =! 0.5134).! This!

relationship! suggests! that! patrol! strategies! in! higher-crime! areas,! like! the! Central! division,!

which!is!home!to!both!the!city’s!highest!crime!rate!and!highest!stop!rate,!will!be!substantially!

different!than!in!the!Northern!division,!where!both !crime!and!stop!rates!are!closer!to!citywide!

averages.! In! addition! to! other! factors! such! as! staffing! levels! and! the! availability! of! other!

resources,!these!data!highlight!the!unique!division-level!circumstances!that!may!shape!patrol!

decisions,! and! which! in! turn! may! contribute! to! division-level! differences! in! the! racial/ethnic!

distribution! of! stops.! Finally,! as! we! discussed! in! Chapter! 2,! crime! and! poverty! tend! to!

concentrate! in! neighborhoods! with! comparatively! high! levels! of! minority! residents.! In! San!

Diego,!most!of!those!neighborhoods!are!found!in!th e!police!divisions!located!below!Interstate!

8.! !

39!

Figure!4.4.!

Examining!the!relationship!between!vehicle!stop!rates!and!crime,!by!SDPD!police!division!

Source:!City!of!San!Diego!and!SDPD!

Note:!Both!vehicle!stop!rate!and!crime!rate!listed!per!1,000!division!residents!over!2014!and!2015.!!

Table! 4.5! lists! the! volume! of! recorded! stops! by! patrol! division,! as! well! as! each! division’s!

population! and! sq uare! mileage.! The! Northern! division! was! the! city’s! busiest,! accounting! for!

37,203!stops,!or! 14.7! percent! of!those!recorded!between!January! 1,! 2014! and!December! 31,!

2015.!The!Eastern,!Northeastern,!and!Western!divisions!were!the!next-busiest!in!terms!of!stop!

volume,!followed!by!the!C entral,!Southern,!and!Mid-City!divisions.!Officers!in!the!Northwestern!

division! tallied! the! fewest! stops,! accountin g! for! just! 6.4! percent! of! the! citywide! total.! Stops!

initiated! in! divisions! located! above! Interstate! 8! accounted! for! 58.1! percent! of! all! recorded!

stops,!while!those!recorded!below!I-8!represented!41.9!percent!of!the!total.!

! !

100

120

140

160

Total!crime!rate

Stop!rate

Stop!rate Total!crime!rate

40!

Table!4.5.!

SDPD!vehicle!stops,!by!patrol!division,!2014!and!2015!combined!

Population!

Square!mileage!

Stops!

Above!Interstate!8!

!!!

!!!!!Northern!

225,234!(16.4%)!

41.3!(12.5%)!

37,203!(14.7%)!

!!!!!Northeastern!

234,394!(17.0)!

103.8!(31.5)!

31,692!(12.5)!

!!!!!Eastern!

155,892!(11.3)!

47.1!(14.3)!

31,788!(12.6)!

!!!!!Western!

129,709!(9.4)!

22.7!(6.9)!

30,078!(11.9)!

!!!!!Northwestern!

70,822!(5.1)!

41.6!(12.6)!

16,306!(6.4)!

$$$$$Sub-total$

816,051$(59.3)$

256.5$(77.8)$

147,067$(58.1)$

Below!Interstate!8!

!!!!!Southeastern!

175,757!(12.8)!

19.1!(5.8)!

19,292!(7.6)!

!!!!!Central!

103,524!(7.5)!

9.7!(2.9)!

29,692!(11.7)!

!!!!!Southern!

107,631!(7.8)!

31.5!(9.6)!

29,351!(11.6)!

!!!!!Mid-City!

173,012!(12.6)!

12.8!(3.9)!

27,692!(10.9)!

$$$$$Sub-total!

559,924$(40.7)!

73.1$(22.2)!

106,027$(41.9)!

Total!

1,375,975!(100.0)!

329.6!(100.0)!

253,094!(100.0)!

Source:!City!of!San!Diego.!

Note:!Stop!totals!do!not!include!the!6,475!stop!records!submitted!without!stop!location!information.!!!

Table!4.6!lists!the! results! of! our!comparison!of!stop!rates!among!Black!and! White! drivers,!by!

stop!location,!across!the!combined!dataset!of!2014!and!2015!(for!separate!analysis!of!2014!and!

2015! data,! see! Appendix ! 6).! There! is! some! evidence! to! support! the! notion! th at! drivers! are!

treated! differently! in! certain! neighborhoods.! In! the! Northeastern! division,! strong! statistical!

evidence!indicates!that!disparity!was!present:! Bl ack!drivers!were!60.2!percent!more!likely!to!be!

stopped! in! daylight! than! after! dark,! compared! to! White! drivers.! We! fi nd ! no! meaningful!

difference!in!the!treatment!of!drivers!by!race/ethnicity!in!the!Eastern,!Western,!Northern,!and!

Northwestern! divisions.! Analysis! of! the! aggregated! data! from! these! five! divisions! shows! no!

statistically! significant! difference! in! the! daylight-darkness! stop! patterns! of! Black! and! White!

drivers.!!

! !

41!

Table!4.6.!

Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!for!either!a!

moving!violation!or!an!equipment!violation!in!2014!and!2015!combined,!by!stop!location!!

Odds!ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

Above!Interstate!8!

!!!!!Northern!

1.460!

0.066!

0.300!

0.975,!2.184!

2,319!

!!!!!Northeastern!

1.602!

0.005!

0.271!

1.149,!2.232!

2,062!

!!!!!Eastern!

1.050!

0.752!

0.162!

0.776,!1.421!

1,775!

!!!!!Western!

0.936!

0.670!

0.145!

0.692,!1.267!

2,096!

!!!!!Northwestern!

0.891!

0.687!

0.254!

0.510,!1.599!

925!

$$$$$Sub-total$

1.150$

0.068$

0.088$

0.990,$1.337$

9,452$

Below!Interstate!8!

!!!!!Southeastern!

1.397!

0.077!

0.264!

0.964,!2.024!

1,064!

!!!!!Central!

0.572!

<0.001!

0.080!

0.434,!0.752!

1,891!

!!!!!Southern!

1.070!

0.742!

0.220!

0.716,!1.600!

753!

!!!!!Mid-City!

0.887!

0.269!

0.096!

0.717,!1.097!

1,938!

$$$$$Sub-total$

0.793$

<0.001$

0.051$

0.699,$0.899$

5,646$

We!find!distinct!variation!among!divisions!located!below!Interstate!8!across!2014!and!2015.!In!

the!Central!division,!stops!involving!Blacks!are!nearly!43!percent!less!likely!to!occur!during!the!

day! than! they! are! after! sundown,! compared! to! those! involving! White! drivers.! Analysis! of!

Southern,!Southeastern,!and!Mid-City!stops!shows!no!statistically!significant!disparity.!Perhaps!

on!the!strength!of!the!Central!division!findings,!analysis!of!the!aggregated!data!for!these!four!

divisions! shows! that! compared! to! White! drivers,! Blacks! are! 20.7! percent! less! li kely! to! be!

stopped! during! daylight! hours,! when! driver! race/ethnicity! is! visible,! than! they! are! after!

sundown,!when!race/ethnicity!is!obscured!by!darkness.!

42!

Table!4.7.!!

Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!drivers!will!be!stopped!

citywide!for!either!a!moving!violation!or!an!equipment!violation!!

Odds!

ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!

of!stops!

2014!

!!!!!Hispanic!v.!White!

0.973!

0.561!

0.046!

0.887,!1.067$

11,952!

!!!!!Young!Hispanic!v.!Young!White!

1.052!

0.608!

0.103!

0.868,!1.275!

2,775!

2015!

!!!!!Hispanic!v.!White!

0.935!

0.223!

0.052!

0.839,!1.042!

9,055!

!!!!!Young!Hispanic!v.!Young!White!

0.843!

0.123!

0.093!

0.679,!1.047!

2,392!

Combined!

!!!!!Hispanic!v.!White!

0.949!

0.141!

0.034!

0.885,!1.018!

21,007!

!!!!!Young!Hispanic!v.!Young!White!

0.939!

0.392!

0.069!

0.814,!1.084!

5,167!

Table!4.8.!!

Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!drivers!will!be!stopped!c itywide!for!

either!a!moving!violation!or!an!equipment!violation!during!the!DST!period!

Odds!

ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!

of!stops!

2014!

!!!!!Hispanic!v.!White!

1.044!

0.686!

0.111!

0.847,!1.288!

3,669!

!!!!!Young!Hispanic!v.!Young!White!

1.098!

0.685!

0.254!

0.698,!1.728!

854!

2015!

!!!!!Hispanic!v.!White!

1.295!

0.035!

0.158!

1.019,!1.644!

2,950!

!!!!!Young!Hispanic!v.!Young!White!

0.834!

0.461!

0.206!

0.514,!1.353!

803!

Combined!

!!!!!Hispanic!v.!White!

1.145!

0.090!

0.092!

0.979,!1.340!

6,619!

!!!!!Young!Hispanic!v.!Young!White!

0.950!

0.756!

0.158!

0.685,!1.316!

1,657!

! !

43!

Hispanic!drivers!

Tables!4.7,!4.8,!and!4.9!list!results!of!our!analysis!of!traffic!stops!involving!Hispanic!drivers.!Per!

Table!4.7,!when!aggregated!at!the!city!level,! the!odds!o f!a!stop!involving! a!Hispanic! dri ver!is!not!

affected!by!the!ch ange!from!daylight!to!darkness,!regardless!of!when!the!stop!occurred!o r!the!

comparison!group!used,!as!indicated!by!odds!ratios!that!align!so!closely!to!1.0.!

Table!4.8!displays!the!results!from!several!models!examining!day/night!stop!rates!of!Hispanic!

drivers!stopped!for!either!an!equipment!violation!or!a!moving!violation!during!the!120-day!DST!

period.!Under!these!more!restrictive!analytical!conditions,!the!2014!data!reveal!no!disparity!in!

the!treatment!of!Hispanic!and!White!drivers.!In!2015,!however,!Hispanic!drivers!of!all!ages!were!

29.5!percent!more!likely!to!be!sto pped!during!daylight!hours!than!after!dark,!when!compared!

to!Whites.!This! result!was!statistically!significant!at!the!0.05!level.!When!the!analytical!sample!is!

limited!to!those!drivers!ages!25!and!younger,!we!find!no!indication!of!disparity.!!

Table!4.9.!

Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!drivers!will!be!stopped!for!either!a!

moving!violation!or!an!equipment!violation!in!2014!and!2015!combined,!by!stop!location!

Odds!ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

Above!Interstate!8!

!!!!!Northern!

1.043!

0.751!

0.138!

0.805,!1.350!

2,596!

!!!!!Northeastern!

1.337!

0.020!

0.167!

1.047,!1.707!

2,298!

!!!!!Eastern!

0.956!

0.715!

0.117!

0.753,!1.215!

2,025!

!!!!!Western!

0.953!

0.656!

0.102!

0.773,!1.176!

2,490!

!!!!!Northwestern!

1.145!

0.462!

0.210!

0.799,!1.640!

1,063!

Sub-total$

1.062$

0.268$

0.058$

0.955,$1.181$

10,893$

Below!Interstate!8!

!!!!!Southeastern!

1.084!

0.662!

0.200!

0.755,!1.558!

1,351!

!!!!!Central!

0.544!

<0.001!

0.054!

0.447,!0.663!

2,582!

!!!!!Southern!

0.964!

0.726!

0.101!

0.785,!1.184!

4,547!

!!!!!Mid-City!

0.812!

0.030!

0.079!

0.673,!0.980!

2,476!

Sub-total$

0.716$

<0.001$

0.036$

0.649,$0.790$

10,956$

! !

44!

Table!4.9!shows!th e!results!of!our!division-level!analysis!of!stops!involving!Hispanic!drivers!for!

the! combined! dataset! of! 2014! and! 2015! (for! analysis! of! these! data! broken! out! by! year,! see!

Appendix! 6).! We! find! no! evidence! of! disparity! in! the! Northern,! Eastern,! Western,! or!

Northwestern!divisions,!but!strong!evidence!of!disparity!in!the! Northeastern! division:! c o mpared!

to!White!drivers,!Hispanics!stopped!in!the!Northeastern!division!were!33.7!percent!more!likely!

to!be!stopped!before!sundown!than!after!dark!(p!=!0.020).!

We! find! no! difference! in! the! stop! rates! of! Hispanic! and! White! drivers! stopped! in! the!

Southeastern! or! Southern! divisions.! Central! division! stops! involving! Hispanic! drivers! are! 45!

percent!less!l ikely!to!occur!during!the!day!than!they!are!at!n ight!compared!to!stops!of!Whites.!

Similarly,!Hispanic!drivers!stopped!in!Mid-City!are!18.8!percent!less!likely!to!be!stopped!before!

sundown!than!after!dark.!Analysis!of!the!combined!nearly!11,000!stops!occurring!in!divisions!

below! Interstate! 8! shows! that! Hispanic! drivers! were! 28.4! percent! less! likely! to! experience! a!

daytime!stop!than!one!occurring!in!darkness,!compared!to!White!drivers.!T hese! fi ndi ngs!reach!a!

high!level!of!statistical!significance.!

Asian/Pacific!Islander!drivers!

Tables!4.10!–!4.12!document!the!results!of!our!analysis!of!traffic!stops!involving!Asian/Pacific!

Islander!and!White!drivers.!In!short,!we!find!no!meaningful!difference!in!the!stop!patterns!of!

API!and!White!drivers,!regardless!of!driver!age,!stop!date,!stop!location,!or!modelling!strategy.!

Table!4.10.!!

Modeling!the!effects!of!daylight!on!the!odds!that!Asian/Pacific!Islander!drivers!will!be!

stopped!citywide!for!either!a!moving!violation!or!an!equipment!violation!!

Odds!

ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!

of!stops!

2014!

!!!!!Asian!v.!White!

0.986!

0.801!

0.056!

0.882,!1.102$

8,927!

!!!!!Young!Asian!v.!Young!White!

0.953!

0.695!

0.117!

0.749,!1.212!

1,911!

2015!

!!!!!Asian!v.!White!

0.970!

0.635!

0.062!

0.857,!1.099!

6,845!

!!!!!Young!Asian!v.!Young!White!

0.967!

0.792!

0.123!

0.753,!1.231!

1,721!

Combined!

!!!!!Asian!v.!White!

0.978!

0.596!

0.041!

0.900,!1.062!

15,772!

!!!!!Young!Asian!v.!Young!White!

0.960!

0.646!

0.085!

0.808,!1.141!

3,632!

45!

Table!4.11.!!

Modeling!the!effects!of!daylight!on!the!odds!that!Asian/Pacific!Islander!drivers!will!be!

stopped!citywide!for!either!a!moving!or!an!equipment!violation!during!the!DST!period!

Odds!

ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!

of!stops!

2014!

!!!!!Asian!v.!White!

1.090!

0.520!

0.146!

0.838,!1.417!

2,758!

!!!!!Young!Asian!v.!Young!White!

1.307!

0.340!

0.367!

0.754,!2.266!

614!

2015!

!!!!!Asian!v.!White!

1.244!

0.138!

0.183!

0.932,!1.660!

2,200!

!!!!!Young!Asian!v.!Young!White!

1.413!

0.222!

0.400!

0.812,!2.460!

582!

Combined!

!!!!!Asian!v.!White!

1.161!

0.130!

0.114!

0.957,!1.408!

4,958!

!!!!!Young!Asian!v.!Young!White!

1.322!

0.153!

0.259!

0.901,!1.941!

1,196!

Table!4.12.!

Modeling!the!effects!of!daylight!on!the!odds!that!Asian/Pacific!Islander!drivers!will!be!

stopped!for!either!a!moving!violation!or!an!equipment!violation,!by!stop!location!

Odds!ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

Above!Interstate!8!

!!!!!Northern!

0.927!

0.570!

0.124!

0.713,!1.205!

2,585!

!!!!!Northeastern!

1.117!

0.196!

0.056!

0.944,!1.321!

3,231!

!!!!!Eastern!

1.237!

0.085!

0.153!

0.971,!1.575!

2,016!

!!!!!Western$

0.872!

0.315!

0.119!

0.666,!1.139!

2,196$

!!!!!Northwestern!

0.852!

0.256!

0.120!

0.646,!1.123!

1,310!

Sub-total$

0.945$

0.259$

0.047$

0.858,$1.042$

11,603$

Below!Interstate!8!

!!!!!Southeastern!

1.357!

0.179!

0.308!

0.869,!2.118!

473!

!!!!!Central!

1.022!

0.874$

0.143!

0.777,!1.345!

1,960!

!!!!!Southern!

1.370!

0.132!

0.286!

0.910,!2.063!

767!

!!!!!Mid-City!

1.064!

0.647!

0.144!

0.816,!1.387!

1,499!

Sub-total$

1.010$

0.895$

0.078$

0.868,$1.176$

4,699

! !

46!

Table! 4.13!lists! the! demographic! profile! of! drivers! stopped! i n! 2014! and! 2015,! broken! ou t! by!

year.!We!include!these!data!to!highlight!the!statistical!similarities!between!the!full!dataset!and!

the!i nter-twili ght!and!DST-only!sub-samples.!The!proportions! o f!driver!race/ethnicity!and!driver!

age! categories! are! nearly! identical! across! the! two! sub-samples.! Critically,! the! DST-only! sub-

sample! data! al so! mirror! the! full! data! set! quite! closely.! These! similarities! lend! confidence! in!

projecting!to!the!full!sample!of!stops!the!day-night!disparities!revealed!by!our!review!of!inter-

twilight!stops.!

Table!4.13.!

The!demographic!profile!of!drivers!stopped!in!2014!and!2015!!

Total!Sample!

Inter-twilight!P erio d!

Inter-twilight!–!DST!only*!

Driver!race!

2014!

2015!

2014!

2015!

2014!

2015!

!!!!!Asian/PI!

22,059!(15.6%)!

18,493!(16.2%)!

2,588!(15.4%)!

2,085!(16.3%)!

807!(15.6%)!

674!(16.2%)!

!!!!!Black!!

15,763!(11.1)!

12,162!(10.7)!

2,000!(11.9)!

1,459!(11.4)!

616!(11.9)!

467!(11.3)!

!!!!!Hispanic!

42,888!(30.3)!

33,974!(29.8)!

5,716!(34.1)!

4,348!(34.0)!

1,755!(33.9)!

1,446!(34.8)!

!!!!!White!

61,011!(43.1)!

49,211!(43.2)!

6,480!(38.6)!

4,884!(38.2)!

1,999!(38.6)!

1,563!(37.7)!

Driver!age!

!!!!!25!and!under!

31,544!(23.3%)!

28,949!(25.1%)!

3,917!(24.4%)!

3,455!(27.0%)!

1,223!(24.5%)!

1,163!(28.0%)!

!!!!!Over!25!

103,966!(76.7)!

86,456!(74.9)!

12,137!(75.6)!

9,321!(73.0)!

3,764!(75.5)!

2,987!(72.0)!

*30!days!prior!to!an d!after!the!start!and!end!of!Daylight!Saving!Time:!Feb.!7th!through!Apr.!9th!and!th e!October!

3rd!through!December!2nd.!

Note:!Race/ethnicity!and!age!column!totals!are!unequal!because!of!missing!data.!

Analysis!

Application!of!the!veil!of!darkness!technique!to!SDPD’s!2014!and!2015!data!produced!a!series!

of!mixed!results.!Our!analysis!of!citywide!stops!conducted!in!2014!found!disparities!in!the!stop!

patterns!of!Black!and!White!drivers,!yet!those!disparities!disappeared!under!the!more!rigorous!

parameters!of! the!DST-only!analysis.!Neither!the!2015!data!nor!the!combined!2014/2015!totals!

showed!statistically!significant!differences!in!the!treatment!of!Black!drivers!compared!to!White!

drivers,!regardless!of!driver!age!or!stop!date.!

Our!review!of!stops! in volvin g!Hispanic!drivers!produced!a!similarly!mixed!yet!distinct!pattern!or!

results.!No!disparities!were!evident!in!the!2014,!2015,!or!combined!2014/2015!data.!!Ho wever,!

when! we! limited! the! analysis! to! those! stops! occu rring! within! 30! days! of! the! Daylight! Saving!

Time!changes,!we!found!some!evidence!of!disparity!in!the!2015!stop!data.!Comparison!of!stop!

patterns! involving! API! and! White! drivers! revealed! no! statistically! significant! differences!

between!the!two!groups.!!!

47!

In!addition!to!our!citywide!analysis,!we!also!examined!division-level!stop!patterns.!Our!review!

of! aggregate! data! from! the! five! divisions! located! above! Interstate! 8! revealed! no! statistically!

significant! disparities! in! the! daylight-darkness! stop! patterns! of! Black! and! White! drivers! or!

Hispanics!and!Whites.!Narrowing!the!focus!to!the!division!level!revealed!strong!and!consistent!

disparities! in! the! day-night! stop! rates! among! Black! and! Hispanic! drivers! stopped! in! the!

Northeastern!d ivisi on ,! as!compared!to!Whites.!No! such!disparities!were!evident!amon g! stops!

occurring!in!the!Northern,!Eastern,!Western,!or!Northwestern!divisions.!!

Data!on!stops!conducted!below!Interstate!8!reveal!a!different!set!of!results.!We!find!substantial!

evidence!to!suggest!that!in!the!aggregate,! both!Black!and!Hispanic!drivers!were!less$likely!be!

stopped! during! daylight!hours! than! they!were! after! dark,!compared! to! stops!involving! Whi te!

drivers.!In!other!words,!when!the!police!were!able!to!see!a!driver’s!race,!they!were!more!li kely!

to! stop! a! White! driver! than ! they! were! a! Black! or! Hispan ic! driver.! At! the! division! level,! these!

results! were! evident! in! stops! occurring! in! the! Central! divisi on ! and! among! Hispanic! (but! not!

Black)!drivers!stopped!in!the!Mid-City!division.!!

! !

48!

CHAPTER!5:!ANALYZING!POST-STOP!OUTCOMES!

Introduction!

In! the! previous! section! we! examined! 2014! and! 2015! Vehicle! Stop! Card! data! in! an! effort! to!

discern! if! any! disparity! exists! in! the! way! that! SDPD! officers! initiate! vehicle! stops! by!

race/ethnicity.! In! Chapter! 5,! we! examine! post-stop! outcomes! by! driver! race/ethnicity.! These!

outcomes! include! an! officer’s! decision! to! search! a! driver! following! a! traffic! stop,! whether!

contraband!is!discovered,!and!whether!an!o ffi cer!decides!to!issue!a!ticket!or!give!the!dri ver!a!

warning,!among!others.!!

Unlike! with! vehicle! stops,! where! the! comparison! population! (the! demographic! profile! of! the!

city’s! driving! population)! is! unknown,! the! pattern! of! post-stop! outcomes! can! be! measured!

against!an!established!benchmark:!all!drivers!that!were!stopped.!Thus,!in!examining!post-stop!

outcomes,! we! are! ab le! to! get! a! clear! picture! of! the! extent! to! which! disparities! exist! across!

driver! characteristics,! including! race,! gender,! and! residency! status,! as! well! as! stop!

characteristics!like!location!and!time!of!day.!

Table!5.1.!!

Traffic!stops!and!post-stop!outcomes!in!2014!and!2015,!by!SDPD!patrol!division!!

Stops!(%)!

Search!(%)!

Hit!rate!(%)!

Arrest!(% )!

FI!(%)!

Citation!(%)!

Above!Interstate!8!

!!!!!Northern!

14.7!

3.3!

12.1!

1.5!

1.4!

67.1!

!!!!!Northeastern!

12.5!

2.6!

7.6!

0.9!

1.9!

56.1!

!!!!!Eastern!

12.5!

2.6!

6.6!

0.9!

1.2!

67.7!

!!!!!Western!

11.9!

4.2!

12.4!

1.4!

2.7!

60.8!

!!!!!Northwestern!

6.4!

2.6!

7.1!

0.8!

1.6!

45.1!

!!!!!Sub-total!

58.1!

3.1!

9.9!

1.1!

1.8!

57.8!

Below!Interstate!8!

!!!!!Southeastern!

7.6!

10.1!

9.1!

1.7!

8.8!

46.9!

!!!!!Central!

11.7!

5.1!

6.8!

1.7!

2.5!

60.0!

!!!!!Southern!

11.6!

3.1!

8.0!

1.1!

1.8!

69.4!

!!!!!Mid-City!

10.9!

8.6!

7.9!

2.0!

5.3!

51.4!

!!!!!Sub-total!

41.9!

6.7!

8.0!

1.6!

4.2!

53.3!

Total!

100.0!

4.6!

8.7!

1.3!

2.7!

57.5!

*!Hit!rate!is!the!percentage!of!searches!that!led!to!the!discovery!of!contraband! !

49!

Table!5.1!lists!by!police!division!both!vehicle!stop!totals!and!the!incidence! rates!of!key!post-stop!

outcomes.!In!the!Northern!division,!police!conducted!a!search!in!3.3!percent!of!37,203!vehicle!

stops,! or!1!in!30.!Contrast!that!with!the!Southeastern!division,!where!1!in! 10!stops!resulted!in!a!

formal! search! –! three! times! the! rate! in! the! Northern! division.! The! same! kind! of! variance! is!

present! in! other! raw! post-stop! data.! Drivers! stopped! in! the! Western! division! are! more! than!

twice!as!likely!to!face!a!field!interview!(FI)!than!are!drivers!stopped!in!the!Eastern!division. !A!

similar!pattern!is!visible!in!citation!rates:!45.1!percent!of!stops!conducted!in!the!Northwestern!

division! resulted! in! the! issuance! of! a! ticket,! compared! to! almost! 70! percent! of! stops! in! the!

Southern!division.!!

These! observed! patterns! do!not! appear! to! be! random.! To! some!extent,! they!follow! division-

based! differences! in! terms! of! crime! rates! and! Department! allocation! of! officer! resources.!

Drivers!stopped!in!the!city’s!higher-crime!neighborhoods!tend!to!face!a!greater!police!presence.!

That! the! SDPD! may! police! some! areas! differently! than ! other! locations! is! common! practice!

among!other!major!city!police!departments!and !is!well-supported!in!the!research!literature.

These! data! are! also! consistent! with! the! well-established! notion! that! p oli ce! officers! stop! and!

search!drivers!with!two!strategic!goals!in!mind:!(1)!to!promote!public!safety!through!traffic!law!

enforcement!and!deterrence;!and! ( 2)!to!investigate!the!possibility!that!the!driver!(or!passenger)!

has!engaged!in!other!criminal!activity.

Post-stop!enforcement!patterns!vary!just!as!widely!across!other!metrics!as!well.!As!is!shown!in!

Table! 5.2,! drivers! stopped! in! the! middle! of! the! night! are! more! likely! to! be! searched! and!

ultimately!arrested! than! are!drivers! stopped! in!the!morning! or! afternoon.!Similar! variation! is!

found! across! day! of! the! week,! month,! driver! gender,! and! race,! which! is! shown! in! Table! 5.3.!

These! raw! numbers! suggest! that! on! balance! Black! drivers,! compared! to! drivers! of! other!

races/ethnicities,!were!more!frequently!searched!and!arrested!following!a!stop,!less!frequently!

found!with!contraband,!and!the!least!frequently!ticketed.!!

! !

!Braga,!A.,!Papachristos,!A.,!&!Hureau,!D.!(2012).!Hot!spots!policing!effects!on!crime.!Campbell$Systematic$

Reviews,!8,!1-96;!Weisburd,!D.,!&!Telep,!C.!(2014).!Hot!spots!policing:!What!we!know!and!what!we!need!to!know.!

The$Journal$of$Contemporary$Criminolog y,!30,!2 0 0 -220;!CrimeSolutions.gov!(2015).!Hot$Spots$Policing.!Retrieved!

Aug.!16,!2016!from!https://www.crimesolutions.gov/PracticeDetails.aspx?ID=8.!

!Ashton,!R.J.!(2007,!Jul.).!Bridging!the!legal!gap!between!the!traffic!stop!and!criminal!investigation.!The$Police$

Chief,!74(7).!Retrieved!Aug.!16,!2016 ,!from !

http://www.policechiefmagazine.org/magazine/index.cfm?fuseaction=display_arch&article_id=1229&issue_id=72

007;!Whren$v.$United$States.!(1996).!517!U.S.!806.!

50!

Table!5.2.!!

Traffic!stops!and!post-stop!outcomes,!by!stop!time!!

!Time!of!day!

Stops!

Search!

(%)!

Hit!rate!

(%)!

Arrest!

(%)!

FI!(%)!

Citation!

(%)!

Midnight!-!3:00!AM!

25,201!

7.4!

9.9!

3.2!

3.6!

46.8!

3:00!-!6:00!AM!

7,584!

6.6!

10.6!

2.3!

3.0!

46.0!

6:00!-!9:00!AM!

32,541!

3.1!

6.3!

0.8!

1.7!

63.1!

9:00!-!Noon!

52,309!

2.9!

6.8!

0.7!

1.5!

64.6!

Noon!-!3:00!PM!

33,145!

2.4!

6.3!

0.7!

1.2!

66.8!

3:00!-!6:00!PM!

43,145!

5.0!

7.7!

1.1!

4.2!

54.1!

6:00!-!9:00!PM!

27,703!

5.7!

11.0!

1.5!

3.6!

46.8!

9:00!-!Midnight!

36,613!

5.6!

10.2!

1.8!

3.8!

45.6!

These!disparities!may!be!due!to!the!fact!that!more!Black!drivers!live!in!high!crime!areas!of!the!

city!or!are! more!likely!to!drive!late!at!night! rather!than!during!the!day,!thus!the!natural!result!of!

higher!levels!of!exp osure!to!police;!they!may!also!be!the!p roduct!of!disparate!treatment.!The!

challenge!with!this!kind!of!inquiry!is!to!distinguish !variation!that!may!be!the!result!of!policy,!like!

sending! police! officers! to! higher! crime! areas! or! more! proactively! searching! those! drivers!

stopped!at!after!midnight,!from!that!which!is!motivated!by!some!form!of!bias.!!!

Table!5.3.!!

Traffic!stops!and!post-stop!outcomes,!by!driver!race/ethnicity!

Driver!race!

Stops!

Search!(%)!

Hit!rate!(%)!

Arrest!(% )!

FI!(%)!

Citation!(%)!

Asian/PI$

41,021$

4.5!

5.2$

0.8$

2.0$

57.2$

Black!$

28,535$

9.3$

7.7$

1.8$

8.0$

46.1!

Hispanic$

77,934$

5.9$

7.4$

1.5$

3.0$

56.7$

White$

111,855$

2.9$

11.2$

1.2$

1.5$

57.8$

Total$

259,345$

4.4$

8.5$

1.3$

2.7$

56.1$

51!

Research!Method!!

To! this! end,! we! rely! on! an! analytical! technique!

known! as! propensity! score! matching,! which!

allows! the! researcher! to! match! drivers! across!

several!categories!thought!to!affect!the!likelih ood !

of! certain! post-stop! outcomes.! The! matching!

criteria! include! stop-related! factors! like! location!

and! time! of! day,! and! driver! characteristics,! like!

gender! and! residency! status.! Th is! approach! has!

been! used! to! study! traffic! stop ! data! in! Oakland,!

California,

! Cincinnati,! Ohio,

! and! St.! Louis,!

Missouri,

! among! others.! Though! it! is! not! the!

only!techni qu e!that!can!be!used!to!evaluate!post-

stop!outcomes,

!propensity!score! matchin g!is!the!

most! effective! and! intuitive! means! of! isolating! the! effects! of! driver! race.! In! the! section! that!

follows!we!describe!our!application!of!this!technique.!

A! young! male! stopped! on! Friday! night! at! 2:30! AM! for! speeding! through! a! high-crime!

neighborhood! may! be! more! likely! to! receive! a! ticket! than! an! elderly! woman! stopped! on!

Tuesday!at!1:00!PM!for!a!broken!tail!light!while!driving!in!an!area!of!town!not!associated!with!

crime.!If!the!first!driver!is!ticketed!and!the!second!is!not,!can!we!fairly!attribute!that!decision!to!

the!gender!of!the!driver?!Or!is!it!because!one!was!stopped!at!night!and!the!other!during!the!

day?! Or! because! one! was! stopped! for! a! moving! violatio n! and! the! other! for! an! equipment-

related!problem?!In!reality,!an!officer’s!decision!to!search!is!likely!the!product!of!these!several!

factors!taken!together.!Thus,!we!wan t!to!compare!the!post-stop!outcomes!of,!for!example,!all!

!Ridgeway,!G.!(2006).!Assessing!the!effect!of!race!bias!in!post-traffic!stop!outcomes!u sing!p rop en sity!sco res.!

Journal$of$Qu an tita tive$C rim ino log y,!22,!1 -28.!!

!Riley,!K.J.,!Turner,!S.,!MacDonald,!J.,!Ridgeway,!G.,!Schell,!T.,!Wilson,!J.,!Dixon,!T.L.,!Fain,!T.,!&!Barnes -Proby,!D.!

(2005).!Police-community$relations$in$Cincinnati.!Santa!M o nic a,!C A :!RA N D !C orp o ra tio n.! !

!Rosenfeld,!R.,!Rojek,!J.,!&!Decker,!S.!(2011).!Age!matters:!Race!differences!in!police!searches!of!young!and!older!

male!drivers.!Journ a l$of$Re sea rch $in$C rim e$a n d$D elin qu en cy,!49,!3 1-55.!

!Though!we!believe!that!the!propensity!score!m atching!technique!is!the!most!effective!means!of!isolating!the!

effect!of!race!on!post-stop!outcomes,!the!use!of!this!approach!does!have!the!effect!of!reducing!the!sample!size!

available!for!analysis.!To!account!for!the!possibility!that!this!lim its!the!generalizability!o f!our!findings,!we!also!

analyzed!the!2014!and!2015!data!using!logistic!regression!modeling,!another!statistical!technique!widely!accepted!

for!use!with!data !of!th is!kin d!(S ee ,!for!ex am p le,!B au m ga rtn er,!F.,!Ep p ,!D.,!& !Lov e,!B.!(2 01 4).!Police$Searches$of$

Black$and$White$M otorists.$(Durham,$NC).$Chapel!Hill,!NC:!University!of!N o rth !C a ro lin a-Chapel!Hill!Department!of!

Political!Scien ce.!Engel,!R.,!Cherkauskas,!J.,!Smith,!M.,!Lytle,!D.,!&!Mo ore,!K.!(2009).!Traffic$Stop$Data$Analysis$

Study:$Year$3$Final$Report,$Prepared$for$the$Arizona$Department$of$Public$Safety.!Cincinnati,!O H:!University!of!

Cincinnati!Policing!Institute.!Our!findings,!which!are!detailed!in!Appendix!7,!are!consistent.!

Propensity! score! matching! allows!

researchers! to! match! drivers! of!

different!races!across!the!various!other!

factors!known!to!affect!the!decision!to!

ticket,! search,! arrest,! or! discover!

contraband.

! Put! another! way,!

matching! allows! the! analyst! to!

compare! the! likelihood! that! two!

drivers! who! share! gender,! age,! stop!

reason,! stop! location,! and! so! on,! but!

differ! by! race,! will! be! searched,!

ticketed,!or!fo und!with!contraband.!!

52!

young!men!stopped!late!on!Friday!nights!for!speeding!in!a!high-crime!n ei gh b orh ood,!to!see!i f!

race/ethnicity!is!a!determinative!factor!in!these!outcomes.!

Figure!5.1.!!

The!average!percentage!difference!between!matched!and!unmatched!Black!and!White!

drivers!across!eight!variables!used!to!comple te !matching!process!

Note:!Matched!pairs!consist!of!19,948!Black!and!19,948!White!drivers.!No!matches!were!possible!for!8,579!Black!

and!91,859!White!drivers.!!

Figures! 5.1! and! 5.2! document! the! average! differences! between! matched! and! unmatched!

drivers!across!the!eight!variables!upon!whi ch!the!match!was!based.!These!variables!include!the!

reason!for!and!location!(police!division)!of!the!stop,!the!day!of!the!week,!month,!and!time!of!

day!during!which!the!stop!occurred,!and!the!driver’s!age,!gender,!and!residency!status.!!

Per! Figure! 5.1,! the! stop! location! of! matched! Black! and! White! drivers! di ffers! by! only! 0.44!

percent,! while! the! stop! location! of! unmatched! drivers! differs! by! an! average! of! 8.55!percent.!

Similarly,!matched!drivers!were!of!identical!age!categories!in!99.6!percent!of!cases,!compared!

to! 94.63! percent! of! cases!involving! unmatched!Black! and!White! drivers.!Overall,! th e! average!

disparity!between!matched!Black!and!White!drivers!is!0.67!percent,!compared! to!a!7.38!percent!

difference!between!unmatched!drivers.!Figure!5.2!shows!similar!outcomes!from!the!matching!

process!involving!Hispanic!and!White!drivers.!!

10%

12%

14%

16%

18%

Stop!type Division! Time Stop!day Month Age Gender Residency

Matched! Unmatched

53!

Figure!5.2.!

The!average!percentage!difference!between!matched!and!unmatched!Hispanic!and!White!

drivers!across!eight!variables!used!to!comple te !matching!process!

Note:!Matched!pairs!consist!of!39,252!Hispanic!and!39,252!W hite!drivers.!No!matches!were!possible!for!38,682!

Hispanic!and!72,603!White!drivers.!!

These! figures! illustrate! a! critical! attribute! of! the! propensity! score! matching! approach:! any!

differences!we!find!between!Black!and!His pani c!drivers!and!their!matched!White!counterparts!

in!terms!of !searches!conducted,!citations!issued,!or!contraband!found,!are!not!the!result! o f!any!

of!the!factors!listed.!In!other! words,! based! on! the! information! available,!race/ethnicity!is!the!

only!difference!between!the!two!groups!of!drivers,!and!thus!the!only!factor!that!may!explain!

the!observed!differences!in!post-stop!outcomes.

!See!Ridgeway,!G.,!(2009).!Cincinnati$Police$Department$traffic$stops:$Applying$RAND’s$framework$to$analyze$

racial$disparities.!Santa !M o n ica ,!CA :!RA N D!Corporatio n.!T h ere !are !ot he r!fa cto rs!th o u gh t!to !aff ect !the !like lih oo d !of!

certain!post-stop!outcomes,!including,!for!examples:!officer!demographics!(Rojek,!J.,!Rosenfeld,!R.,!&!Decker,!S.!

(2012).!Policing!rac e:!Th e!rac ial!stratification !of!se arch es!in!p olice!traffic!sto ps .!Criminology,!50,!993-1024;!Tillyer,!

R.!Klahm,!C.F.,!&!Engel,!R.S.!(2012).!The!discretion!to!search:!A!multilevel!examination!of!driver!demographics!and!

officer!characteristics.!Journal$of$C on temporary$Crim ina l$Jus tice,!28,!1 8 4 - 205.)!and!performance!history!(Alpert,!

G.P.,!Dunham,!R.G.,!&!Smith,!M.R.!(2004).!Toward!a!better!benchmark:!Assessing!the!utility!of!not-at-fau lt!traffic!

crash!data!in!racial!profiling!research.!Justice$Research$and$Policy,!6,!43-69),!age!(Giles,!H.,!Linz,!D.,!Bonilla,!D.,!&!

Gomez,!M.L.!(2012).!Police!stops!of!and!interactions!with!Hispanic!and!W hite!(non-Hispanic)!drivers:!Extensive!

policing!and!comm unication!accommodation.!Communication$Monographs,$79(4),!40 7-427),!make,!model,!and!

condition!of!the!vehicle!stopped!(Engel,!R.S.,!Frank,!J.,!Klahm ,!C.F.,!&!Tillyer,!R.!(2006,!Jul.).!Cleveland$Division$of$

Police$Traffic$Stop$Data$Study:$Fin al$Report.!Cincinnati,!O H :!U n ive rsity !o f!Cin cin n at i!Div isio n !of !Criminal!Justice),!

10%

12%

14%

16%

18%

20%

Stop!type Division Time Weekday Month Age Gender Residency

Matched Unmatched

54!

Results!

What! follows! are! the! results! of! our! comparative! analysis! of! post-stop! outcomes! for! Black,!

Hispanic,!and!API! d rivers!and!their!matched!White!counterparts,!beginning!with!the!decision!to!

search.!!

The!decision!to!search!

Police!searches!can!be! classified! based! on!the!legal!rules!that!define!them.!The!SDPD! vehicle!

stop!card!lis ts!four!such!search!types:!consent!search,!Fourth!waiver!search,!search!incident!to !

arrest,! and! inventory! search.! We! frame! each! search! type! in! terms! of! the! level! of! officer!

discretion!that!may!determine!the!decision!to!initiate!the!search.!!

We! classify! searches! occurring! incident! to! an! arrest! and! inventory! searches! as! involving! low!

levels!o f!discretionary!authority.!Officers!are!within!their!legal!rights!to!conduct!a!search!when!

an! arrest! is! made,

! and! when! a! vehicle! is! impounded.

! Because! most! such! searches! occur!

automatically,! race-based! disparities! that! exist! say! less! about! officer! behavior! than ! they! do!

about!the!factors!that!led!to!the!arrest!or!impound.!

Consent!searches!are!classified!as!involving!higher!levels!of!officer!discretion.!A!consent!search!

occurs! after! an! officer! has! requested! and! received! consent! from! the! driver! to! search! the!

driver’s! person! or! vehicle.! When! granting! consent,! the! driver! waives! his! or! her! Fourth!

Amendment!protection!against!unreasonable!search!and!seizure.

!A!consent!search!involves!a!

high! degree! of!police! discretion,!as! there! are!few! if! any!legal! strictures!in! place! to!guide! the!

request! for! or! the! nature! of! a! search! following! the! grant! of! consent.! We! would! expect! that!

whatever! disparity! exists! would! manifest! more! clearly! in! the! execution! of! discretionary!

searches.!!

In!the!case!of!a!F ourth!waiver!search,!police!officers!are!permitted!to !search!a!person!and/or!

vehicle!if!and!when!they!d etermine! that! the! driver! or!passenger!is!either!on!probation!or!on!

parole.!By!virtue!of! this! legal! statu s,! the!driver! impl ici tly! agrees!to! waive! Fourth!Amendment!

protection.!As!a!result,!these!searches!often!occur!in!the!absence!of!probable!cause.

and!the!demeanor!of!the!driver!(Engel,!R.S.,!Klahm,!C.F.,!&!Tillyer,!R.!(2010).!Citizens’!demeanor,!race,!and!traffic!

stops.!In!S.K.!Rice!&!M.D.!White!(Eds.),!Race,$ethnicity,$and$policing:$N ew$and$essential$readings.!Ne w !Y or k:!N ew!

York!University!Press),!among!others.!Because!the!SDPD!does!not!collect!these!data,!it!is!impossible!to!include!

them!in!our!ma tch ing!p roto co l.!!

!U.S.$v.$Robinson.!(1973).!41 4!U .S .!21 8 ;!Arizona$v.$Gant.!(2009).!556!U.S.!33 2 .!

!South$Dakota$v.$Opperman.!(1976).!42 8 !U .S.!3 64 .!

!Schneckloth$v.$Bustamonte.!(1973).!412!U.S.!218 .!

!People$v.$Schmitz.!(2012).!55!Cal.4th!90 9 .!

55!

Fourth! wai ver!searches!involve!an!ambiguous!level!of!officer!discretion.

!On!one!hand,!officers!

who!are!legally!permitted!to!conduct!a!Fourth!waiver!search!have!the!discretionary!authority!to!

opt! against! doing! so.! Similarly,! officer! discretion! is! used! in! determining! whether! a! driver! or!

passenger!is!on!probation!or!parole.!In!each!case,!this!d iscretion ary!authority!may!be!applied!

differently! based! on! driver! race.

! On! the! other! hand,! o nce! it! is! determined! that! a!

driver/passenger!is!either!on!probation!or!parole,!the!officer!has!full!legal!authority!to!conduct!

a!search,!which!reduces!the!import!of!th e!decision!to!initiate!the!search.!Relatedly,!we!have!no!

knowledge! of! the! demographic! profile! of! the! City’s! probation/parole! population! or! of! the!

population! of! stopped! drivers! on! probation/parole.! To gether,! these! factors! complicate! our!

ability!to!assign!meaning!to!results!generated!by!an!analysis!of!Fourth!waiver!searches.!

Table!5.4.!!

Comparing!search!rates!among!matched!Black!and!White!drivers!!

Matched!Black!

drivers!(%)!

Matched!White!

drivers!(%)!

Difference!

(%)

p-value!

All!searches$

8.65!

5.04!

52.70$

<0.001!

Consent$

1.39!

0.75!

60.09!

<0.001!

Fourth!waiver$

2.90!

1.30!

76.37$

<0.001!

Inventory $

1.91!

1.30!

42.29$

<0.001!

Incident!to !a rre s t$

0.90!

0.89!

0.56$

!0.480$

Other!(uncategorized)$

1.56!

0.86!

58.09$

<0.001!

Note:!The!analysis!is!based!on!a!total!of!19,948!Black!drivers!and!19,948!matched!White!drivers.!

An! additional! search! type,! the! probable! cause! search,! may! occur! after! an! officer! has!

determined!that!there!is!sufficient!probable!cause!to!believe!that!a!crime!has!been!or!is!about!

to!be!committed.

!The!law!grants!officers!a!substantial!degree!of!leeway!in!determining!when!

the! probable! cause! threshold! has! been! met,! which! makes! the! evaluation! of! probable! cause!

search! incidence! potentially! very! important.! The! SDPD! Vehicle! Stop! card! does! not! include! a!

‘probable! cause! search’! category.! Given! the! legal! and! practical! importance! of! the!

demonstration! of! probable! cause! prior! to! a! search,! this! category! of! searches! should! be!

!Hetey,!R.,!Monin,!B.,!M aitreyi,!A.,!&!Eberhardt,!J.!(2016).!Data$for$change:$A$statistical$analysis$of$police$stops,$

searches,$handcuffings,$and$arrests$in$Oakland,$Calif.,$2013-2014.!Stanford!University,!CA:!Stanford!SPARQ.!

!E.g.,!Burks,!M.!(2014,!Jan.!30).!What!it!means!when!police!ask:!‘Are!you!on!probation!or!parole.’!Voice$of$San$

Diego.!Retrieved!Nov.!21 ,!20 1 6,!fro m!http://www.voiceofsandiego.org/racial-profiling-2/what-it-means-when-

police-ask-are-you-on-probation/.!

!To!calculate!the!percentage!difference!used!in!this!an d!subsequent!tables,!we!divide!the!absolute!value!o f!the!

difference!between!the!first!two!columns!(3.61)!by!the!ave rage !of!th e!first!tw o!co lu m ns !–! in !t h is!c as e,!se a rc h !rat es !

(6.85).!3.61/6.85!=!5 2.7 !perc en t.!

!Illinois$v.$G a te s.!(1 98 3 ).!4 63 !U .S.!2 1 3.!

56!

captured.! As! a! result! of! this! omission,! we! were! unable! to! analyze! this! category! of! police!

action.

As! is! documented! in! Table! 5.4,! we! found! statistically! significant! evidence! of! a! Black-White!

disparity! across! all! search! types! combined,! and! in! four! out! of! five! types! of! searches.! For! all!

search!types!combined,!8.65! percent!of!matched!Black!drivers!were!searched!in!2014!and!2015,!

compared! to! 5.04! of! matched! White! drivers.! 2.90! percent! of! stopped! Black! drivers! were!

subjected!to!a!Fourth!waiver! search,!compared!to!1.30!percent!of!matched!White!drivers.!Black!

drivers!were!also!more!likely!to!face!consent!searches!th an!were!matched!Whites.!To!a!certain!

extent,! these! disparities! were! also! evident! in! low-discretion! searches,! including! inventory!

searches!and!unclassified!search!types.!We!found!no!statistical!difference!between!the!rate!of!

searches! conducted! incident! to! the! arrest! of! a! Black! motorist! when! compared! to! those!

involving!matched!White!drivers.!!

Table!5.5.!!

Comparing!search!rates!among!matched!Hispanic!and!White!drivers!!

Matched!Hispanic!

drivers!(%)!

Matched!White!

drivers!(%)!

Difference!

(%)!

p-value!

All!searches!

6.56!

3.93!

50.22!

<0.001!

Consent!!

0.92!

0.60!

42.69!

<0.001!

Fourth!waiver!!

1.07!

0.90!

17.62!

0.004!

Inventory !!

2.68!

1.06!

86.49!

<0.001!

Incident!to !a rre s t!

0.91!

0.68!

29.86!

<0.001!

Other!(uncategorized)!

0.99!

0.70!

33.84!

<0.001!

Note:!The!analysis!is!based!on!a!total!of!39,252!Hispanic!drivers!and!39,252!matched!White!drivers!

!The!data!file!we!received!from!the!SDPD!included!several!uncategorized!searches!(i.e.,!a!search!was!recorded,!

but!the!officer!involved!either!did!not!consider!it!a!Fourth!waiver!search,!a!consent!search,!a!search!incident!to!

arrest,!or!an!inventory!search,!or,!simply!neglected!to!categorize!it!as!such).!These!incidents!are!referred!to!as!

‘Other!(u n c at e g o rized )’!search es .!The !curren t!veh icle!sto p!d ata!ca rd!d oe s!inclu de !fields!tha t!allow !th e!office r!to!!

describe!the!nature!of!the!probable!cause!used!to!justify!the!search,!including!“Contraband!visible,”!“Odor!of!

contraband,”!“Canine!alert,”!“Observed!evidence!related!to!criminal!activity,”!or!“Other”!(See!Appendix!2!for!

details).!Yet!in!most!cases,!the!officers!are!not!consistent!in!this!documentation.!In!2014,!for!example,!the!‘Other!

(uncategorized )’!categ ory !inclu de d!9 38 !sea rche s.!O f!thes e,!595 !(63 .4!pe rcent)!were!unlabeled,!while!another!145!

(15.5!percent)!w ere!d esc ribed !as!‘O th er,’!in!m os t!case s!w itho ut!a ny !ad ditio na l!info rm atio n.!B eca u se!w e!ca nn o t!

confidently!characterize!some!78.9!percent!of!these!data!as!meeting!the!probable!cause!standard,!we!neglected!to!

create!such!a!category.!!!!

57!

Table!5.5!displays!the!results!of!our!comparison!of!Hispanic! drivers!and! their! matched! White!

counterparts.!We!find!statistically!significant! evidence! of! a! Hispanic-White! disparity!across! all!

search! types! combined,! as! well! as! in! all! five! types! of! searches.! In! the! aggregate,! officers!

conducted!a!search!in!6.56!percent!of!stops! involving! Hispanic! drivers,! compared! to! the!3.93!

percent!of!stops!involving!matched!White!drivers.!

Though!cons ent!searches!are!relatively!rare!occurrences,!regardless!of!driver!race,!in!2014!and!

2015! Hispanic! drivers! were! subject! to! consent! searches! more! often! than! their! White!

counterparts.!We!find!statistically!significant!differences!between!Hispanic!and!matched!White!

drivers!across! all! search!types,! including! consent!searches,!Fourth! waiver!searches,!inventory!

searches,! those! conducted! incident! to! arrest,! and! other! uncategorized! searches.! Hispanic!

drivers!were!also! significantly! more! likely! to!face!an!inventory!search!than!are!their!matched!

White!counterparts.!!

Table!5.6!lists!the!results !of!our!analysis!of!searches!involving!matched!API!and!White!drivers.!

Under! certain! conditions,! we! find! statistically! significant! evidence! that! White! drivers! were!

searched! at! greater! rates! than! matched! APIs.! In! the! aggregate,! matched! White! drivers! were!

searched! following! 3.48! percent! of! stops,! compared! to! a! 2.61! percent! search! rate! for! API!

drivers.!We!also!find!that!Whites!were!subject!to!higher!rates!of!inventory!searches,!searches!

conducted!incident!to!arrest,!and!uncategorized!searches.!There!was!no!statistically!significant!

difference!in!either!consent!or!Fourth!waiver!search!rates.!!

Table!5.6.!

Comparing!search!rates!among!matched!Asian/Pacific!Islander!and!White!drivers!!

Matched!Asian/PI!

drivers!(%)!

Matched!White!

drivers!(%)!

Difference!

(%)!

p-value!

All!searches$

2.61!

3.48!

-28.57!

<0.001!

Consent$

0.48!

0.49!

-2.06!

0.390!

Fourth!waiver$

0.64!

0.74!

-14.49!

0.063!

Inventory $

0.69$

1.02$

-38.60$

<0.001$

Incident!to !a rre s t$

0.35$

0.68$

-64.08$

<0.001$

Other!(uncategorized)$

0.50!

0.64!

-24.56!

0.006!

Note:!The!analysis!is!based!on!a!total!of!34,068!Asian/PI!drivers!and!34,068!matched!White!drivers!

In!sum,!we!find!that!Black!and!Hispanic!drivers!were!more!likely!to!be!the!subject!of!a!police!

search!following!a!traffic!stop!than!were!matched!Whites.!These!disparities!are!consistent!with!

58!

those!generated!by!recent!analyses!of!police!search!decisions!in!Minneapolis,!Minnesota,

!St.!

Louis,!Missouri,

!and!Portland,!Oregon,

!among!several!other!jurisdictions.

Hit!rates!

The!term!‘hit!rate’!is!used!to!describe!the!frequency!that!a!police!officer’s!search!leads!to!the!

discovery! of! unlawful! contraband,! which! the! SDPD! defines! as! “property! that! is! illegal! to!

possess.”

!This!metric!is!a!reflection!of!the!quality!and!efficiency!of!a! pol ice! of ficer’s!d ecisio n! to!

search!and!is!a!well-accepted!means!of!identifying!racial/ethnic!disparities.

! !

Our!hit! rate! analysis!was! complicated! by!several!challenges! stemming! from!the!way! that! the!

SDPD!captures!data!on!the!discovery!of!contraband.!The!first!involved!how!to!treat!the!tens!of!

thousands!of!ambiguously!labeled!cases!included!as!part!of!the!raw!data!compiled!by!the!SDPD.!

As!is! documented! in! Table!5.6,!a!very!high! number! –!over! 90!percent! –!of!cases!were!either!

missing! info rmation! on! the!discovery! of!contraband! or!coded! ambiguously.! We! acknowledge!

that!these!missi ng! data! are!likely!the!product!of!the!SDPD’s!data! management! system!rather!

than! officer! non-compliance.! Indeed,! our! hit! rate! analysis! reflects! the! assumption! that! these!

missing/ambiguous!data!in di cate!that!no!contraban d!was!discovered.!With !that!said,!we!cannot!

offer! any! evidence! to! substantiate! this! assumption,! and! thus! make! these! calculations! with!

slightly!less!confidence!than!some!of!our!others.!

!Briggs,!S.J.!(2016).!The!impact!of!police!deployment!on!racial!disparities!in!discretionary!searches.! Race$and$

Justice.!Availab le!o nlin e !be fo re!p rin t.!DOI:!10.1177/2153368716646163.!!

!Rojek,!J.,!Rosenfeld,!R.,!&!Decker,!S.!(2012).!Policing!race:!The!racial!stratification!of!searches!in!police!traffic!

stops.!Criminology,!50,!993-1024.!

!Renauer,!B.C.!(2012).!Neighborhood!variation!in!police!stops!and!searches:!A!test!of!consensus!and!conflict!

perspectives.!Justic e$ Quarterly,!15,!219-240.!

!Tillyer,!R.,!&!Klahm,!C.F.!(2015).!Discretionary!searches,!the!impact!of!passengers,!and!the!implications!for!

police-minotity!encounters.!Criminal$Justice$Review.!Available!onlin e !be fo re!p rin t.!DOI:!

10.1177/0734016815581049;!Tillyer,!R.,!Klahm,!C.F.,!&!Engel,!R.S.!(2012).!The!discretion!to!search:!A!multilevel!

examination!of!driver!demographics!and!officer!characteristics.!Journal$of$Contem p ora ry $Crim in al$Ju stice,!28,!1 8 4 -

205;!Fallik,!S.W.,!&!Novak,!K.J.!The!decision!to!search:!Is!race!or!ethnicity!imp orta nt? !Jou rna l$of$C on temporary$

Criminal$Justice,!28,!46-165.$

!The!Department!also!notes!that,!“Determ ining!whether!property!is!contraband!is!contextual—some!prop erty!

that!is!generally!legal!to!poss ess!m ay !be!illegal!in!ce rtain!circum stances.!!For!example,!an!open!container!of!alcohol!

is!generally!le ga l!fo r!a d u lts !2 1!y ea rs !o r!o ld er ,!ho wever!is!illegal!w h e n !p o ss es se d!in !a !v eh ic le.!!S imilarly,!parole es !

may!have!restrictions!regarding!possession!of!specific!weapons!that!would!otherwise!be!legal.!

!Persico,!N.,!&!Todd,!P.E.!(2008).!The!hit!rate!test!for!racial!bias!in!motor-vehicle!searches.!Po lice$Quarterly,!25,!

37-53;!Ridgeway,!G.!&!MacDonald,!J.!(2010).!Methods!for!assessing!racially!biased!policing.!In!S.K.!Rice!&!M.D.!

White!(Eds.)!Race,$ethnicity,$and$policing:$New$and$essential$readings!(pp.!180-204).!New!York:!New!York!

University!Press;!Tillyer,!R.,!Engel,!R.S.,!&!Cherkauskas,!J.C.!(2010).!Best!practices!in!vehicle!stop!data!collection!and!

analysis.!Policing :$An$International$Journal$of$Police$Strategies$&$Management,!33,!69-92.!

59!

Table!5.7.!

Raw!data!on!the!discovery!of!contraband!

!Search!conducted?!!!!

Contraband!found?!!

Yes!

No!

Missing!

Total$

!!!!!Yes!

981!

26!

1,007$

!!!!!No!!

6,775!

9,554!

31!

16,360$

!!!!!Null!

337!

63,488!

722!

64,547$

!!!!!Missing!

3,434!

163,453!

10,777!

177,664$

$$$$$Total$

11,527$

236,521$

11,530$

259,578$

The! second! and! related! challenge! resulted! from! the! fact! that! according! to! the! SDPD,!

contraband! discovery! should! be! considered! valid! for! the! purposes! of! our! analysis! only! if! it!

follows!a!search.!Per!Table!5.7!there!were!26!cases!where!contraband!was!discovered,!but!n o!

search! was! recorded.! Furthermore,! there! are! 3,771! cases! where! a! search! o ccurred,! but! the!

outcome! of! the! search! was! either! missing! or! ambiguously! coded.! Final ly,! there! were! 11,499!

cases! where!search!data!was!missing!or!listed!as!null,!including!31!cases!where!‘no!contraband ’!

was!listed.!

To! address! these! data! issues,! we! excluded! the! 11,499! cases! where! search! data! was!

missing/null,!and!the!26!cases!where!the!discovery!of!contraband!was!reported,!but!no!search!

was! conducted.! From! there,! we! classified! cases! where! information! on! the! discovery! of!

contraband!was!either!missing!or!null!as!indicative!of !a!‘no!contraband’!finding.!We!recognize!

that!there!are!possible!impl icati ons! for!treating!these!missing!cases!differently!and!thus!have!

included! the! results! of! addi tion al! analyses,! including! models! where! we! drop! all! missing/null!

cases,!in!Appendix!8.!

To!generate!the!data!shown!in!Tab le!5.8,!we!interpreted!all!missing!and!null!cases!as!indicating!

that!no!contraband!was!discovered!(n=242,211).!From!there,!we!calculated!hit!rates!using!the!

19,948! matched! Black! and! 19,948! matched! White! drivers! that! we! used! to! analyze! the!

Department’s!search! decisions. ! Police!searched! 1,726! (8.65!percent)!of! Black! drivers!stopped!

and! discovered! contraband! on! 137! occasions,! o r! 7.9! percent! of! the! time.! Of! matched! White!

drivers,!1,005!(5.04!percent)!were!searched,!with!125!of!those!searched!(12.4!percent)!found!to!

be! holding! contraband.! Matched! Whites! were! more! likely! to! be! found! with! contraband!

following!F ourth! waiver!searches!and! consent!searches.!There!were!no!statistically!significant!

differences!in!the! hit! rates!of!matched!Black!and! White! drivers!following!searches!conducted!

incident!to!arrest,!inventory!searches,!or!other,!uncategorized!searches.$

! !

60!

Table!5.8.!!

Comparing!hit!rates!among!matched!Black!and!White!drivers!

Matched!Black!

drivers!(%)!

Matched!White!

drivers!(%)!

Difference!

(%)!

p-value!

All!searches!

7.9!

12.4!

-44.2!

<0.001!

Consent!

7.2!

14.8!

-68.6!

0.013!

Fourth!waiver!

7.4!

14.3!

-63.2!

0.002!

Inventory !

3.4!

4.8!

-34.6!

0.368!

Incident!to !a rre s t!

14.0!

13.5!

3.5!

0.897!

Other!(uncategorized)!

11.6!

17.5!

-41.0!

0.069!

Note:!The!analysis!is!based!on!a!total!of!19,948!Black!drivers!and!19,948!matched!White!drivers.!Missing!and!null!cases!coded!

as!no!contraband.!!

Table!5.9.!!

Comparing!hit!rates!among!matched!Hispanic!and!White!drivers!!!

Matched!Hispanic!

drivers!(%)!

Matched!White!

drivers!(%)!

Difference!(%)!

p-value!

All!searches!

7.4!

11.9!

-46.2!

<0.001!

Consent!!

9.1!

17.5!

-62.9!

0.002!

Fourth!waiver!!

11.0!

13.1!

-17.6!

0.368!

Inventory !!

2.8!

4.3!

-44.2!

0.126!

Incident!to !a rre s t!

8.9!

13.2!

-38.6!

0.089!

Other!(uncategorized)!

13.2!

15.6!

-17.1!

0.373!

Note:!The!analysis!is!based!on!a!total!of!39,252!Hispanic!drivers!and!39,252!matched!White!drivers.!Missing!and!null!cases!

coded!as!‘no!contraband.’!!

We!used! an! identical!four-part!process! to! evaluate! hit!rates!of! matched! Hispanic!drivers!and!

their!matched! White! counterparts.! The!results!are!shown! in! Table!5.9. ! Police!searched!2,576!

(6.56! percent)! of! the! 39,252! matched! Hispanic! drivers,! finding! contraband! 191! times! (7.4!

percent).!This!figure!is!46.2!percent!lower!than!the!11.9!percent!hit!rate!(183!of!1,542!searches!

uncovered!contraband)!of!the!matched!White!drivers!who!were!searched.!White!drivers!were!

more! likely! to! be! found! carrying! contraban d! following! consent! searches! than! were! matched!

Hispanics.! We! found!no! meaningful! difference!in! the!hit!rates! following!either! Fourth! waiver!

searches,!inventory!searches,!those!conducted!incident!to!arrest,!or!unclassified!searches.

! !

!The!SDPD!also!captures!data!on!incidence!of!property!seizure!fo llowing!traffic!stops,!though!the!Department!

does!not!document!what!type!of!property!was!seized!or!the!circumstances!under!which!the!seizure!occurred.!

Despite!the!ambiguity!that!accompanies!these!data,!we!analyzed!them!using!the!same!analytical!approach!applied!

61!

Table!5.10.!!

Comparing!hit!rates!among!matched!Asian/Pacific!Islander!and!White!drivers!

Matched!API!

drivers!(%)!

Matched!White!

drivers!(%)!

Difference!

(%)!

p-value!

All!searches!

9.42!

10.39!

-9.78!

0.465!

Consent!

9.68!

16.56!

-52.44!

0.075!

Fourth!waiver!

9.22!

12.90!

-33.33!

0.208!

Inventory !

5.15!

3.17!

47.60!

0.230!

Incident!to !a rre s t!

12.61!

12.23!

3.04!

0.920!

Other!(uncategorized)!

12.29!

12.79!

-3.95!

0.881!

Note:!The!analysis!is!based!on!a!total!of!68,136!Asian/Pacific!Islander!drivers!and!68,136!matched!White!drivers.!Missing!and!

null!cases!coded!as!‘no!contraband.’!!

In!Table!5.10,!we!document!the!h it!rates!of!searches!involving!68,136!matched!API!and!White!

drivers.!There!were!no!statistically!significant!differences!evident.!!

To!review,!we!compared!the!hit!rates!–!the!percentage!of!searches!that!led!to!the!discovery!of!

contraband! –! of! searches! involving! API,! Black,! and! Hispanic! drivers! with! those! of! matched!

White!drivers.!Despite!h aving! higher! search! rates,!Black!and!Hispanic!drivers!were!either!less!

likely!or!just!as!likely!to!be!found!carrying!an!illegal!substance,!a!finding!that!is!consistent!with!

those!generated!by!other!recent!studies.

!Matched!White!and!API!drivers!were!equally!likely!

to!be!found!carrying!contraband.!

Arrest!

We! also! used! propensity! score! matching! to! compare! the! arrest! rates! of! Black! and! Hispanic!

drivers!with!White!drivers!who!were!stopped!under!similar!circumstances.!As!is!shown!in!Tab le!

5.11,!1.79!percent! (20,872!stops!led!to!374!arrests)! of!matched!Black!drivers!were! ultimately!

arrested,!compared!with!1.84!percent!(384!of!20,872)!of!matched!White!drivers.!This!difference!

was!not!statistically!significant.!!

! !

to!the!discovery!of!co ntra ban d .!Prop erty !wa s!seized !from !8.9 !perce nt!o f!Black!d rivers!se arch ed ,!a!rate!28!p ercen t!

fewer!than!the !11 .8!p erce n t!seizu re!rate !of!matched!W hite !drive rs!(differenc e!statistica lly!significa nt!at!th e!0.01 !

level).!Simila rly ,!pr op e rty !w a s!s eiz ed !f ro m !1 1 .1 !p erc e n t!o f!H isp a n ic !d rive rs !sto p p e d !an d !s ea rc h ed !b y !th e !SDPD,!

compared!to!the!seizure!rate!of!12.3!percent!of!matched!Whites!(difference!not!statistically!significant).!!

!Tillyer,!R.,!&!Klahm,!C.!(2011).!Searching!for!contraband:!Assessin g!the!use!of!discretion!by!police!officers.!Police$

Quarterly,!14,!166-185;!Warren,!P.Y.,!&!Tomaskovic-Devey,!D.!(2009).!Racial!profiling!and!searches:!Did!the!politics!

of!racial!profiling!change!police!behavior?.!Criminal$Justice$&$Public$Policy,!8,!343-369;!Williams,!B.N.,!&!Stahl,!M.!

(2008).!An!analysis!of!police!traffic!stops!and!searches!in!Kentucky:!A!mixed!methods!approach!offering!heuristic!

and!practical!implications.!Policy$Scien ces,!Vo l.!41,!2 2 1 -243.!

62!

Table!5.11.!!

Comparing!arrest!rates!for!matched!Black!and!White!drivers!!

Matched!

Black!

drivers!(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-value!

Matched!

pairs!

Arrest!

1.79!

1.84!

-2.8!

-0.69!

20,872!

Note:!Missing!and!null!data!considered!as!indicative!of!‘no!arr es t .’!!

As!we!document!in!Table!5.12,!651!o f!41,220!stops!involving!matched!Hispanic!drivers!resulted!

in!an!arrest,!or!an!arrest!rate!of!1.71!percent.!Stops!involving!matched!White!drivers!ended!i n!

arrest!slightly!less!often!(537!times,!or!a!rate!of!1.41!percent),!tho ugh!the!di fference!between!

the!two!groups!proved!to!be!statistically!significant.!!

Table!5.12.!!

Comparing!arrest!rates!for!matched!Hispanic!and!White!drivers!!

Matched!

Hispanic!

drivers!(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-value!

Matched!

pairs!

Arrest!

1.71!

1.41!

19.2!

<0.001!

41,220!

Note:!Missing!and!null!data!considered!as!indicative!of!‘no!arrest.’!

Table! 5.13! documents! our! analysis! of! arrests! involving! matched! API! and! White! drivers.! API!

drivers! were! arrested! following! 0.85! percent! of! stops! (304! arrests! out! of! 35,847! stops),! 44!

percent! lower!than!the!1.33!percent!arrest! rate!for!matched!Whites!(477!of!35,847!stops!led!to!

an!arrest).!This!disparity!is!statistically!significant!at!the!0.001!level.!

Table!5.13.!!

Comparing!arrest!rates!for!matched!Asian/Pacific!Islander!and!White!drivers!!

Matched!

Asian/PI!

drivers!(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-value!

Matched!

pairs!

Arrest!

0.85!

1.33!

-44.04!

<0.001!

35,847!

Note:!Missing!and!null!data!considered!as!indicative!‘no!arrest.’!

The! findings! involving! Black! and! Hispanic! drivers! are! inconsistent! with! much! of! the! existing!

research!on!the!effects!of!race/ethnicity!on!police!arrest!decisions. !In!fact,!according!to!a!2011!

63!

paper,!24!of!the!27!studies!published!on!the!iss ue!found!that!Blacks!and!other!minorities!were!

more!likely!to!be!arrested!than!Whites!encountering!the!police!under!similar!circumstances.

Field!Interviews!

Per!SDPD!Procedure!6.03,!which!establishes!Department!guidelines!for!the!use!and!processing!

of!Field!Interview!Reports,!a!field!interview!is!defined!as! “ any! co ntact!or!stop!in!which!an!officer!

reasonably! suspects! that! a! person! has! committed,! is! committing,! or! is! about! to! commit! a!

crime.”! According! to! one! SDPD! Sergeant,! FIs! are! “the! bread! and! butter! of! any! gang!

investigator”!and!important!for!identifying!criminal!suspects.

The! traffic! stop! data! card! includes! space! for! officers! to! document! these! encounters.! Our!

analysis! of! the! SDPD’s! field! interview! records! also! showed! statisticall y! significant! differences!

between!matched!pairs.!As!we!show!in!Table!5.14,!matched!Black!drivers!were!subject!to!field!

interview! questioning! 1,203! times! (6.60! percent! of! stops)! between! January! 1,! 2014! and!

December!31,!2015,!while!552!White!drivers!were!given!field!interviews!(2.75!percent)!during!

that!same!period,!a!difference!of!just!over!82!percent.!!!!

Table!5.14.!!

Comparing!field!interview!rates!for!matched!Black!and!White!drivers!!

Matched!

Black!

drivers!(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-value!

Matched!

pairs!

Field!interview!

6.60!

2.75!

82.4!

<0.001!

20,060!

Note:!Missing!and!null!cases!considered!as!indicative!of!‘no!field!interview.’!

Table!5.15!documents!the!results!of!our!analysis!of!matched!Hispanic!and!White!drivers.!SDPD!

officers!conducted!field!interviews!with!2.98!percent!of!matched!Hispanics,!a!rate!37!percent!

greater!than!the!2.05!percent!experienced!by!White!drivers.!!

!Kochel,!T.R.,!Wilson,!D.B.,!&!Mastrofski,!S.D.!(2011).!Effect!of!suspect!race!on!officers’!arrest!decisions.!

Criminology,!49,!473-512.!See!also,!Alpert,!G.!P.,!Becker,!E.,!Gustafson,!M.!A.,!Meister,!A.!P.,!Smith,!M.!R.,!&!

Strombom,!B.!A.!(2006).!Pedestrian$and$motor$vehicle$post-stop$data$analysis$report.!Los!Angele s,!C A :!An aly sis !

Group.!Retrieved!Oct.!3,!2016,!from!

http://assets.lapdonline.org/assets/pdf/ped_motor_veh_data_analysis_report.pdf;!Smith,!M.!R.,!&!Petrocelli,!M.!

(2001).!Racial!profiling?!A!multivariate!analysis!of!police!traffic!stop!data.!Police$Quarterly,$4,!4-27;!Withrow,!B.!L.!

(2004).!Race-based!policing:!A!descriptive!analysis!of!the!Wichita!stop!study.!Police$Practice$and$Research,$5,!223-

240.!!!

!O'Deane,!M .,!&!Murphy,!W.P.!(2010,!Sept.!23).!Identifying!and!documenting!gang!members.!Police$Ma gazine.!

Retrieved!Aug.!16,!2016,!from!http://www.policemag.com/channel/gangs/articles/2010/09/identifying-and-

documenting-gang-members.aspx.!

64!

Table!5.15.!!

Comparing!field!interview!rates!for!matched!Hispanic!and!White!drivers!!

Matched!

Hispanic!

drivers!(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-value!

Matched!

pairs!

Field!Interviews!

2.98!

2.05!

37.0!

<0.001!

39,505!

Note:!Missing!and!null!cases!considered!as!indicative!of!‘no!field!interview.’!!

Table!5.16!documents!the!results!of!our!analysis!of!field!interviews!involving!matched!API!and!

White!drivers.!Though!field!interviews!were!relatively!rare!occurrences!overall,!we!find!that!the!

FI!rate!of!matched!API!drivers!(1.98!percent,!or!710!FIs!following!35,847!stops)!was!higher!than!

that!of!matched!Whites!(1.67!percent,!or!599!FIs!following!35,847!stops).!!

Table!5.16.!!

Comparing!field!interview!rates!for!matched!Asian/Pacific!Islander!and!White!drivers!!

Matched!

Asian/PI!

drivers!(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-value!

Matched!

pairs!

Field!interview!

1.98!

1.67!

16.99!

<0.001!

35,847!

Note:!Missing!and!null!cases!considered!as!indicative!of!‘no!field!interview.’!!

It!is!difficult!to!position!these!findings!in!context!with!data!generated!by!other!departments,!as!

the! vast! majority! of! pu bl ish ed! research! examining! field! interviews! considers! those! FIs! that!

occur!following!pedestrian!stops.!We!note!that!SDPD’s!current!data!management!regime!does!

not! allow! officers! to! distinguish! a! field! interview! conducted! pursuant! to! a! traffic! stop! from!

those!involving!pedestrians.!

Citation!or!warning!

We! close! Chapter! 5! with! a! review! of! data! on! the! issuance! of! citations.! As! with! the! previous!

analyses,!we!use!propensity!score!matching!to!account!for!the!several!factors!that!may!affect!

an!officer’s!decision!to!issue!a!citation!rather!than!a!warning,!including!when,!why,!and!where!

the! stop! occurred.! This! allows! us! to! attribute! any! disparities! we! observe! to ! driver! race.! We!

interpreted!missing!data!and!those!cases!l isted!as!‘null’!(n!=!11,550)!to!indicate!that!the!driver!

received!a!warning!rather!than!a!citation.

!To!account!for!the!possibility!that!our!findings!are!influenced!by!this!interpretation!of!the!missing!and/or!null!

data,!we!examined!the!citation/warning!data!under!several!other!assumption!conditions.!The!full!results,!which!

are!consistent!with!those!described!above,!are!found!in!Appendix!10.!

65!

The! findings,! listed! in! Table! 5.17,! show! that! matched! Black! drivers! receive! a! citation! in! 49.6!

percent! of! stops,! as! compared! to! matched!White! drivers,! who! were! cited! in! 56.1! percent!of!

stops.! To! account! for! th e! possibil ity! that! those! factors! that! led! to! a! search! may! affect! the!

likelihood!that! a! driver!will!receive! a! citati on,! we!also!limited! the! analysis!to!those! motorists!

who! were! stopped! by! the! SDPD! but! not! searched.! After! dropping! searched! drivers! from! the!

sample,!we!re-matched!the!remaining!drivers!using!the!same!set!of!variabl es!and!procedure!as!

described! above.

! T he! results,! also! displayed! in! Table! 5.17,! suggest! that! the! relationship!

between!the! initiation!of!a!search!and! the!decision!to!issue!a! citation!is!unrelated!to!race.! In!

fact,!the!percentage!of!citations!increased!slightly!for!both!matched!Black!and!White!drivers.!!

Table!5.17.!!

Comparing!citation!rates!for!matched!Black!and!White!drivers!!

Matched!

Black!drivers!

(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-value!

Matched!

pairs!

Searched!drivers!includ ed!

49.60!

56.10!

-12.3!

<0.001!

20,922!

Searched!drivers!excluded!

51.97!

58.03!

-11.0!

<0.001!

19,353!

Note:!Missing!and!null!cases!coded!as!indicative!of!‘no!citation!given.’!

As!shown!in!Table!5.18,!SDPD!officers!cite!matched!Hispanic!and!White!drivers!at!very!similar!

rates.!When!searched! drivers! are! included! as!part! of!the! matched!sample,! the!percentage! of!

drivers! given! a! citation! is! nearly! identical! across! races.! When! searched! drivers! were! omitted!

from! the! sample,! the! re-matched! Hispanic! drivers! were! ticketed! 60.67! percent! of! the! time,!

compared!to!59.72!for!Whites.!!

Table!5.18.!!

Comparing!citation!rates!for!matched!Hispanic!and!White!drivers!!

Matched!

Hispanic!

drivers!(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-value!

Matched!

pairs!

Searched!drivers!includ ed!

58.44!

58.36!

0.1!

0.833!

41,340!

Searched!drivers!excluded!

60.67!

59.72!

1.6!

0.007!

39,006!

Note:!Missing!and!null!cases!coded!as!indicative!of!‘no!citation!given.’!

Finally,!as!is!shown!in!Table!5.19,!we!relatively!small!yet!statistically!significant!differences!in!

the!citation!rates!of!matched!API!and!White!drivers.!

!The!categorical!balancin g!requirements!(no!statistical!difference)!were!met!for!each!of!the!independent!

variables!used!to!match!Black/Hispanic!and!White!drivers.!

66!

Published!research!on!the!relationship!between!driver!race/ethnicity!and!the!citation/warning!

decision! has! generated!inconsistent! findings.! In!some! studies,! analysts!have! found! that!Black!

and!Hispanic!drivers!are!less!likely!to!receive!a!traffic!citation!than!White!drivers.

!In!others,!

data!show! that! minority! drivers!receive!citations!at! greater! rates! than! Whites!stopped!under!

similar!conditi ons.

!No!published!research!that! we!are!aware! o f!examines!the!citation!patterns!

of!API!drivers.!

Table!5.19.!!

Comparing!citation!rates!for!matched!Asian/Pacific!Islander!and!White!drivers!!

Matched!

Asian/PI!

drivers!(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-value!

Matched!

pairs!

Searched!drivers!includ ed!

59.13!

57.39!

2.99!

<0.001!

35,847!

Searched!drivers!excluded!

60.11!

58.66!

2.44!

<0.001!

34,884!

Note:!Missing!and!null!cases!coded!as!indicative!of!‘no!citation!given.’!

Summary!

We!used!the!propensity!score!matching!technique!to!pair!API,!Black,!and!Hispanic!drivers!with!

White!drivers!who!were!stopped!by!the!SDPD!under!similar!circumstances.!By!matching!drivers!

along! these! lines! we! were! able! to! isolate! the! effect! that! driver! race/ethnicity! has! on! the!

likelihood!that!each!group!will!experience!one!of!several!post-stop!outcomes.!We!found!that:!!

• 8.65! percent! of! stops! involving! Black! drivers! involved! a! search,! a! rate! 52.7! percent!

greater!than!the!5.04!percent!of!matched!White!drivers!who!were!searched.!Similarly,!

Hispanics!were!searched!in!6.56!percent!of!stops,!50.22!percent!greater!than!matched!

Whites! (3.93! percent).! With! few! exceptions,! these! disparities! were! robust! across! all!

search!types.!

!Engel,!R.!S.,!Frank,!J.,!Tillyer,!R.,!&!Klahm,!C.F.!(2006).!Cleveland$division$of$police$traffic$stop$data$study:$Final$

report.$Cincinnati,!OH:!University!of!Cincinnati.!Submitted!to!the!Cleveland!Division!of!Police,!Cleveland,!OH;!

Schafer,!J.A.,!Carter,!D.L.,!Katz-Bannister,!A.,!&!Wells,!W.M.!(2006).!Decision-!making!in!traffic!stop!encounters:!A!

multivariate!analysis!of!police!behavior.!Police$Quarterly,$9,!184-209.!!

!Engel,!R.!S.,!Tillyer,!R.,!Cherkauskas,!J.!C.,!&!Frank,!J.!(2007).!Traffic$stop$data$analysis$study:$Year$1$Final$Report.$

Cincinnati,!OH:!University!of!Cincinnati.!Submitted!to!the!Arizona!Department!of!Public!Safety,!Phoenix,!AZ;!

Regoeczi,!W.C.,!&!Kent,!S.!(2014).!Race ,!po verty ,!and !the !traffic!ticke t!cyc le:!Exp lorin g!th e!situa tion a l!con tex t!of!the !

application!of!police!discretion.!Policing:$An$International$Journal$of$Police$Strategies$&$Management,!37,!190 –205.!

Tillyer,!R.,!&!Engel,!R.S.!(2013).!The!impact!of!drivers’!race,!gender,!and!age!during!traffic!stops:!Assessing!

interaction !te rms!and!the !so c ial!c o n d itio n in g!model.!Crime$&$Delinquency,!59,!369-395.!

67!

• Despite! occurring! at! greater! rates,! police! searches! of! Black! and! Hispanic! drivers! were!

either!less!likely! than! or! just!as!likely!to! be! found!with!contraband!as! matched! White!

drivers.!The!size!and!statistical!strength!of!the!disparity!vary!by!search!type.!!

• Matched! Black! drivers! were! su bject! to! field ! interviews! in! 6.60! percent! of! stops,! 2.4!

times! the! rate! of! matched! White! drivers! (2.75! percent).! Police! conducted! field!

interviews!in!2.98!percent!of!stops!involving!matched!Hispanic!drivers,!37!percent!lower!

than! the! 2.05! percent! FI! rate! of! their! matched! White! counterparts.! Police! cond ucted!

field! interviews! with! 1.98! percent! of! matched! API! d rivers,! nearly! 17! percent! greater!

than!the!1.67!percent!FI!rate!of!matched!Whites.!!

• There! was! no! statistical! difference! in! the! arrest! rates! of! matched! Black! and! White!

drivers.!Hispanic!drivers!were!arrested!at!a!slightly!higher!rate!than!their!matched!white!

counterparts,!while!Whites!were!arrested!at!a!greater!rate!than!matched!API!drivers.!

• Black! drivers! were! issued! citations! less! often! than! their! matched! White! peers,! while!

matched!API,!Hispanic,!and!White!drivers!were!cited!at!nearly!identical!rates.!

! !

68!

CHAPTER!6:!SUMMARY!AND!RECOMMENDATIONS!

Summary!of!research!method!and!findings!

In! this! Report,! we! analyzed! several! data! sources! –! including! records! of! 259,569! traffic! sto ps!

conducted! between! January! 1,! 2014! and! December! 31,! 2015,! data! gathered! from! 10!

community!focus! groups,! an! electronic!survey!of!the!SDPD! (n=365),! and!foll ow-u p! interviews!

with!officers!from!all!nine!patrol!divisions!(n=52)!–!in!an!effort!to!address!four!broad!questions:!

1. To!what!extent!is!there!a!department-level!pattern!of!racial/ethnic!disparity!in!the!

initiation!of!traffic!stops?!!

2. To!what!extent!are!racial/ethnic!disparities!in!the!in iti atio n!of!traffic!stops!evident!at!

the!patrol!division!level?!!

3. To!what!extent!is!there!a!department-level!pattern!of!racial/ethnic!disparity!in!the!

outcome!of!traffic!stops?!!

4. How! does! SDPD’s! traffic! enforcement! regime! affect! police-community! relations! in!

San!Diego?!!

The! research! methodology! and! findings! detailed! over! the! previous! several! chapters! are!

summarized!below.!In!the!subsequent!recommendations!section,!we!draw!on!our!findings!from!

the! community! focus! groups,! electronic! survey,! and! officer! interviews! to! contextualize! and!

support!our!recommendations!to!the!Department.!

Method!of!analysis:!Traffic!stops!

To!properly!assess!the!effect!that!a!driver’s!race/ethnicity!has!on!the!likelihood!that!he!or!she!

will!be!stopped,!researchers!must!develop!a!benchmark!that!enables!the!comparison!of!actual!

stop! rates! with! a! driver’s! risk! of! being! stopped! in! the! absence! of! bias.

100

! An! appropriate!

benchmark!must!incorporate! the!various!legal!and!non-legal!factors!that! shape!this!stop!risk,!

including:! when,!where,!and!how!often!they!drive;!the!make,!model,!and!condition!of!their!car;!

and!their!behavior!and!demeanor!while!driving.

101

The!ch all enge! that!has!plagued!past!efforts!to!perform!this!kind! of!analysis!is!driven!by!what!

police!accountability!expert!Sam!Walker!calls!the!“denominator”!problem:!researchers!do!not!

100

!Tillyer,!R.,!Engel,!R.S.,!&!Cherkauskas,!J.C.!(2010).!Best!practices!in!vehicle!stop!data!collection !an d!an alys is.!

Policing:$An$International$Journal$of$Police$Strategies$&$Management,!33(1),!69-92.!

101

!Fridell,!L.A.!(2004).!By$the$numbers:$A$guide$for$analyzing$race$data$from$Vehicle$Stops.!Washington,!D.C.:!Po lice !

Executive!Research!Forum;!Ridgeway,!G.!&!MacDonald,!J.!(2010).!Methods!fo r!assessing!racially!b iased!policing.!In!

S.K.!Rice!&!M.D.!White!(Ed s.)!Race,$ethnicity,$and$policing:$New$and$essential$readings!(pp.!180-204).!New!York:!

New!York!University!Press;!Tillyer,!R.,!Engel,!R.S.,!&!Cherkauskas,!J.C.!(2010).!Best!practices!in!vehicle!stop!data!

collection!and!analysis.!Policing:$An$International$Journal$of$Police$Strategies$&$Management,!33(1),!69-92;!and!

Walker,!S.!(2001).!Searching!for!the!den ominator:!Problem s!w ith !po lice!tra ffic!sto p!d ata !an d!a n!e arly!w a rnin g!

system!solution.!Justice$Researc h$a n d$P olicy,!3(1),!63-95.!

69!

have! an! accurate! way! to! measure! the! demographic! profile! of! a! city’s! driving! population.

102

There! are! several! weaknesses! in! using! Census! data! as! a! proxy,! including! well-establish ed!

racial/ethnic!and!age-based!d isp arities!between!those!who!live!in!a!city!and!those!who!drive! on!

its!roads.

103

!Further,!a!city’s!drivin g!population!is!fluid;!those!who!drive!at!8!am!may!look!and!

act!substantially!different!than!those!who!drive!at!8!pm!across!many!relevant!stop-related!risk!

factors.!!

We!circumvent!this!problem!by!employing!what!is!known!as!the!veil! of ! d arkness!techn iq ue.!This!

approach!rests!on!the!assumption!that!if!stop!disparities!exist,!whether!driven!by!race,!age,!or!

other!factors,!they!wil l!be!more!apparent!among!stops!made!in!daylight,!when!drivers’!physical!

profile!and!demeanor!are!more!readily!detectable,!than!at!night,!when!these!characteristics!are!

obscured!by!darkness.!In!an!attempt!to!isolate!the!effect!of!driver!race,!the!analysis!is!confined!

to! the! “inter-twilight! period,”! or! the! period! between! the! earliest! end! of! civil! twilight!

(approximately! 5:09! pm!on! Nov.!27)! and! the!latest! (approximately!8:29! pm! on! Jun.! 27).!This!

allows! us!to! account! for! changes! to!the! driving! population! during! the!course! of! the! day! and!

obviates!the!need!for!an!external!benchmark!against!which!to!compare!actual!stop!patterns.!!

Findings:!Traffic!stops!!

Comparative!analysis!of!discretionary!traffic!stops!involving!Black!and!White!drivers!revealed!an!

inconsistent! pattern! of! results.! Our! review! of! the! 2014! data! (aggregated! at! the! city! level)!

indicated!that!Black!drivers!were!19.6!percent!more!likely!to!be!stopped!during!daylight!hours,!

when! driver! race/ethnicity! was! visible,! than! after! sundown,! when! driver! race/ethnicity! was!

obscured! by! darkness,! compared! to ! White! drivers.! Though! the! 2014! disparities! were! more!

pronounced!when!the! samp le!was!limited!to!drivers!under!the!age!of!25,!they!were!not!present!

in! the! 2015!data! or! in!the! co mbin ed! 2014/2015!data.! Similarly,!our! analysis! of!citywide! data!

revealed! no! indication! that! officers’! decision! to! stop ! Hispanic! drivers! was! affected! by! the!

change! from! daylight! to! darkness,! regardless! of! when! the! stop! occurred! or! the! comparison!

group!used.!

In! addition! to! our! citywide! analysis,! we! also! compared! stop! patterns! by! location.! Analysis! of!

stops!initiated!in!divisions!located!above!Interstate!8!showed!that!in!the!aggregate!police!were!

no! more!likely! to! stop! either! Black! or!Hispanic! drivers! during! daylight! hours! than! after! dark,!

compared! to! White! drivers.! We! found! no! evidence! that! Blacks! or! Hispanics! were! treated!

differently! in! the! Northern,! Eastern,! Western,! or! Northwestern! divisions,! but! statistically!

102

!Walker,!S.!(2001).!Searching!for!the!denominator:!Problems!with!police!traffic!stop!data!and!an!early!warning!

system!solution.!Justic e$Re sea rch $a nd $Po licy,!3(1),!63-95.!

103

!Tillyer,!R.,!Engel,!R.S.,!&!Cherkauskas,!J.C.!(2010).!Best!practices!in!vehicle!stop!data!collection!and!analysis.!

Policing:$An$International$Journal$of$Police$Strategies$&$Management,!33,!69-92.!!!!

70!

significant!evidence!of!disparity!among!stops!initiated!in!the!Northeastern!division.!Compared!

to! White! drivers,! Black! and! Hispanic! drivers! stopped! in! Northeastern! division! neighborhoods!

were!60.2!and!33.7!percent!more!likely!to!be!stopped!in!daylight!than!after!dark,!respectively.!

Conversely,!when!the!analysis!was!confined!to!stops!occurring!in!divisions!below!Interstate!8,!

we!found!that!in!the!aggregate!Blacks!were!nearly!20.7!percent!less!likely!to!be!stopped!during!

daylight!hours,!when!driver!race/ethnicity!is!more!likely!to!be!visible,!than!after!sundown,!when!

race/ethnicity! is! obscured! by! darkness.! Similarly,! our! review! of! the! nearly! 11,000! stops!

occurring! below! Interstate! 8! shows! that! Hispanic! drivers! were! 28.4! percent! less! likely! to!

experience! a! daytime! stop! than! one! occurring! in! darkness,! compared! to! White! drivers.! We!

found!no!stati sti cal!disparity!among!drivers!stopped!in!the!Southeastern!or!Southern!divisions.!

Central!division!stops! involving! Black!drivers!were!42.8!percent! less!likely!to!occur!during! the!

day!than!they!are!at! ni ght!compared!to!stops!of!Whites.!Hispanic!drivers!stopped!in!the!Central!

division! were! 45.6! percent! less! likely! to! experience! a! stop! du ring! daylight! hours! than! in!

darkness.! Similarly,! Hispan ic! drivers! stopped! in! Mid-City! were! 18.8! percent! less! likely! to! be!

stopped!before!sundown!than!after!dark,!compared!to!Whites.!

Finally,! we! found! no! difference! in! the! pattern! of! stops! involving! Asian/Pacific! Islander! and!

White!drivers,!regardless!of!the!analytical!approach!taken!(citywide!and!location-based,!as!well!

as! the! annual!and! DST-only! analyses)!or! the! nature!of! the! comparison!(all! drivers,! drivers!25!

and!under).!$

Method!of!analysis:!Post-stop!outcomes!

In! an! effort! to! eliminate! potentially! confounding! explanations! for! racial/ethnic! disparities! in!

post-stop! outcomes,! we! matched! Black,! Hispanic,! and! API! drivers! with! White! counterparts!

across!a!set!of!demographic!and!stop-based!characteristics!using!a!statistical!technique!known!

as!propensity!score!matching.!Propensity!score!matching!allows!researchers!to!pair!drivers!of!

different! races! across! the! various! other! factors! known! to! affect! the! likelihood! of! receiving! a!

citation,!being!searched,!arrested,!subject!to!a!field!interview,!or!being!found!with!contraband.!

In! other! words,!this! technique! enables!a! much! more!careful! and! nuanced!comparison! of!the!

treatment! of!drivers!who!share!gender,!age,!stop!reason,!stop!lo cation ,!and!so!on,!but!differ!by!

race.!!

Analysis! of! the! post-stop! outcomes! between! matched! pairs! shows! stati stical ly! significant!

differences! in! the! experiences! of! Black! and! Hispanic! drivers! and ! their! matched! White!

counterparts.!!

71!

Findings:!Search!

After! accounting! for! several! possible! explanatory! factors,! we! found! that! Black! drivers! were!

searched! by! the! SDPD! following! 8.65! percent! of! discretionary! traffic! stops,! while! matched!

Whites! were! searched! 5.04! percent! of! the! time.! Analysis! of! specific! search! types! revealed!

similar! levels! of! disparity.! Black! drivers! were! 1.85! times! more! likely! to! submit! to! a! consent!

search! and! 1.47! ti mes! more! likely! to! face! an! inventory! search.! The! differences! were! most!

extreme! in! the! ad mini stration ! of! Fourth!waiver! searches,!where! Black!drivers! were!searched!

more!than!2.23!times!more!often!than!matched!Whites.!!

The! data! also! show! similar! differences! in! the! search! rates! involving! Hispanic! drivers.! In! fact,!

depending!on!the!nature!of!the!search,!Hi spani c!drivers!were!between!17!and!87!percent!more!

likely! to! be! searched! following! a! routine! traffic! stop! than! were! their! matched! White!

counterparts.! Analysis! of! search! rates! in volvin g! matched! API! and! White! drivers! showed! that!

White!drivers!were!1.33!times!more!likely!to!be!searched!than!their!matched!API!peers.!!!

Findings:!Hit!rate!

Despite! being! subject! to! higher! search! rates,! Black! drivers! were! less! likely! to! be! found! with!

contraband!than! were!matched!White!drivers.!Hispanic!drivers!were!also!less!li kely!to!be!found!

holding! contraband,! again! despite! being! subject! to! more! searches.! In! fact,! contraband!

discovery! rates! were! lower! for! searches! involving! Hispanic! drivers,! though! the! statistical!

strength! of! the! d iff erences! with! paired! White! drivers! varied! by! search! type.! No! meaningful!

differences!were!evident!in!the!hit!rates!of!matched!API!and!White!drivers.!!

Findings:!Field!interview,!arrest,!and!citation!

Finally,! we! found! statistically! signifi cant! disparities! in! the! field! interview! rates! of! minority!

drivers,!and!mixed!results!regarding!the!citation!and!arrest!rates!of!Black!and!Hispanic!drivers!

compared! to! matched! Whites.! For! Black! drivers,! 6.60! percent! of! stops! involved! a! fi eld!

interview,! some! 2.4! times! higher! than! the! rate! at! which! matched! White! drivers! were!

interviewed!(2.75!percent).! T he!arrest!rate!of!Black!drivers!was!not!meaningfully!different!from!

that!of!matched!Whites,!despite!the!Department’s!more!proactive!approach!to!searching!and!

interviewing!Black!drivers.!We!found!that!Black!drivers!were!cited!at!lower!rates!(49.6!percent)!

than!White!drivers!(56.1!percent)!who!were!stopped!by!the!SDPD!under!similar!circumstances.!!

Our! analysis! showed! that! Hispanic! drivers! were! subject! to! field! interviews! more! often! than!

matched!White!drivers,!though!the!disparity!was!less!pronounced!than!was!the!case!with!Black!

drivers.!The!observed!disparity!between!Hispanics!and!matched!Whites!did!not!extend!to!either!

arrest!or!the!decision!to!issue!a!citation.!Hispanic!drivers!were!given!citations!at!almost!exactly!

72!

the! same! rate! as! matched! White! drivers! and! though! we! found! statistical! differences! in! the!

arrest!rates!of!the!two!matched!groups,!the!practical!difference!was!rather!small!(1.71!percent!

arrest!rate!for!Hispanics!compared!to!1.41!percent!for!Whites).!!

In!sum,!we!find!statistically!significant!and!meaningful!differences!in!the!post-stop!treatment!of!

Black!and!Hispanic!drivers! compared!to!White!drivers!across!several!important!outcomes.!In!an!

effort!to!put!some!of!these!data!into!context,!we!highlight!the!substantial!race-based!disparities!

in!the!search!rate/hit!rate!data.!!

In!San!Diego,!matched!Black!drivers!were!1.72!times!more!likely!to!be!searched,!and!–!despite!

being!searched! more!frequently!–!were!44.2!percent!less!likely!to!be!foun d! with!contraband.!

Similarly,!SDPD! offi cers!searched!Hispanic!drivers!at!1.67!times!the!rate!of!matched!Whites,!yet!

were! 46.2! percent! less! likely! to! discover! contraband! following! searches! of! Hispanic! drivers!

compared!to!matched!Whites.!!

Compare! these! rates! to! those! of! two! cities! recently! investigated! by! the! U.S.! Department! of!

Justice.!In!Ferguson,!Missouri,!the!DOJ!found!that!Black!drivers!were!2.07!times!more!likely!to!

be!searched,!yet!26!percent!less!likely!to!be!found!with!contraband! th an!were!White!drivers.

104

These! disparities! contributed! to ! the! DOJ’s! conclusion! that! the! Ferguson! Police! Department!

engaged! in! systematic! bias! against! the! city’s! Black! population.

105

! In! Baltimore,! another! city!

recently!found!by!the!DOJ!to!have!engaged!in!a!pattern!or!practice!of!“discriminatory!policing!

against! African! Americans,”

106

! Black! drivers! were! 23! percent! more! likely! than! Whites! to! be!

searched! following! a! traffic! stop,! yet! 74! percent! less! likely! to! be! found! with! contraband.

107

Analysis! of! data! from! Los! Angeles,! California,! a! city! that! spent! nine! years! under! federal!

oversight! to! address! a! pattern! or! practice! of! unlawful! police! behavior,! revealed! a! similar!

pattern.

108

By!contrast,!recent!reports!from!two!other!jurisdictions!found!to!have!engaged!in!a!pattern!or!

practice!of!practice!of!unlawful!conduct,!Cinci nn ati,!Ohio!and!Oakland,!California,!showed!that!

104

!United!States!Department!of!Justice,!Civil!Rights!Division.!(2015,!Mar.!4).!Investigation!of!the!Ferguson!Police!

Department,!p.!65.!Retrieved!Sept.!8,!2016,!from!

https://www.justice.gov/sites/default/files/crt/legacy/2015/03/04/ferguson_findings_3-4-15.pdf.!

105

!United!States!Department!of!Justice,!Civil!Rights!Division.!(2015,!Mar.!4).!Investigation!of!the!Ferguson!Police!

Department.!Retrieved!Sept.!8,!2016,!from!

https://www.justice.gov/sites/default/file s/ crt /le ga cy /2 0 1 5/ 0 3/ 0 4/ fer gu s o n _fin d in g s_ 3 -4-15.pdf.!

106

!United!States!Department!of!Justice,!Civil!Rights!Division.!(2016,!Aug.!10).!Investigation!of!the!Baltimore!City!

Police!Department,!p.!47.!Retrieved!Sept.!8,!2016,!from !https://www.justice.gov/crt/file/883296/download.!

107

!United!States!Department!of!Justice,!Civil!Rights!Division.!(2016,!Aug.!10).!Investigation!of!the!Baltimore!City!

Police!Department.!Retrieved!Sept.!8,!2016,!from! https://www.justice.gov/crt/file/883296/download.!

108

!Ayres,!I.,!&!Borowsky,!J.!(2008),!A!study!of!racially!disparate!outcomes!in!the!Los!Angeles!Police!Department,!

Prepared!for!the!ACLU!of!Southern!California.!!

73!

Black!drivers!were!more!likely!to!be!searched!than!Whites,!but!found!little!difference!in!the!rate!

of!contraband!discovery.

109

To!be!clear,!we!do!not!intend!to!suggest!that!these!similarities!indicate!that!the!SDPD!suffers!

from! the! same! level! of! the! far-reaching,! systemic! dysfun ction ! revealed! by! the! DOJ’s!

investigation!of!police!departments!in!Ferguson!or!Baltimore,!or!those!that!li e!at!the!center!of!

reform!initiatives!pursued!in!the!other!three!jurisdictions.!Rather,!the!comparison!is!made!to!

highlight! the! gravity! of! these! particular! findings! and! the! pattern! of! disparate! treatment! that!

exists!across!several!post-stop!outcomes.!!

Recommendations!

As! other! researchers! have! recently! acknowledged,

110

! a! risk! in! conducting! analyses! of!

racial/ethnic!differences!in!th e!rates!of!contact!with!police!and!the!outcomes!of!those!contacts!

is! to! oversimplify! the! results.! Either! the! police! are! racists! who! purposefully! target! people! of!

color,! or! there! are! no! differences! in! how! people! are! treated! by! the! police,! despite! the!

disparities!regularly!witnessed!and!experienced!by!communities!of!color.!While!shedding!light!

on! an ! important! topic,! these! approaches! –! either! attacking! the! police! or! denying! that!

racial/ethnic! bias! exists!–! inevitably! miss!the! complexity! of!the! issue!and! thus! do!not! offer! a!

productive!way!forward.!!

We! fol l ow! other! recent! research! on! police-community! relations! in! taking! a! problem-solving$

approach!to!the!interpretation!of!o ur!analyses!of!police!traffic!stop!data.!That!is,!in!this!chapter,!

we! offer! potential! ways! of! reducing! racial/ethnic! disparities! in! traffic! stops! and! thereby!

repairing!the!harm!such!disparities!have!inflicted!on!police-community!relations.!In!order!to!do!

so,!we!draw!on!not!only!the!SDPD!traffic!stop!data,!but!also!data!gathered!from!three!other!

sources,! as! described! in! Chapter! 3:! focus! groups! with! residents! of! communities! with! high!

numbers!of!traffic! stops;! an! SDPD-wi de! electronic!survey;! and!in-depth! interviews!with! SDPD!

officers.! Here,! we! draw! on! all! of! these! data! to! present! a! set! of! recommendations! that! we!

believe,! if!earnestly! implemented,! will! enable! the!SDPD! to! eliminate! racial/ethnic! disparities.!

We!focus!our!recommendations!on!three!themes:!addressing!racial/ethnic!disparities;!building!

stronger!police-community!relations;!and!improving!data!collection!practices.!!

109

!Ridgeway,!G.,!(2009).!Cincinnati$Police$Department$traffic$stops:$Applying$RAND’s$framework$to$analyze$racial$

disparities.!Santa!Monica,!CA:!R AND!Corporation;!Hetey,!R.,!Monin,!B.,!Maitreyi,!A.,!&!Eberhardt,!J.!(2016).!Data$for$

change:$A$statistical$analysis$of$police$stops,$searches,$handcuffings,$and$arrests$in$Oakland,$Calif.,$2013-2014.!

Stanford!University,!CA:!Stanford!SPARQ,!p.!136.!

110

!See:!Hetey,!R.,!Monin,!B.,!Maitreyi,!A.,!&!Eberhardt,!J.!(2016).!Data$for$change:$A$statistical$analysis$of$police$

stops,$searches,$handcuffings,$and$arrests$in$Oakland,$Calif.,$2013-2014.!Stanford!U n ive rs ity,!C A :!Stan fo rd !SP A R Q ;!

Eberhardt,!J.!(2016).!Strategies!for!change:!Research!initiatives!and!recommendations!to!improve!police-

community!relations!in!Oakland,!Calif.!Stanford!University,!CA:!Stanford!SPARQ.!!!

74!

Addressing!racial/ethnic!disparities!

The! racial /ethn ic!disparities!we!found!in!the!treatment!of!Black!drivers!–!and!to!a!lesser!extent,!

Hispanic!drivers!–!are!by!no!means!unique!to!the!SDPD.!In!recent!years,!analyses!of!data!from!

state!and!local!jurisdictions!across!the!country!have!identified!similar!disparities!in!the!rates!of!

stops,!searches,!and!arrests.

111

!Moreover,!we!did!not!find!evidence!that!these!disparities!were!

the!result!of!deliberate!discrimination!or!racism!on!the!part!of!S DPD!officers.!Rather,!as!other!

researchers! of! racial/ethnic! disparities! in! policing! have! suggested,! “many! subtle! and!

unexamined!cultural!norms,!beliefs,!and!practices!sustain!disparate!treatment.”

112

Here,!we!discuss!4!recommendations!aimed!toward!the!elimination!of!systemic!disparities:!

Systemic$disparities$

1. Acknowledge! the! existence! of! racial/ethnic! disparities! and! make! combatting! such!

disparities!a!priority;!

2. Continue!to!enhance!training!and!supervision!around!issues!of!racial/ethnic!disparities;!

3. Make!traffic!stop!practices!more!transparent;!and!

4. Make!traffic!stop!practices!more!systematic!and!data-driven.!!

Acknowledge$that$racial/ethnic$disparities$exist$and$make$combatting$such$disparities$a$priority$

Previous! research! has! shown! that! there! is! a! strong! race–crime! association! not! just! among!

police!officers,!but!across!the!general!population!as!a!whole:!Black!faces!are!more!frequently!

associated!with!criminal!behavior!than!are!no n-Black!faces,!and!this!association!extends!to!how!

Black! people! –!youth! and! adult!alike! –!are! treated! throughout!the! criminal! justice!system.

113

This!is!known!as!implicit$or$unconsci ous$bias,!which!may!be! perpetuated!even!by!the!most!well-

meaning!people.!The!post-stop!disparities!noted!earlier!in!thi s!Report!suggest!that!implicit!bias!

may!exist!among!SDPD!officers.!

111

!See,!for!examples:!Baumgartner,!F.,!Epp,!D.,!&!Love,!B.!(2014).!Police$Searches$of$Black$and$White$Motorists.$

(Durham,$NC ).$Chapel!Hill,!NC:!University!of!North!Carolina-Chapel!Hill!Department!of!Political!Science.!Engel,!R.,!

Cherkauskas,!J.,!Smith,!M.,!Lytle,!D.,!&!Moore,!K.!(2009).!Traffic$Stop$Data$Analysis$Study:$Year$3$Final$Report,$

Prepared$for$the$Arizona$Department$of$Public$Safety.!Cincinnati,!OH:!Univ e rsity !of !Cin cin n ati!P o licin g!In stitu te ;!

Ross,!M.!Fazzalaro,!J.,!Barone,!K.,!&!Kalinowski,!J.!(2016).!State$of$Connecticut$Traffic$Stop$Data$Analysis$and$

Findings,$2014-2015.!Connecticut!Racial!Profiling!Prohibition!Project.!

112

!Eberhardt,!J.!(2016).!Strategies!for!change:!Research!initiatives!and!recommendations!to!improve!police-

community!relations!in!Oakland,!Calif.!Stanford!University,!CA:!Stanford!SPARQ,!p.!4.!!

113

!Eberhardt,!J.,!G off,!P.,!Purdie,!V.,!&!Davies,!P.!(2004).!Seein g!Black:!Race,!crime,!and!visual!processing.!Journa l$of$

Personality$and$Social$Psychology$87(6),!876-893;!Rattan,!A.,!Levine,!C.,!Dweck,!C.,!&!Eberhardt,!J.!(2012).!Ra c e !

Race!and!the!fragility!of!the!legal!distinction!between!juveniles!and!adults.!PLoS$ONE$7(5);!Hetey,!R.!&!Eberh ard t,!J.!

(2014).!Racial!disp arities!in !inca rcera tion !inc rea se!ac cep tan ce !of!p un itive!p olicie s.!Psychological$Science$25(10),!

1949-1954.!

75!

The!first!step!in!addressing!the!issue!of!racial/ethnic!disparities!is!acknowledging!that!they!exist!

and!making!it!a!departmental!priority!to!combat!such!disparities.!We!acknowledge!the!SDPD’s!

recent!efforts!to!do!this!by!incorporating!curricula!on!implicit!b ias,!emotional!intelligence,!and!

cultural!competency!into!its!training!for!front-line!officers!and!supervisors!(see!Appendix!11!for!

a!description!of!the!SDPD’s!current!officer!training!requirements).!!

Perhaps!partly!due! to!these!recent!training!efforts,! S DPD!officers!appear!to!already!be!aware!of!

these!issues!to!some!extent.!In!our!electronic!su rvey!of!the!department,!we!asked!officers!to!

assess!whether!they!believed!various!racial/ethnic!groups!feel$comfortable!interacting!with!the!

SDPD.! Just! over! a! third! –! 38.8! percent! –! of! offi cers! who! responded! to! our! survey! strongly!

agreed! or! agreed! that! Blacks! feel! comfortable! interacting! with! the! SDPD.! In! contrast,!

substantially! more! officers! believed! non-Black! citizens! feel! comfortable:!61.5! percent! believe!

Hispanics! feel! comfortable;! 80! percent! believe! Asians! feel! comfortable;! and! 87.5! percent!

believe!Whites!feel!comfortable!interacting!with!the!SDPD.!!

We!also!asked!officers!whether!they!believe!these!racial/ethnic!groups!have!confidence!in!the!

SDPD.! Th e! officers! who! responded! to! our! survey! believe! Blacks! have! the! lowest! confidence!

levels!in!the!SDPD:!35.2!percent!either!strongly!agreed!or!agreed!that!Blacks!have!confidence!in!

the!SDPD,!while!60.5!percent!believed!Hispanics!have!confidence;!78.9!percent!believed!Asians!

have! confidence;! and! 85.9! percent! believed! Whites! have! confidence! i n! the! SDPD.! These!

responses! indicate! that! officers! are! aware! of! how! they! may! be! perceived! by! different!

racial/ethnic!groups.!

However,! only! 4.23! percent! of! our! electronic! survey! respondents! strongly! agreed! or! agreed!

that!racial/ethnic$bias$is$a$genuine$problem$for$the$SDPD.!In!interviews!with!officers,! we!sought!

to!p robe!deeper!into!these!beliefs.!When!asked!whether!th ey!would!be!surprised!if!we!found!

racial/ethnic!disparities!in!our!analysis!of!the!traffic!stop!data,!the!vast!majority!of!officers!we!

spoke! to! expressed! beliefs! in! line! with! our! survey! respondents,! stating! that! they! would! be!

surprised!if!racial/ethnic!bias!were!to!b e!found!to!exist!in!how!traffic!stops!are!conducted!by!

the! Department.! A! typical! explanation! offered! to! us! by! officers! is! that! the! demographics! of!

drivers!who!are!stopped!are!a!reflection!of!the!composition!of!the!patrol!area.!As!one!officer!

explained,!!

The!community!I!work!in!is!a!predominantly!Hispanic!community.!The! p eople!I!pull! over,!

if!you!pull!my!data,!it's!gonna!show!that!the!people!I!pull!over!are!Hispanic…!So!there's!

disparity!there,!that!I'm!pulling!over!Hispanics!more!than!any!other!group!out!there.!But!

it's!not!because!of!my!perception!or!of!a!racist!view!I!have,!it's!because!of!where!I!work.!

76!

Indeed,!many!of!the!individual!officers!we!spoke!to!adamantly!stated!that!not!only!do!they!not!

make!individual!decisions!based!on!race/ethnicity,!but!also!that!in!the!traffi c!stop!context,!they!

frequently!cannot!see!the!race/ethnicity!of!the!driver!prior!to!pulling!them!over.!!

Only! a! handful! of! officers! directly! stated! that! race/ethnicity! is! a! factor! –! whether! explicit! or!

implicit!–!in!how!traffic!stop!decisions!are!made.!These!officers!spoke!ab out!the!“race/ethnicity!

out!of!place”!approach,

114

!in!which!officers!deliberately!target!individuals!whose!race/ethnicity!

does!not!fit!the!dominant!d emographics!of!the!area.!Offi cers!readily!offered!examples!of!this,!

such!as!stopping!a!White!person!in!a!predominately!Black!area!of!the!Southeastern!division,!or!

a!Black!person!in!a!majority-White!area!such!as!La!Jolla.!As!one!officer!candidly!noted,!“I'm!not!

going!to!lie.!If!I!see!somebody!that's!totally!out!of!place!and!there's!a!reason!to!stop!them,!I'm!

going!to!stop!them!and!ask!th em!what!they're!doing.!I!mean,!I'm!being!truth ful .!Unfortunately,!

it!sucks.!It's!not!like!I'm!trying!to.”!Most!other!officers,!however,!denied!using!race/ethnicity!in!

this!way.!One!officer!who!voiced!a!typical!statement!about!this!explained,!“I!am!not!looking!at!

who!the!dri ver!is,!whether!they!are!male,!female,!or!what!ethnicity!they!are.!That!is!not!what!I!

am!looking!fo r!because!I!do!not!write!a!citation!based!on!your!ethnicity.!I!write!it!based!on!the!

moving!violation!or!traffic!violation!that!you!did.”!!

Continue$to$enhance$training$and$supervision$

In!response! to! the! PERF!report,!the!San! Diego! Pol ice! Department!has!already!made! progress!

toward! establishing! a! comprehensive! training! program! for! its! patrol! officers! and! supervising!

officers! (see! Appendix! 11).! As! of! the! July! 2016! Public! Safety! and! Livable! Neighborhoods!

Committee! meeting,! the! SDPD! had! not! only! implemented! an! annual! supervisor! training! on!

procedural! justice,! but! had! also! added! competency! in! procedural! justice! and! community!

policing!concepts!to!its!promotional!testing!process.!The!SDPD!has!also!incorporated!a!two-day!

“effective!interactions”!class!on!unconscious!bias!for!all!new!officers.

115

The!Department!should!be!credited!for!its!prompt!response!to!these!recommendations.!As!the!

SDPD! makes! implicit! bias! curriculum! a! mandatory! part! of! how! both! new! and! veteran! patrol!

officers,!sergeants,!and!command!staff!are!trained,!it!should!track!officer!satisfaction!with!the!

training!to!ensure!maximal!efficacy!of!and!officer!buy-in!to!training!on!these!important!topics.!

114

!Carroll,!L.!&!Gonzalez,!M.L.!(2014).!Out!of!place:!Racial!stereotypes!and!the!ecology!of!frisks!and!searches!

following!traffic!stop s.!Jo urn a l$of$Re sea rch $in$C rim e$& $D elin qu en cy ,$51(5),!559 -584;!Novak,!K.!& !Chamlin,!M.!(2012).!

Racial!threat,!suspicion,!and!police!behavior:!The!impact!of!race!and!place!in!traffic!enforcement.!Crime$&$

Delinquency,$58(2),!275-300.!

115

!Zimmerman,!S.!(July!2016).!Update$of$the$San$Diego$Police$Department’s$response$to$the$Police$Executive$

Research$Forum$(PERF)$recommendations.!Testimony!submitted!to!the!Public!Safety!an d!Livable!Neighborhoods!

Committee!of!the!San!Diego!City!Council.!

77!

While! not! indicated! in! Chief! Zimmerman’s! testimony,! the! unconscious! bias! traini ng! may!

currently!be!drawn!from!two!providers.

116

!First,!the!Fair!and!Impartial!Policing!(FIP)!program

117

educates! patrol! officers! about! how! such! bias! affects! people’s! perceptions! and! can! thereby!

affect! the! actions! that! they! take,! as! well! as! providing! tools! to! help! offi cers! recognize! their!

conscious!and!unconscious!biases!and!in stead!take!actions!that!are!unbiased.!Traini ng!for!first-

line! supervisors! (sergeants)! helps! these! officers! to! identify! when! their! supervisees! may! be!

engaging!in!biased!behavior!as!well!as!to!effectively!address!such!behavior.!!!

Second,! the! Principled! Policing! training! has! been! developed! by! California’s! Department! of!

Justice! in! partnership! with! Stanford! University’s! Social! Psychological! Answers! to! Real-world!

Questions! (SPARQ)! organization.! Principled! Policing! is! the! first! Commission! on! Peace! Officer!

Standards! and! Training! (POST)-certified! training!on! procedural!justice! and!implicit! bi as! in! the!

U.S.! Thus! far,! it! has! been! offered! to! po li ce! leaders! throughout! Cali forni a,! including! to!

representatives!of!the!SDPD,!with!positive!results.

118

When! we! asked! our! community! focus! group! participants! about! how! to! improve! police-

community! relations,! many! agreed! that! law! enforcement! would! benefit! from! training! that!

would! enhance! their! ability! to! understand! –! and! effectively! respond! to! –! local! residents,!

particularly!those!from!diverse!cul tural!backgrounds.!Two!residents!from!different!divisions!put!

it!this!way:!

It!needs!to!be!more! of !a!partnership!model.!Police!are!in!the!power!position!and!instead!

of! being! more! militarized,! they! need! to! be! more! emotional ly! trained.! They! are! not!

soldiers;!they!are!here!to!keep!peace.!Come!around!more,!smile.!(Central!division)!

I! wish! [the! police]! took!a! body!language! class.! A! lot! of! things! that!are! going!wrong! is!

because!they!don’t!understand!the!body!language!of!the!community!or!the!cultures!of!

people! of! color.! We! speak! really! loud.! If! these! officers! are! not! from! our! culture! they!

don’t!understand!that.!(Southeastern!division)!

We! note! that! the! SDPD! has! recently! added! training! in! emotional! intelligence! and! effective!

interactions! to! its! new! officer! phase! training! and! we! encourage! the! tracking! of! officer!

satisfaction!with!such!training.!

116

!However,!we!note!that!a!third,!more!comprehensive!intervention,!consisting!not!only!of!implicit!bias!training,!

but!also!training!around!procedural!justice!and!reconciliation,!is!currently!being!piloted!in!six!U .S.!cities!by!the!

National!Initiative!for!Building!Community!Trust!and!Justice.!See:!https://trustandjustice.org/.!!

117

!http://www.fairandim partialpolicing.com.!!

118

!https://oag.ca.gov/sites/all/files/agweb/pdfs/law_enforcement/principled-policing-white- paper.pdf.!

78!

Make$traffic$stop$practices$more$transparent$$

Traffic!stops!can!be!one!of!the!most!d angerous!activities!a!patrol!officer!engages!in!on!a!regular!

basis;!there!is!no!such!thing!as!a!“routine”!traffic!stop.!Indeed,!a!vast!majority!of!officers!who!

responded!to!our!electronic!survey!–!96.1!percent!–!strongly!agreed!or!agreed!that!conducting!

a!traffic!stop!is!an!inherently!dangerous!activity.!Recent!events!involving!the!deaths!of!drivers!

and!of!police!officers!–!including!a!tragic!incident!in!the!summer!of!2016!here!in!San!Diego

119

!–!

further!heighten!the!tension!for!all!involved.!SDPD!officers!receive!extensive!training!on!how!to!

manage!their!own!safety!and!the!safety!of!the!cars!they!pull!over,!from!how!to!position !their!

vehicles!in!relation!to!that!of!the!cars!they!have!stopped!to!how!to!approach!a!car!and!identify!

potential! threats! to! their! safety.! Yet! this! training! does! not! eliminate! the! palpable! sense! that!

anything!can!happen!during!a!traffic!stop.!As!one!officer!described!it!to!us!during!an!interview,!

“Every!time!I!stop!a!car,!I!have!no!clue.!I!am!stopping!them!for!a!violation .!I!have!no!clue!what!

they!have!just!done,!what!they!were!going!to!go!do!or!what!they!might!have...!It!is!your!most!

dangerous![part!of!the!job]!–!you!are!rolling!the!dice!every!time.”!

Some!traffic!stops!may!further!impair!police-community!relations,!particularly!in!communities!

where! these! relations! may! already! be! strained.! Several! San ! Diego! residents! we! spoke! with!

expressed! a! beli ef! that! traffic! stops! are! conducted! in! a! discriminatory! fashion.! As! o ne!

Southeastern!resident!put! it,! “nine!times!out!of!ten,!it's! people!of!color![being!pulled!over]...!

That!will!make!them!feel!worse!about!the!police!because!they!make!you!feel!alienated!because!

of!your!skin!color.”!!!

Several! focus! group! members! also! expressed! concern! over! the! practice! of! calling! multiple!

patrol!vehicles!to!the!scene!of!a!vehicle!stop.!A!common!refrain!was!that!such!practices!have!

the!effect!of!heightening!the!anxiety!of!the!driver,!thereby!contributing!to!the!volatility!of!the!

interaction!and!alienating!other!members!of !the!community,!many!of!whom!see!thi s!practice!as!

a!gratuitous!or!even!provocative!demonstration!of!force.!As!one!resident!of!the!Southeastern!

division!stated,!!

If!they!are!pulling!people!over,!it!doesn't!take!four![cars]!to!pull!someone!over.!It's!very!

disrespectful! and! makes! more! of! a! scene.! I! don't! know! if! it's! to! show! power.! I!

understand! if! it's! two...if! someone! doesn't! have! a! partner! they! need! help.! It's! always!

three!or!more.!

119

!Kennedy,!M.!(2016,!July!29).!San!Diego!police!officer!shot!and!killed,!another!injured!following!traffic!stop.!

Southern$California$Public$Radio.!Retrieved!on!Aug.!24,!2016!from!

http://www.scpr.org/news/2016/07/29/63075/san-diego-police-officer-shot-and-killed-another-i/.!!

79!

In!interviews,!officers!u nderscored!the!value!of!the!routine!practice!of!officers!providing!back-

up!during!traffic!stops!du e!to!the!perceived!potential!dangers!of!such!stops.!While!this!back-up!

was!appreciated!(and!reciprocated)!by!the!patrol!officers!we!interviewed,!it!tends!to!engender!

resentment! among! community! residents,! particularly! those! who! may! not! understand! the!

perceived! and! real! risks! that! officers! face! during! these! encounters.! Reducing! the! number! of!

stops! made! for! violations! not! directly! related! to! public! safety! may! indirectly! improve!

community!relations,!given!community!members'!perceptions!about!such!stops.!!

Make$traffic$stop$practices$more$systematic$and$data-driven!!

Amongst!the!many!recommendations!recently!issued!by!President!Obama’s!Task!Force!on!21

Century!Policing

120

!was!the!following:!!

Law! enforcement! agencies! and! municipalities! should! refrain! from! practices! requiring!

officers!to!issue!a!predetermined!number!of!tickets,!citations,!arrests,!or!summonses,!or!

to! initiate! investigative! contacts! with! citizens! for! reasons! not! directly! related! to !

improving!public!safety,!such!as!generating!revenue.!!

We!found!no!evidence!of!the!use!of!quotas,!nor! pressure!to!issue! citati ons!to!increase! revenue.!

The!SDPD!and!the!City!of!San!Diego! s hou ld !be!commended!for!this,!in!light!of!recent!findings!of!

a!profit!motive!underlying!the!issuance!of!citations!in!other!jurisdictions!across!the!country.!!

However,!we!urge!the!SDPD!to!make! i ts!traffic!stop!practices!more!systematic!and!data-driven.!

Traffic!stops!in!San!Diego!appear!to!be!inconsistently!used!as!an!enforcement!tool,!which!may!

further! contribute! to! negative! perceptions! of! SDPD! activity.! In! interviews,! SDPD! officers!

described!highly!varying!approaches!to!and!justifications!for!making!traffic!stops.!Some!officers!

we!spoke!with!frequently!described!traffic!stops!as!being!useful!for!educational!purposes,!such!

as!reminding!drivers!that!they!should!not! be!texting!while! dri ving,!while!others!s tated!that!they!

hardly!conduct!any!traffic!stop s!at!all.!Still!others!touted!the!investigative!usefulness!of!traffic!

stops!to!uncover!criminal!activity.!This!speaks!to!a!highly-individualized!app roach! to! th is! f orm!o f!

law!enforcement,!which!suggests!one!way!in!which!disparate!treatment!can!arise.!!

As!noted!in!Chapter!5,!our!analysis!of!traffi c!stop!data!revealed!that!out!of!the!259,569!stops!

conducted!in!2014!and!2015,!only!981!resulted!in!the!discovery!of!contraband.!This!means!that!

contraband!was!found!in!fewer!than!one!out!of!every!260!traffic!stops!conducted!by!the!SDPD!

in!the!p ast!two!years.!Other!post-stop!outcomes!indicative!of!criminal!investigation!activity!are!

120!

President’s!Task!Force!on!21

!Century!Policing.!(2015).!Final$Report$of$the$President’s$Task$Force!on$21

$Century$

Policing.!Washington,!D C :!Off ice !of!C o m m u n ity !O rien te d!P o licin g!S erv ice s,!p .26 . $Retrieved!Aug.!24,!2016,!from!

http://www.cops.usdoj.gov/pdf/taskforce/Implementation_Guide.pdf.

80!

similarly!rare:!across!the!two!years,!roughly!4.4!percent!of!all!stops!led!to!a!search,!2.7!percent!

led! to! a! field! interview,! and! 1.3! percent! led! to! an! arrest.! Collectively,! the! finding! that! traffic!

stops! yield! minimal! crime!control! value!while! potentially! contributing! to! the! deterioration! of!

police-community! relations! point! to! the! need! for! a! reconsideration! of! how! traffic! stops! are!

used!in!law!enforcement.! This! recommendation! is!in! l in e! with!what! other!researchers! of! this !

topic! have! noted! –! that! “the! benefits! of! investigatory! stops! are! modest! and! greatly!

exaggerated,!yet!their!costs

are!substantial!and!largely!unrecognized.”

121

Given!the!post-stop!disparities!discovered!in!our!analyses,!we!urge!the!Department!to!con si d er!

how!it!might!devise!and!implement!policy!guiding!traffic!stops!to!address!this!issue.!

Strengthening!police-community!relations!

Drawing! primarily! on! the! data! we! collected! from! our! community! focus! groups! and! in-depth !

interviews!wi th!SDPD!officers,!as!well!as!the!evidence-based!recommendations!recently!made!

by! other! researchers,! we! discuss! two! recommendations! for! strengthening! p oli ce-communi ty!

relations,!particularly!in!police!divisions!where!these!relations!may!currently!be!strained:!

1. Make!community!engagement!a!core!departmental!value,!and!

2. Improve!communication!and!transparency!regarding!police!practices.!!

Make$community$engagement$a$core$departmental$value$

Community!residents!who!partici pated!in!our!focus!groups!indicated!a!strong!desire!to!see!and!

interact! with! police! officers! in! their! neighborhoods,! and! to! get! to! know! them! in! non-crime!

control! situations.! Residents! expressed! their! belief! that! the! best! way! to! improve! police-

community! relations! is! to! expand! opportunities! for! positive! p ol ice-commu n i ty! i nteraction .!

Likewise,! many! of! the! officers! we! interviewed,! particularly! those! who! work! in! division s! with!

higher!levels!of!crime!and!p ol ice!activity,!expressed!awareness!that!police-community!relations!

must!be!improved.!These!findings!are!wholly!consistent!with!those!of!the!PERF!report,!which!

found! a! beli ef! among! some! members! of! the! community! that! the! SDPD! has! become!

disconnected! from! the! communities! i t! serves.

122

! Thus,! we! urge! the! Department! to ! make!

community! engagement! a! core! departmental! value.! We! no te! that! this! is! a! central!

recommendation! of! President’s! Task! Force! on! 21

! Century! Policing,! which! stated! that! “in!

communities!that!have!high!numbers!of!interactions!with!authorities!for!a!variety!of!reasons,!

121

!Epp,!C.,!Maynard-Moody,!S.,!&!Haider-Markel,!D.!(2014).!Pulled$over:$How$police$stops$define$race$and$

citizenship.$Chicago,!IL:!University!of!Chicago!Press,!p.!153.!

122

!Police!Executive!Research!Forum!(PERF).!(2015).!Critical$response$technical$assessment$review:$Police$

accountability$-$findings$and$na tion al$im plica tion s$of$a n$as sessment$of$the$San$Dieg o$P olice $dep artm e nt.!

Washington,!DC:!Office!of!Community!Oriented!Policing!Services,!U.S.!Department!of!Justice,!p.!55.!

81!

police!should!actively!create!opportunities!for!interactions!that!are!positive!and!not!related!to!

investigation!or!enforcement!action.”

123

The! most! frequent! example! officers! offered! of! the! sign! that! police-community! relations! are!

suffering! in! at! least! some!parts! of!San! Diego! was! the! prevalence!of! the!“one-finger”! (middle!

finger)! wave! rather! th an! the! “five-finger”! wave.! Officers! use! this! as! an! indication! that! their!

presence!isn’t!welcome,!and!that!any!efforts!at!outreach!would!be!futile.!As!one!officer!put!it,!!

I!know!that!the! people! are! not! always!very! pol ice-friend ly.! I!would! never!stop! my! car!

and!just!say,!‘how!are! you! doing?’! because! I! am!going! to! get!the! one-finger! salute…! I!

think! in! a! community! where! people! are! mo re! police-friendly,! as! you! drive! down! the!

street,!i f!I!were!to!wave!at!someone,! th ey!would!wave!back!or!smile.!You!learn!people’s!

body!language.!They!intenti onal ly!turn!away…!You!get!the!feeling!that!they!do!not!like!

police!in!that!area.!!

These!officers!expressed!a!desire!for! greater! community! connection,! and! some!lamented!the!

fact! that! there! was! little! or! no! time! for! community! engagement! or! proactive! policing,! given!

staffing! constraints! and! the! ongoing! demands! of ! calls! for! service.! It! was! clear! from! these!

interviews!that!patrol!officers'!participation!in!community!events!across!the!nine!SDPD!divisions!

is!highly!variable!and!voluntary.!!!

When! asked! what! strong,! positive! police-commun ity! relations! would! look! like,! residents!

emphasized!that!they!would!involve!more!non-service!and!non-enforcement!interactions!with!

the!officers!who!police!their!communities.!The!residents!we!spoke!with!h ad!many!suggestions!

for!the!types!of!activities!they!would!like!to!engage!in!with!th e!officers.!It!is!important!to!note!

that!some!of!these!activities!are!already!occurring,!but!unevenly!across!the!city.!One!resident!of!

the!Southeastern!described!her!attendance!at!one!such!event!and!how!this!experience!made!

her!long!for!more!similar!opportunities!to!engage!with!officers:!

I!went!to!an!event!in!Skyline!and!it!was!awesome!to!connect!with!the!community.!The!

police! low-riders! were! out! and! they! were! bumping! old-sch ool ! and! it! was! cool! to! see!

STAR!PAL! (Sports! Training,!Academics,!Recreation/Police! Athletic! League).!It!made! me!

wish!there!were!more!programs!to!help!kids!respect!the!police.!This!experience!last!year!

made!me!feel! more!connected!to!the!police,!like!when!I!was!a!kid!(and!there!were!many!

more!events!between!police!and!residents).!

123

!President’s!Task!Force!on!21

!Century!Policing.!(2015).!Final$Report$of$the$President’s$Task$Force!on$21

$Century$

Policing.!Washington,!D C :!Off ice !of!C o m m u n ity !O rien te d!P o licin g!S erv ice s. $Retrieved!Aug.!24,!2016,!from!

http://www.cops.usdoj.gov/pdf/taskforce/Implementation_Guide.pdf.!

82!

Similarly,! a! resident! from! the! Southern! division! had! these! suggestions! for! fostering! positive!

relations:!!

…! a! carnival! to! get! to! know! each! other--for! residents! and! police! to! say! hi! and! get! to!

know!each!other;!a!community!meeting!every!month!where!we!tal k!about!our!fears!and!

concerns;!community!outreach!by!the!cops!in!our!community.!It’s!not!us!agains t!them—

they!are!here!to!help,!so!let’s!work!together.!

The!residents!we!spoke!with!want!to!get!to !know!their!local!police!officers!and!want!the!police!

get!to! know!them;!they!would!like!to!see!police!out!of!their!cars!and!interacting!with! residents .!

Several!residents!stressed!the!importance!of!nu rturing!relationships!between!police!and!youth,!

so!that!future!relationships!with!the!community!and! l aw! enf orcement! wil l!improve.!As!a!Central!

division!resident!observed,!!

If!officers!would!attend!community!events!with!kids!or!teenagers,!that!would!go!far!with!

respect.! Be! a! part! of! the! community…not! in! your! uniform.! Go! to! schools,! go! to! the!

community! garden.! It! will! just! take! the! police! Department! to! want! to! do! that.! When!

people!see!that!they!are!o n!the!same!level!they!will! feel!freer!to!express!themselves!and!

get!the!help!they!need.!

Our!focus!group!participants'!suggestions!echo!those!noted!in!the!recent!analysis!of!the!S DPD!

conducted! for! the! PERF! report,! in! which! the! most! frequent! suggestions! from! community!

members! were! related! to! maximizing! police-community! engagement! “through! proactive! and!

positive!interactions.”

124

We!acknowledge!the!SDPD’s!existing!community!engagement!activities. !In!our!interviews!with!

officers! at! all! ni ne! SDPD! divi sion s,! it! was! evident! that! each! division's! Community!

Liaison/Resource!Officers!have!attempted!to!connect!with!residents!through!a!wide!variety!of!

meetings! and! events! and! are! disseminating! information! and! sharing! resources! in! multiple!

venues.!Further,!it!is!clear!from!both!our!officer!interviews!and!community!focus!groups!that!

many!patrol!officers!are!community-minded!and!enjoy!opportunities!to!positively!engage!with!

residents!while!on!patrol.!In!addition! to!the!various!community!safety!and!prevention!programs!

offered!through!the!SDPD,!including!the!youth!programs!STAR!PAL!and!KIDZWATCH!Academy,!

the! Department! also! collaborates! with! local! clergy! and! advocacy! groups! in! various!

124

!Police!Executive!Research!Forum!(PERF).!(2015).!Critical$response$technical$assessment$review:$Police$

accountability$-$findings$and$na tion al$im plica tion s$of$a n$as sessment$of$the$San$Dieg o$P olice $dep artm e nt.!

Washington,!DC:!Office!of!Community!Oriented!Policing!Services,!U.S.!Department!of!Justice,!p.!22.!

83!

neighborhood-based! initiatives.

125

! Another! way! the! SDPD! currently! promotes! community!

engagement! is! through! a! program!called! Inside!SDPD,! in!which! some!sessions! of! new! officer!

training!that!every!new!recruit!attends!are!open!to!the!public.!Inside!SDPD!all ows!citizens!the!

opportunity! to!receive! some! of! the! same! training! the!Department! provides! to! its! officers!on!

topics!such!as!use!of!force,!procedural!justice,!and!non-biased!based!policing.!

We!recommend!that!the!SDPD!create!a!system!to!make!positive,!community-based!interactions!

and!activities!a!fundamental!component! of !officers’!roles!and!to!incentivize!officers’!commun ity!

engagement! activities.! We! also! recommend! that! the! SDPD! further! publicize! and! raise!

awareness!about!existing!community!meetings!and!events,!and!create!additional!opportunities!

for!officers!and!the!communi ty!to!interact.!We!suggest!that!such!interactions!involve!more!of!

each! police! division's! officers! –! not! just!Community! Liaison/Resource!officers! –!perhaps! on!a!

rotating!basis,!and!that!the!communities!with!higher!crime!and!lower!police!trust!are!prioritized!

in!this!process.!!

Improve$communication$and$transparency$regarding$police$practices$$

Both! community! residents!and! law! enforcement!officers! i nterviewed! in!our! study!recognized!

that!tension!exists!and!desired!better!communication!and!understanding.!!Several!officers!we!

spoke!with!wished! community! members! b etter! understood!the! challenges!and! constraints!of!

their!jobs,!and!many!community!members!desired!more!information!about!local!crime!issues!

and!poli ce!decision-making.!Police!officers! expressed!a!desire!for!more!citizens!to!request!to!go!

on! police! ride-alongs! so! they! could! witness! the! challenges! officers! regularly! face.

126

! Citizens!

wished!officers!would!share!more!information!about!crime!problems!in!their!communities!and !

efforts!underway!to!address!them.!As!previously!noted,!they!also!desired!more!interaction!and!

collaboration.!!

Expanding!and!improving!the!lines!of!communication!between!police!and!residents!should!be!a!

high! priority.! The! SDPD! should! seek! additional! op portun iti es! for! information-sharing! and!

clarification! of! police! practices! and! procedures! in! the! communities! they! serve.! Greater!

transparency! and! communication! about! these! practices! will ! strengthen! community! trust! and !

perceptions! of! police! legitimacy.

127

! Ongoing! communication! strategies! utilizing! social! media!

outlets!(Facebook,!Twitter,!Nextdoor,!etc.)!and!websites!should!continue,!but!more!face-to-face!

125

!See!the!SDPD!website!for!more!information!on!community!policing!and!crime!prevention!activities:!

https://www.sandiego.gov/police/services/prevention/programs!(Retrieved!Sept.!28,!2016).!!!

126

!Any!member!of!the!commun ity!ca n!re qu es t!a!ride-along!through!this!online!form:!

https://www.sandiego.gov/sites/default/files/legacy/police/pdf/RideAlong.pdf!(Retriev ed !Sep t.!28 ,!20 1 6).!

127

!See:!Advancement!Project!and!PolicyLink.!(2014).!Engaging$communities$as$partners:$Strategies$for$problem$

solving.!Part!of!the!Beyond$confrontation:$Community-centered$policing$tools!series.!Los!Angeles,!CA:!Urban!Peace!

Institute.!R et rie ve d !S e p tem ber!8,!2016,!from:!http://www.urbanpeaceinstitute.org/key-projects/.!!

84!

outreach!is!needed,!especially!in!the!communiti es!where!police!trust!is!low!and!residents!are!

concerned! about! crime! and! safety,! yet! suspicious! of! police! crime! control! strategies.! In! our!

study,! Southeastern ! and! Mid-City! were! the! communities! that! were! most! vocal ! in! asking! for!

greater!police!communication.!!As!two!Southeastern!residents!noted:!!!!!!!!!!

!!!!

They! could! do! more! meetings,! maybe! get! involved! in! neighborhood! watches.! The!

community!needs!to!have!awareness!(about!local!crime!problems)!and!get!to!know!the!

cops;!give!us!their!cards!and!do!outreach…!build!a!relationship!between!the!police!and!

the!school!district.!!

If!they!would!actually!walk!beats!and!get!to!know!people;!I!would!like!if!they!have!an!

officer!meet!and!greet!to!introduce!yourself!or!share!input!or!su ggestions—to !increase!

familiarity.!!!

The! SDPD! is! to! be! credited! for! the! communication! and! information/resource! dissemination!

already! underway,! but! additional ! work! is! needed.! As! noted! in! the! previous! section,! several!

residents! expressed! concern! and! confusion! abou t! traffic! stop! practices! in! their! communities,!

particularly!related!to!the!number!of!cars!and!officers!involved!in!such!stops.!The!SDPD!should!

explain!the!rationale!behind!these!decisions!and!ad dress!communities’! concerns.!Obtaining!the!

support! of! community! members! in! local! law! enforcement! can! be! a! challenging! task,! but! we!

note! that! there! are! several! effective! model s! for! doin g! so.

128

! We! recommend! that! the! SDPD!

consider! adopting! one! of! these! models,! and! in! doing! so,! identify! new! ways! to! promote!

transparency!and!communicate!information!about!local!crime!and!police!enforcement!practices!

with!community!residents,!particularly!in!neighborhoods!with!higher!levels!of!police!presence,!

where!police-community!relations!are!most!strained.!!!

Improving!data!collection!

Finally,! we! incl ud e! five! broad! recommendations! germane! to! the! collection,! analysis,! and!

dissemination!of!data!related!to!SDPD’s!traffic!enforcement!regime:!

1. Revise!the!current!data!collection!system;!

2. Coordinate!existing!data!collection!efforts;!

3. Collect!additional!data;!and!

128

!See:!President’s!Task!Fo rce!on!21

!Century!Policing.!(2015).!Final$Report$of$the$President’s$Task$Force!on$21

Century$Policing.!Washington,!D C :!Of fice !of!C o m munity!Orien ted !P olic ing !Se rv ices . $ Retrieved!Aug.!24,!2016,!from!

http://www.cops.usdoj.gov/pdf/taskforce/Implementation_Guide.pdf;!Advancemen t!P ro je ct !a n d !P o licy Lin k .!

(2014).!Engaging$communities$as$partners:$Strategies$for$problem$solving.!Part!of!the!Beyond$confrontation:$

Community-centered$policing$tools!series.!Los!Angeles,!CA:!Urban!Peace!Institute.!Retrieved!September!8,!2016!

from!http://www.urbanpeaceinstitute.org/key-projects/.!

85!

4. Strengthen!accountability!and!oversight!of!data!collection!and!management!$

Revise$the$current$data$collection$system!

The!Department’s!current!traffic!stop!data!collection!system,!which!relies!heavily!on !the!traffic!

stop! data! card,! produces! duplicative,! often! inaccurate! and! unreliable! data,! is! unnecessarily!

time-consuming,! and! harmful! to! officer! morale.! For! these! reasons,! we! recommend! that! the!

SDPD!discontinue!the! use!of!the!traffic!stop!data! card!in!favor!of!a! system!that!captures!and!

compiles!data!gathered!by!officers!through!other!means.!

Stop$ ca rd$data$are$duplicative.$At!the!conclusion!of!a!traffic!stop,!SDPD!officers!must!document!

the!contact!in!several!different!ways.!If!the!stop!involved!th e!issuance!of!a!citation!or!a!written!

warning,! the! officer! must! complete! the! requisite! paperwork.! The! officer! must! complete! an!

additional!set!of!forms!if!they!conduct!a!field!interview,!a!search,!or!an!arrest.!Next,!they!must!

describe!every!encounter!in!a!separate!form,!called!a!“journal,”!an!internal!mechanism!used!to!

track!officer! productivity.! They!must! then! submit!an!additional! form!logging!their! body-worn!

camera!footage.!Finally,!they!must!then!complete!the!traffic!stop!data!card.!$

In!interviews,!SDPD!officers!described!this!documentation!process!as!both!time-consuming!and!

filled!with!redundancy.!Many!also!noted!that!much!of!the!data!captured!by!vehicle!stop!cards,!

including!driver!race,!gender,!age,!and!stop!location,!is!information!already!captured!by!many!

of!the!other!forms!they!submit.!This!is!a!key!point:!Eli minati ng!the!traffi c!stop! d ata!card!wil l!not!

hinder!the!Department’s!ability!to!document!traffic!enforcement!patterns,!nor! will!the!public!

lose!oversight!ability.!

Excessive!paperwork!is!a!noted!source!of!officer!stress,

129

!a!fact!no!doubt!amplified!by!staffin g!

shortages!and!other!resource!deficiencies.!Whether!owed!to!the!time!it!takes!to!complete!the!

paperwork,!the!notion!that!they!are!not!trusted!and!thus!must!document!every!action!taken,!or!

some!other!reason,!we!believe!that!the!stress!associated!with!the!use!of!the!traffic!stop!cards!

contributes!to!relatively!low!morale!Department-wide.!!

Stop$ cards$ harm$ officer$ morale.$ Lingering! questions! about! the! broad! purpose! of! the! data!

collection!effort!and!the!stop!card!data!in!particular!li kely!contribute!to!the!sense!that!the!stop!

cards!represent!unnecessary,!extraneous,!and!even!frivolous!work.!In!the!words!of!one!officer,!

“The! coll ection ! of! traffic! stop! data! is! useless.”! Others! called! the! process! a! “waste! of! time,”!

129

!Crank,!J.!P.,!&!Caldero,!M.!(1991).!The!production!of!occupational!stress!in!medium-sized!police!agencies:!A!

survey!of!line!officers!in!eight!mu n icip a l!d e p ar tments.!Journa l$of$C rim ina l$Jus tice,!19,!339-349;!Zhao,!J.S.,!He,!N.,!&!

Lovrich,!N.!(2002).!Predicting!five!dimensions!of!police!officer!stress:!Looking!more!deeply!into!organizational!

settings!for!sources!of!police!stress.!Police$Quarterly,!5,!43-62.!

86!

“worthless,”! “stupid,”! and! a! “joke.”! Off icer! survey! responses! make! the! point! more!

systematically:! 72! percent! of! respond ents! either! disagreed! or! disagreed! strongly! with! the!

notion!that!“completing!the!traffic!stop!data!card!is!a!worthwhile!use!of!officer!time.”!Several!

officers! also! reported! feeling!as! though! the! data! gathered! would! be! used! to!unfairly! portray!

their! work! as! biased.! As! one! officer! put! it,! “[r]egardless! of! the! outcome,! the! data! will! be!

misconstrued! and! manipulated.”!In! the! words!of! another,! “[in!completing! the! card],!I! feel! as!

though!I’m!having!to!prove!I’m!not!a!racist!after!every!traffic!stop.”!$

The!effects!of!officer!cynicism!over!use!of!the!stop!cards!appears!to!stretch!beyond!morale.!In!

an!effort!to!avoid!being!characterized!as!bi ased,!several!officers!discussed!instances!where!they!

chose!not!to!submit!a!stop!card!following!a!stop!involving!minority!drivers,!or!mislabelin g!th e!

driver’s! race/ethnicity! on! the! stop! card.! Others! acknowledged! choosing! not! to! stop! minority!

drivers! altogether! i n! hopes! of! avoiding! the! possible! ramifications! of! the! encounter.! That! the!

data!collection!regime!is!contributing!to!what!scholars!refer!to!as!depolicing!suggests!strongly!

that!there!is!need!for!reform.!

Stop$card$ data$ are$ unreliable.$As!we!noted! in! Chapter!3,! and!very!much!related! to! the! point!

about!depolicing,!the!traffic!stop!records!used!in!this!analysis!was!of!relatively!low!quality.!The!

dataset!contained!several!instances!of!missing!data,!a!problem!that!was!most!apparent!among!

post-stop!variables.!Data! charting!the!issuance!of!citations!or! warnings!was!absent!from!10.6!

percent!of! the!259,569!stops!recorded!between!2014!and!2015.!Data!on!field!interviews!(7.9!

percent),!searches!(4.4!percent),!and!arrests!(4. 1!percent),!were!also!missing!in!relatively!high!

volume.!Of!the!poorest!q ual ity!were!data!associated!with!the!discovery!of!contraband!and!the!

seizure!of!property,!where!over!93!percent!were!either!left!blank!or!ambiguously!labeled,!‘null.’!$

The!problems!associated!with!missing!cases!are!amplified!by!what!appears!to!be!the!substantial!

under-reporting! of! traffic! stops.! As! we! have! noted! previously,! SDPD! records! in di cate! that!

183,402!traffic! tickets!were!issued!between!January!1,!2014!and!December! 31,!2015.!Yet!the!

Department’s! stop! card! database! includes! records! of! only! 145,490! stops! where! drivers! were!

issued! a! citation.! Th e! sizable! difference! between! actual! citations! and! reported! citations!

suggests!that!tens!of!thousands!of!traffic!stops!went!undocumented.!!

This!disparity!raises!significant!questions!about!the!reliability!of!data!set!used!for!this!analysis,!

particularly! in! light! of! missing! stop! card! data! and! the! inconsistent! month-to-month!

enforcement!trends.!These!data!quality!issues!are!n ot!new.!In!fact,!Cordner!and!his!colleagues!

raised!a!very!similar!set!of!concerns!in!their!2001!analysis!of!SDPD!traffic!stops:!!

87!

This! very! substantial! [year-to-year]! decrease! [in! stop! card! records]! raises! serious!

questions!about!the!validity!of!the!vehicle!stop!d ata.!One!question!is!whether!officers!

always! filled! out! the! vehicle! stop! forms! –! the! answer! to! this! is! clearly! no.! A! natural!

follow-up!question!asks!what!the!compliance!rate!was!–!this!can!only!be!estimated,!but!

it!appears!to!have!been!about!60%.

130

The! consistency! of! our! findings! with! those! articulated! by! Dr.! Cordner! speaks! to ! a! series! of!

systemic!weaknesses!that!must!be!addressed!before!the!SDPD!is!able!to!generate!a!thorough,!

accurate!reporting!of!officer!traffic!enforcement.!For! these!reasons,!we!recommend!eliminatin g!

the! use! of ! the! traffic! stop! data! card! and! replacing! the! current! system! with! a! modified! data!

collection!and!management!infrastructure.!!

Coordinate$existing$data$collection$efforts$

The!recommendation!to!replace!the!traffic!stop!data!card!is!predicated!on!th e!development!of!

a! more! effective,! more! efficient! system! for! tracking! vehicle! stops! and! post-stop! outcomes.!

Collection!of!stop!card!data!should!not!be!discontinu ed!unless!and!unti l!a!viable!replacement!

system!is!up!and!fully!operational.!!

The! current! SDPD! system! of! data! collection! and! management! is! defined! by! duplication! and!

siloed! information. ! We! believe! the! department’s! current! architecture! contains! many! of! the!

necessary! components! of! a! more! usable,! and! thus! more! valuable! system! based! on! the! data!

collected!via!the!CAD!system,

131

!traffic!citations!and!written!warnings,!as!well!as!forms!officers!

are! required! to! submit! in! documentation! of! field! interviews,! search/seizure! incidence,! and!

arrests.!!!

Additional$data$collection$

In!addition!to!the!data!currently!collected,!we!recommend!the!SD PD!capture!and!incorporate!

the!following!information!into!the!new!database:!!

• Police!officer!race,!gender,!unit!(e.g.,!Gang!Unit,!Auto!Theft!Unit,!etc.)!and!division!(e.g.,!

Traffic!division)!

• Specific!stop!location!(address,!intersection,!and/or!landmark)!!

• Vehicle!make,!model,!and!condition!

130

!Cordner,!G.,!W illiams,!B.,!&!Zuniga,!M.!(2001).!San$Diego$Police$Department$vehicle$stop$study:$Yea r-end$report.!

San!Diego,!CA.,!p.!1-2.!

131

!For!an!introduction!to!police!CAD!systems!and!a!usefu l!description!of!the!standard!capability!of!such!systems,!

see!Law!Enforcement!Information!Technology!Standards!Council!(LEITSC).!(n.d.).!Standard$Functional$Specifications$

for$Law$Enforcement$Compu ter$A id e d$D is p at ch $(C A D )$S y ste m s .!B ur ea u!o f!Ju stic e!A ss ista n ce,!O ff ice!o f!Ju stic e!

Programs,!U.S.!Department!of!Justice.!Retrieved!Aug.!14,!2016,!from,!!

!https://www.it.ojp.gov/documents/LEITSC_Law_Enforcement_CAD_System s.pdf.!

88!

• Description!of!driver!behavior!and!demeanor!

• Probable!cause!search!

• Nature!and!amount!of!contraband!discovered!and!property!seized!

Augmenting! the! current! data! collection! efforts! with! th ese! additional! data! would! put! SDPD!

squarely!in!line!with!best!practices!and!woul d!yield!significant!benefits!both!for!the!SDPD!and!

the!City!of!San!Diego.!!

Officer$ information.! SDPD’s! current! traffic! stop! data! card! contains! no! information! about! the!

officer! conducting! the! stop,! and! thus! no! such! information! was! available! for! the! present!

analysis.! To! our! knowledge,! most! if! not! all! of! the! existing! data! co ll ection! mechanisms,! from!

traffic!citations!to!search!detail!forms,!are!associated!with!officer!badge!numbers,!which!seems!

to!suggest!that!the!i ncl usi on!of!basi c!information!ab out!the!officer!may!not!represent!a!major!

challenge.!!

Officer!data!are!essential!for!charting!enforcement!patterns!at!the!officer!level!–!necessary!for!

identifying! so-called! “rotten! apple”! officers.

132

! The! Department’s! existing! early! intervention!

system,!a!point!of!emphasis!in!the!2015!PERF!report,

133

!has!the!potential!to!be!very!useful!in!

this! regard.! We! also! believe! that! officer! data! may! hold! the! key! to! more! effectively!

understanding!the!role!that!race/ethnicity!plays!in! d riving!stop!and!post-stop!patterns.!Scholars!

have!found!in!several!ins tances!that!disparities!are!most!pronounced!in!cases!where!the!officer!

and!the!driver!are!of!different!racial!or!ethnic!backgrounds!(for!example,!when!a!White!officer!

stops,!searches,!or!arrests!a!Black!driver).

134

!The!quality!of!future!analysis!of!SDPD’s! traffi c!stop!

patterns! would! be! strengthened! consi derably! by! the! capture! of! officer! race/ethnicity! and!

gender!data.!!

Stop$ location.! In!one-on-one!interviews,! several! SDPD!officers!noted! that! traffic!enforcement!

patterns!follow!closely!th e!crime!and!demographic!trends!of!the!stop!location.!In!the!words!of!

one!officer,!“The!popul atio n!in!the!area!I!patrol!is!mainly!Hispanic!or!Black.!Therefore,!majority!

of!the!traffic! stops,!criminals,!etc.!are!going!to! be!those!ethnicities.!It!has!nothing! to! do!with!

132

!For!an!example!of!what!this!analysis!might!look!like,!see!Rid geway,!G.,!(2009).!Cincinnati$Police$Department$

traffic$stops:$Applying$RAND’s$framework$to$analyze$racial$disparities.!Santa!Monica,!CA :!RAN D!Corporation,!pp.!

43-48.!

133

!Police!Executive!Research!Forum!(PERF).!(2015).!Critical$response$technical$assessment$review:$Police$

accountability$-$findings$and$na tion al$im plica tion s$of$a n$as sessment$of$the$San$Dieg o$P olice $dep artm e nt.!

Washington,!DC:!Office!of!Community!O rien ted !Po licing !Se rvice s,!U.S .!De pa rtm en t!o f!Justice .!

134

!Tillyer,!R.!Klahm ,!C.F.,!&!Engel,!R.S.!(2012).!The!discretion!to!search:!A!multilevel!examination!of!driver!

demographics!and!officer!characteristics.!Journal$of$Contempo ra ry$C rim inal$Justice,!28(2),!184-205;!Brown,!R.A.,!&!

Frank,!J.!(2006).!Race!and!officer!decision!making:!Examining!differences!in!arrest!outcomes!between!Black!and!

White!officers.!Justice$Quarterly,!23,!96-126.!

89!

race,!but!the!population!itself!in!the! ci ty.”!Other!officers!suggested!that!traffic!stops!are!used!as!

a!means!of!investigating!and!controlling!crime.!We!believe! an alysis !of!the!relationship!between!

traffic! enforcement! and! crime!control! is!hugely! important! and! potentially! benefi cial! both! for!

law!enforcement!purposes!and!for!enhancing!external!oversight!and!accountability.!!

Yet! this! type! of! place-driven! analysis! is! not! possible! when! limited! to! division-level! data.!

Criminological! research! has! established! definitively! that! crime! is! not! randomly! dispersed!

throughout! a! city! or! even! a! neighborhood.

135

! Instead,! what! we! heard! from! SDPD! officers! is!

largely!consistent!with !the!current!research:!hot!spots!o f!illegal!activity!vary!by!crime!type!and!

are! a! function!of! time!of! day,! time! of! year,!and,! most!importantly,! by! very!narrowly! defined!

spaces.

136

! In! fact,!the! relatio nsh ip ! between! crime! and!place! is! most! effectively! considered!at!

the!“micro”!level.

137

!According!to!one!recent!study,!these!crime!places!“can!be!as!small!as!the!

area! immediately! next! to! an! automatic! teller! machine! or! as! large! as! a! block! face,! a! strip!

shopping!center,!or! an!apartment!building.!Often!places!are!thought!of!as!addresses,!specific!

types!of!businesses,!or! blockfaces.”

138

!As!such,!we! recommend! that!stop!data!be!captured!in!

terms!of!the!specific!location!of!the!encounter,!rather!than!by!division.!!!

Further$ stop-related$ detail.! We! recommend! that! the! SDPD! incorporate! into! existing! data!

collection! efforts!the! make,!model,! and! condition! of! the! driver’s! vehicl e,! as! well!as! stop!and!

post-stop!data!on!stops!involving!cyclists!and!pedestrians.!!

An!officer’s! knowl edge!of!his!or!her!beat!is!critical!to!good!police!work!in!part!because! it!allows!

the! officer! to! recognize! and! act! on! incongruities.

139

! Community! policin g! is! premised! on! this!

notion:!police!work!to!get!to!know!the!community!not!only!to!foster!trust,!but!also!to!develop!

the! skills! to! be! able! to! distin guish ! interlopers! from! residents.

140

! The! same! is! true! of! patrol!

officers.!A!consistent!theme!from!our!interviews!with!SDPD!staff!was!the!importance!of!traffic!

stops! for! investigating! circumstances! or! individual s! that! may! appear! out! of! place.! Language!

used!to!d escribe!vehicles!that!appear!incongruous!often!goes!hand-in-hand!with!discussion!of!

an!individual! of! a!particular!race/ethnicity! who! appears!out!of!place! in! certain!neighborhood!

135

!Braga,!A.A.,!&!Weisburd,!D.L.!(2010).!Policing$Problem$Places.!Oxford,!UK:!Oxford!U niv er sity !Pre ss .!

136

!Sherman,!L.!W.,!Gartin,!P.!R.,!&!Buerger,!M.!E.!(1989).!Hot!spots!of!predatory!crime:!Routine!activities!and!the!

criminology!of!place.!Criminology,!27,!27-56.!

137

!Groff,!E.R.,!W eisburd,!D.,!&! Yan g,!S.!(2010).!Is!it!important!to!examining!crime!trends!at!the!‘micro’!level?:!A!

longitudin a l!a n aly sis !o f!st ree t!v ar iab ilit y!in !c rime!trajectories .!Journ a l$of$Q u an titativ e$Criminology,!26,$7-32.$

138

!Eck,!J.E.,!&!Weisburd,!D.!L.!(2015).!Crime!places!in!crime!theory.!Crime$and$place:$Crime$prevention$studies ,! 4.!

Retrieved!Aug.!10,!2016,!from!

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.362.1293&rep=rep1&type=pdf.!

139

!Mastrofski,!S.!D.!(1983).!Police!knowledge!of!the!patrol!beat:!a!performance!measure.!Police$at$Work:$Policy$

Issues$an d $A n a ly sis ,$S a g e$P u b lic a tio n s,$B e v erly $H ills ,$C A,!45 -64.!

140

!Greene,!J.!R.!(2000).!Comm unity!policing!in!America:!Changing!the!nature,!structure,!and!function!of!the!

police.!Criminal$justice,!3(3),!299-378.!

90!

contexts.!As!one!officer!put!it,!“I!do!not!write!many!tickets.!I!pull!people!over!that!I!think!might!

be!doing!bad!things.!Am!I!going!to!pull!over!the!guy!coming!home!from!work!because!he!does!

not!have!a!front!license!plate?!No.!If!I!see!two!peopl e!wearing!hoodi es!with!their!hoods! u p!in! a!

Tesla,!yeah!chances!are!I!am!pulling!them!over.”!

Relatedly,!we!believe!that!the!SDPD!would!benefit!from!capturing!data!on!individual!behavior!

and! demeanor.! Particular!behaviors! on! the!part! of! either!the! driver! or!passenger! –! apparent!

nervousness,! aggression! or! combativeness,! even! obsequiousness! –! are! often! associated! with!

suspicion!and!thus!used!to!justify!a!field!interview,!request!for!permission!to!search,!or,!when!

combined!with!other!factors,! a!probabl e!cause!search.

141

!That!an!officer’s!p erception!of!certain!

behavior!may!be!unwittingly!influenced!by!driver/pedestrian!race/ethnicity!(and!stop!context)!

is! a! fundamental! component! of! implicit! bias! in! law! enforcement.! More! to! the! point,!

racial/ethnic! differences! in! the! characterization! of! a! vehicle! as! being! out! of! place! or! in! the!

interpretation!of!certain!behavior,!have!been!consistently!linked!to!racial/ethnic!disparities!in!

the! treatment! of! drivers.

142

! This! i s! critically! important! in! light! of! the! wide! search! and! field!

interview!disparities!found!between!White!and!minority!drivers.!!

Collection! of! vehicle! data! and! driver! behavior/demeanor! information,! which! is! wi dely!

considered!best!practices,

143

!would!add! depth! and! insight!i nto! future!anal ysis,! in!the! process!

allowing!the!SDPD!to!more! effectively! disentangle! manifestations! of! bias!from! those! of ! solid,!

proactive!policing.!

We!further!recommend!that!the!SDPD!collect!and!track! an !additional!mechanism!for!evaluating!

racial/ethnic! disparities! in! the! enforcement! of! traffic! regulations:! stop! duration.! From! mere!

inconvenience! to! other! job-! or! family-related! costs,! the! length! of! a! traffic! stop! can! have!

substantial! ramifications! for! drivers,! regardless! of! whether! the! stop! ends! with! a! citation,! a!

warning,! or! some! other! outcome.! Discussion! of! the! issue! among! community! focus! group!

members!often!reflected!research!that!has!found!that!these!costs!are!often!weigh!more!heavily!

141

!Alpert,!G.!P.,!MacDonald,!J.!M.,!& !Dunham,!R.!G.!(2005).!Police!suspicion!and!discretionary!decision!making!

during!citizen!stops.!Criminology,!43(2),!407-434.!

142

!Eberhardt,!J.,!G off,!P.,!Purdie,!V.,!&!Davies,!P.!(2004).!Seein g!Black:!Race,!crime,!and!visual!processing.!Journa l$of$

Personality$and$Social$Psychology$87(6),!876-893;!Novak,!K.!&!Chamlin,!M.!(2012).!Racial!threat,!suspicion,!and!

police!behavior:!The!impact!of!race!and!place!in!traffic!enforcement.!Crime$&$Delinquency,$58(2),!27 5-300.!

143

!Tillyer,!R.,!Engel,!R.S.,!&!Cherkauskas,!J.C.!(2010).!Best!practices!in!vehicle!stop!data!collection!and!analysis.!

Policing:$An$International$Journal$of$Police$Strategies$&$Management,!33(1),!69-92.;!Ramirez ,!D .,!McDevitt,!&!

Farrell,!A.!(2000).!A!resource!guide!on!racial!profiling!data!collection!systems:!Prom ising!practices!and!lessons!

learned.!U.S.$Department$of$Justice.!Retrieved!Aug.!15 ,!20 1 6,!fro m!

https://www.ncjrs.gov/pdffiles1/bja/184768.pdf.!

91!

on!minority!drivers,!as!their!stops!have!been!shown!to!last!longer!than!those!involving!White!

drivers.

144

Finally,! we! recommend! that! the! SDPD! take! steps! to! increase! the! specificity! of! their!

documentation! of! post-stop! outcomes! in! two! ways:! (1)! begin! tracking! searches! justified! by!

probable!cause;!and!(2)!documenting!the!specific!nature!and!amount!of! con traband!discovered!

and!property!seized.!!

Pedestrian$ and$ bicycle$ stop$ data.$On! October!3,! 2015,! Governor!Jerry! Brown! signed! into! law!

Assembly!Bill!953,

145

!which!requires!all!law!enforcement!agencies!in!the!State!of!California!to!

collect!and!disseminate!data!on!all!traffic!and!pedestrian!stops.!The!SDPD!must!submit!its!first!

report!to!th e! State’s! Attorney!General!by!April!1,!2019.!We! urge!the!Department!to!institute!

and! implement! policy! mandating! data! collection! for! pedestrian! and! bicycle! stops! well! in!

advance!of!the!AB!953!mandate.!Further,!we!urge!the!department!to !distinguish!by!stop!type!

(vehicle,! bicycle,! or! pedestrian)! data! on! relevant! post-stop! outcomes,! including! search,!

contraband!discovery,!and!property!seizure,!as!well!as!field!interview,!arrest,!and!citation.$

Strengthen$accountability$and$oversight$of$data$collection$and$management$$

Regardless!of!which! ap proach!the!SDPD!takes!toward! f uture!data!collection!efforts,!we!strongly!

recommend!that!the!Department!institute!a!more!robust!set!of!data!imputation!qu ali ty!control!

mechanisms.! Adoption! of! the! recommendation! to! replace! the! current! system! with! o ne! that!

draws!more!heavily!on!data!from!the!CAD!system!and!incorporates!inf ormation!generated!by!

judicial!records,!including!traffic!citations!and!other!post-stop!forms,!would!likely!reduce!some!

of! the! quality! assurance! requirements,! as! their! value! as! legal! documents! is! predicated! on!

thoroughness! and! accuracy.! However,! we! recommend! that! during! the! transition! to ! the! new!

system!(or!in!the!alternative,!should!the!Department!opt!to!continue!within!the!parameters!of!

the!current! approach),! there!be!much! more! careful!organizational!attention! paid ! to!ensuring!

data!quality.!!

A!possible! first! step!toward! this! end!is!the! incorporation! of!traffic! stops,! citations,!and!other!

post-stop!outcomes!into!th e!Department’s!early!intervention!system.!Doing!so!would!seem!to!

obviate!the!need!for!officers!to!submit!a!“journal”!entry!for!each!stop!(though!the!use!of!daily!

activity! journals! may! continue! to! be! relevant! for! other! Department! purposes),! freeing! up!

144

!Engel,!R.S.,!&!Calnon,!J.M.!(2004).!Comparing!benchmark!methodologies!for!police-citizen!contacts:!T raffic!stop!

data!collection!for!the!Pennsylvania!State!Police.!Police$Quarterly,!7,!97 -125;!Ridgeway,!G .!(2006).!Assessing!the!

effect!of!race!bias!in!post-traffic!stop!outcomes!u sing !prop en sity!sco res.!Jou rn al$o f$Qu a ntita tive$C rim in olo gy,!22,!1 -

28.!

145

!Racial!and!Identity!Profiling!Act!of!2015,!Cal.!Assemb.!B.!953!(2015-2016),!Chapter!466!(Cal.!Stat.!2015).!

92!

additional!time!for!other!work.!Further,!it!would!allow!mid-!and!high-level!supervisors!to!track!

individual,!squad,!division,!and!department-wide!trends!in!real!time.!!

Relatedly,! we! recommend!that! the!Department! begin! to! brief! officers!on! the!purpose! of! the!

data! collection! effort! and! include! traffic/pedestrian! stop! and! post-stop! outcomes! as! part! of!

regular! line-up! level! briefings.! Finally,! we! recommend! that! the! Department! work! to! include!

open! format! traffic! and! pedestrian! stop! data! files! (e.g.,! .csv! [comma-separated! values]! files!

rather!than!PDF)!as!part!of!the!City!of!San!Diego!Open!Data!Portal.

146

!Doing!so!would!increase!

the!visibility!of!these!data!and!facilitate!third-party!oversight.

147,148!

! !

146

!San!Diego!Open!Data!Portal.!(n .d.).!Retrieved!Aug.!15,!2016,!from!http://data.sandiego.gov/.!

147

!Ross,!D.!(2015,!May!17).!How !to!jumpstart!the!release!of!open!data!on!policing.!Code$for$America.!Retrieved!

Aug.!15,!2016,!from!https://www.codeforamerica.org/blog/2015/05/17/5-ways-to-jump sta rt -the-release-of-open-

data-on-policing/.!

148

!The!SDSU!research!team!is!investigating!funding!opportunities!to!assist!the!SDPD!in!building!th e!robust!data!

collection!infrastructure!we!recommend.!One!promising!funding!source!is!the!Research!Network!on!Misdemeanor!

Justice!at!John!Jay !Co llege !of!C rim in al!Jus tice.!With!funding!from !th e!L au ra!a nd !Joh n !Arn o ld!Fo u nd a tion ,!the !

Network!is!in!the!process!of!identifying!seven!jurisdictions!in!which!to!bring!together!law!enforcement!agencies!

and!research!institutions!to!build!data!analytic!infrastructure!and!capacity!to!examine!trends!in!various!forms!of!

low-leve l!en fo r ce m e n t!a ct ivity :!misdemea n o r!a rre sts ,!cita tio n s ,!an d !p e d est rian!and!traffic!stops.!See:!

http://johnjay.jjay.cuny.edu/mjp/RN_MJ_Solicitation.pdf.!!

93!

Appendix!1!

Detailed!data!on!SDPD!staffing!and!crime!in!San!Diego!

Table!A1.1.!! !

SDPD!patrol!staffing,!by!division,!watch,!and!year!

1st!Watch!

2nd!Watch!

3rd!Watch!

Total!

2014!

!!!!!Northern!

28!

32!

27!

87!

!!!!!Northeastern!

20!

24!

17!

61!

!!!!!Eastern!

19!

27!

21!

67!

!!!!!Western!

35!

33!

24!

92!

!!!!!Northwestern!

11!

29!

$$Above$Interstate$8$

113$

125$

98$

336$

!!!!!Southeastern!!

25!

39!

23!

87!

!!!!!Central!

36!

34!

30!

100!

!!!!!Southern!

22!

24!

15!

61!

!!!!!Mid-City!

35!

42!

38!

115!

$$Below$Interstate$8$

118$

139$

106$

363$

!!!!!Traffic!

41!

10!

60!

!!Annual!total!

272!

273!

214!

759!

2015!

!!!!!Northern!

36!

39!

26!

101!

!!!!!Northeastern!

20!

21!

16!

57!

!!!!!Eastern!

21!

25!

21!

67!

!!!!!Western!

29!

38!

22!

89!

!!!!!Northwestern!

27!

$$Above$Interstate$8$

115$

132$

94$

341$

!!!!!Southeastern!!

24!

30!

28!

82!

!!!!!Central!

32!

36!

38!

106!

!!!!!Southern!

16!

22!

19!

57!

!!!!!Mid-City!

28!

36!

40!

104!

$$Below$Interstate$8$

100$

124$

125$

349$

!!!!!Traffic$

38!

14!

61!

!!Annual!total!

253!

270!

228!

751!

Source:!San!Diego!Police!Department!

94!

Table!A1.2.!

Crime!in!San!Diego,!CA,!by!crime!type,!location,!and!year!

!!!

Population!

Violent!Crime!(rate)!

Property!crim e!(rate)!

Total!crime!(rate)!

2014!

!!!!!Northern!

225,234!

599!(2.7)!

5,111!(22.7)!

5,710!(25.4)!

!!!!!Northeastern!

234,394!

226!(1.0)!

2,211!(9.4)!

2,437!(10.4)!

!!!!!Eastern!

155,892!

372!(2.4)!

3,486!(22.4)!

3,858!(24.7)!

!!!!!Western!

129,709!

684!(5.3)!

4,055!(31.3)!

4,739!(36.5)!

!!!!!Northwestern!

70,822!

58!(0.8)!

791!(11.2)!

849!(12.0)!

$$Above$Interstate$8$

816,051$

1,939$(2.4)$

15,654$(19.2)$

17,593$(21.6)$

!!!!!Southeastern!!

175,757!

846!(4.8)!

2,408!(13.7)!

3,254!(18.5)!

!!!!!Central!

103,524!

1,099!(10.6)!

3,336!(32.2)!

4,435!(42.8)!

!!!!!Southern!

107,631!

303!(2.8)!

1,905!(17.7)!

2,208!(20.5)!

!!!!!Mid-City!

173,012!

1,023!(5.9)!

3,509!(20.3)!

4,532!(26.2)!

$$Below$Interstate$8$

559,924$

3,271$(5.8)$

11,158$(19.9)$

14,429$(25.8)$

!!Annual!total!

1,375,975!

5,210!(3.8)!

26,812!(19.5)!

32,022!(23.3)!

2015!

!!!!!Northern!

225,234!

626!(2.8)!

5,499!(24.4)!

6,125!(27.2)!

!!!!!Northeastern!

234,394!

267!(1.1)!

2,361!(10.1)!

2,628!(11.2)!

!!!!!Eastern!

155,892!

446!(2.9)!

4,109!(26.4)!

4,555!(29.2)!

!!!!!Western!

129,709!

714!(5.5)!

4,450!(34.3)!

5,164!(39.8)!

!!!!!Northwestern!

70,822!

70!(1.0)!

847!(12.0)!

917!(13.0)!

$$Above$Interstate$8$

816,051$

2,123$(2.6)$

17,266$(21.2)$

19,389$(23.8)$

!!!!!Southeastern!!

175,757!

888!(5.1)!

2,523!(14.4)!

3,411!(19.4)!

!!!!!Central!

103,524!

1,183!(11.4)!

3,549!(34.3)!

4,732!(45.7)!

!!!!!Southern!

107,631!

328!(3.0)!

2,006!(18.6)!

2,334!(21.7)!

!!!!!Mid-City!

173,012!

1,046!(6.0)!

3,813!(22.0)!

4,859!(28.1)!

$$Below$Interstate$8$

559,924$

3,445$(6.2)$

11,891$(21.2)$

15,336$(27.4)$

!!Annual!total!

1,375,975!

5,568!(4.0)!

29,157!(21.2)!

34,725!(25.2)!

Source:!San!Diego!Police!Department!

95!

Appendix!2!!

The!San!Diego!Police!Department!Vehicle!Stop!Data!Card!

! !

96!

Appendix!3!

SDPD!Officer!Survey!

A research team from San Diego State University is gathering the opinions of SDPD officers as a

part of the ongoing review of traffic stop data and police-community relations in the City of San

Diego. As a part of this process, we are asking you to complete the following survey. It should take

no more than 5 or 10 minutes of your time.

As the recent Department Announcement made clear, your input is extremely important. This is why

we ask that you please be as honest as you can and select the response to each question that best

describes your opinion about each topic.

No personally identifiable information will be collected in this survey. Your participation is voluntary

and your responses will be kept confidential. Responses will not be identified by individual, but

rather will be compiled together and analyzed as a group.

If you have any questions or concerns about this survey or your rights as a research subject,

please contact SDSU professor Joshua Chanin at [email protected].

Thank you very much for your time and for the work you do.

SDPD Officer Survey - May 2016

Police-Community Relations in San Diego

SDPD Officer Survey - May 2016

Strongly Agree Agree Disagree Strongly Disagree Not Sure

1. San Diego residents trust the San Diego Police Department.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

2. San Diego residents trust my division of the San Diego Police Department.

97!

Strongly Agree Agree Disagree Strongly Disagree Not Sure

Asian

Black

Hispanic

White

3. The following racial/ethnic groups feel comfortable interacting with the SDPD:

Strongly Agree Agree Disagree Strongly Disagree Not Sure

4. Recent events involving police in cities like Ferguson and Baltimore have made my job as a police officer

more difficult.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

5. The community in my patrol area is appreciative of police presence.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

6. The community in my patrol area is willing to work with the police to solve neighborhood problems.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

7. The Department should do more to reach out to members of the community in my patrol area.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

Asian

Black

Hispanic

White

8. The SDPD treats the following racial/ethnic groups fairly:

98!

Strongly Agree Agree Disagree Strongly Disagree Not Sure

Asian

Black

Hispanic

White

9. The following racial/ethnic groups have confidence in the SDPD:

10. Please use the space below to add any additional thoughts you might have about police-community

relations in San Diego. Is there anything we haven't asked about this topic that you believe should be

addressed?

11. Do you have any suggestions for improving police-community relations in San Diego?

Race, Crime, and Police Patrol

SDPD Officer Survey - May 2016

Strongly Agree Agree Disagree Strongly Disagree Not Sure

Identifying criminal

behavior

Identifying gang-related

activity

Discovering illegal

drugs, guns, or other

contraband

Enforcing traffic laws

12. When you do not have the description of a suspect, a person's race or ethnicity is an important factor

for:

99!

Strongly Agree Agree Disagree Strongly Disagree Not Sure

Asian

Black

Hispanic

White

13. In your experience, the following racial/ethnic groups are more likely to commit crime than members of

other groups:

Strongly Agree Agree Disagree Strongly Disagree Not Sure

Asian

Black

Hispanic

White

14. In your experience, the following racial/ethnic groups are more likely to carry illegal drugs, weapons, or

other contraband than members of other groups:

Strongly Agree Agree Disagree Strongly Disagree Not Sure

Asian

Black

Hispanic

White

15. In your experience, the following racial/ethnic groups are subject to a disproportionate number of police

stops compared to drivers of other racial/ethnic backgrounds:

Strongly Agree Agree Disagree Strongly Disagree Not Sure

16. Racially or ethnically biased policing is justified if it helps keep the community safe.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

17. Conducting a traffic stop is an inherently dangerous activity.

100!

18. Please use the space below to add any additional thoughts you might have about police patrol in San

Diego. Is there anything we haven't asked about this topic that you believe should be addressed?

Traffic Stop Data Cards

SDPD Officer Survey - May 2016

Strongly Agree Agree Disagree Strongly Disagree Not Sure

19. Completing the Traffic Stop Data Card is a worthwhile use of officer time.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

20. Officers who submit incomplete or inaccurate Traffic Stop Data Cards are held accountable.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

21. Sharing traffic stop data (where, when, and of whom stops are made) with the public increases trust in

the police.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

22. Sharing traffic stop data with the public hurts morale among SDPD officers.

23. Please use the space below to add any additional thoughts you might have about the use of Traffic

Stop Data Cards. Is there anything we haven't asked about this topic that you believe should be

addressed?

Officer Training and SDPD Culture

SDPD Officer Survey - May 2016

101!

Strongly Agree Agree Disagree Strongly Disagree Not Sure

24. Officer racial/ethnic bias is a genuine problem for the San Diego Police Department.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

25. SDPD policy is clear on the appropriate use of race/ethnicity in making law enforcement decisions.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

26. Additional training on racial/ethnic bias would make me a more effective officer.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

27. The Department does an effective job identifying officers who are acting in a racially/ethnically biased

manner.

Strongly Agree Agree Disagree Strongly Disagree Not Sure

28. Officers who engage in biased policing are held accountable for their actions.

29. Please use the space below to add any additional thoughts you might have about SDPD policy,

training, or officer culture. Is there anything we haven't asked about this topic that you believe should be

addressed?

Demographics

SDPD Officer Survey - May 2016

102!

30. What is your current rank?

Police Office I or II

Sergeant, Detective, or Lieutenant

Captain or above

Other

31. How long have you been a member of the San Diego Police Department?

1 or fewer years

Between 2 and 5 years

Between 6 and 10 years

Between 11 and 20 years

21 or more years

32. What is the highest level of education you have completed?

High School Graduate

Some College

College Graduate

Post-Graduate Degree

33. What is your age?

24 or Younger

Between 25 and 34

Between 35 and 44

Between 45 and 54

55 or Older

34. What is your race/ethnicity?

Asian

Black

Hispanic

White

Other

103!

! !

35. How many hours per week do you spend enforcing traffic laws?

0-5

6-10

11-15

16-20

21+

36. To which division are you currently assigned?

Central

Eastern

Mid-City

Northern

Northeastern

Northwestern

Southern

Southeastern

Western

Not Applicable

Thank you again for your time. We are seeking volunteers to participate in short, confidential

follow-up interviews on the topics covered in this survey. If interested, please contact Joshua

Chanin at [email protected].

Follow-up Interview

SDPD Officer Survey - May 2016

104!

Appendix!4!

Limiting!the!veil!of!darkness!analysis!to!stops!involving!moving!violations!

The!authors!of!a!recent!paper!analyzing!traffic!stops!in!Syracuse,!New!Y ork!argued!that!“some!

kinds! of! equipment! violations! (e.g.,! malfunctioning! headlights)! are! uniquely! nighttime!

violations,! and! it! is! conceivable! that! the! incidence! of! such! equipment! violations! is! also!

correlated! with! drivers’! race.”

149

! Worden! goes! on! to! argue! that! the! inclusion! of! equipment!

violations! may! bias! the! veil! of! darkness! analysis.! To! account! for! this! possibility,! we!excluded!

equipment!violations!and!re-applied!the!veil!of!darkness!technique!to!a!sub-sample!of!records!

generated! for! stops! involving! only! moving! violations.! Our! findings! are! shown! in! Tables! A4.1!

through!A4.4.!

Table!A4.1.!

Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!citywide!for!a!

moving!violation!!

Odds!

Ratio!

p-value!

Standard!

error!

95%!

Confidence!

Interval!

Number!

of!Stops!

2014!

!!!!!Black!v.!White!

1.165!

0.066!

0.097!

0.990,!1.374!

5,884!

!!!!!Young!Black!v.!Young!White!

1.269!

0.128!

0.198!

0.934,!1.724!

1,544!

2015!

!!!!!Black!v.!White!

0.793!

0.016!

0.076!

0.656,!0.957!

4,381!

!!!!!Young!Black!v.!Young!White!

0.649!

0.019!

0.120!

0.452,!0.932!

1,112!

Combined!

!!!!!Black!v.!White!

0.985!

0.809!

0.062!

0.871,!1.114!

10,265!

!!!!!Young!Black!v.!Young!White!

0.952!

0.676!

0.113!

0.755,!1.120!

2,656!

Table! A4.1! shows! the! results! of! an! analysis! of! citywide! stops! made! during! the! intertwilight!

period!involving!Black!and! White! drivers!stopped!for!a!moving!violation.!Th ese! data! show!no!

statistically!significant!difference!in!the!2014!stop!patterns!of!Blacks!and!Whites.!When!limited!

to! moving! violation! stops! occurring! in! 2015,! our! analysis! shows! that! Black! drivers! were! less!

likely!to!be!stopped!during!daylight!hours!than!after!dark,!compared!to!Whites.!Analysis!of!the!

combined! 2014/2015!data! showed!no! meaningful!disparity! in! the! stop! patterns! of! Black! and!

149

!Worden,!R.E.,!McLean,!S.J.,!&!Wheeler,!A.P.!(2012).!Testing!for!racial!profiling!with!the!veil!of!darkness!method.!

Police$Quarterly,!15,!92-111.!

105!

White!drivers.!

Table!A4.2.!

Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!drivers!will!be!stopped!citywide! for!

a!moving!violation!!

Odds!

Ratio!

p-value!

Standard!

error!

95%!

Confidence!

Interval!

Number!

of!Stops!

2014!

!!!!!Hispanic!v.!White!

1.039!

0.463!

0.054!

0.938,!1.151!

8,619!

!!!!!Young!Hispanic!v.!Young!White!

1.102!

0.382!

0.123!

0.886,!1.372!

1,849!

2015!

!!!!!Hispanic!v.!White!

0.793!

<0.001!

0.047!

0.706,!0.891!

6,681!

!!!!!Young!Hispanic!v.!Young!White!

0.711!

0.005!

0.087!

0.559,!0.904!

1,639!

Combined!

!!!!!Hispanic!v.!White!

0.915!

0.023!

0.036!

0.848,!0.988!

15,300!

!!!!!Young!Hispanic!v.!Young!White!

0.893!

0.165!

0.073!

0.761,!1.048!

3,488!

Table!A4.2!shows!results!of!our!comparative!analysis!of!Hispanic!and!White!drivers!stopped!for!

moving! violations.! We! find! no! statistically! significant! differences! in! the! 2014! data! or! in! the!

combined! 2014/2015! data.! Analysis! of! the! 2015! data! shows! that! Hispanic! drivers! were! less!

likely!to!be!stopped!for!a!moving!violation!during!the!day,!when!driver!race/ethnicity!is!more!

apt!to!be!visible!to!the!naked!eye,!than!were!Whites.!

106!

Table!A4.3.!

Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!for!a!moving!

violation,!above!and!below!Interstate!8!

Odds!

Ratio!

p-value!

Standard!

error!

95%!

Confidence!

Interval!

Number!

of!Stops!

2014!

!!!!!Above!Interstate!8!

1.358!

0.019!

0.177!

1.052,!1.752!

3,771!

!!!!!Below!Interstate!8!

0.773!

0.024!

0.088!

0.618,!0.967!

2,240!

2015!

!!!!!Above!Interstate!8!

1.050!

0.752!

0.162!

0.775,!1.422!

2,983!

!!!!!Below!Interstate!8!

0.597!

<0.001!

0.077!

0.463,!0.770!

1,514!

Combined!

!!!!!Above!Interstate!8!

1.191!

0.077!

0.118!

0.981,!1.446!

6,754!

!!!!!Below!Interstate!8!

0.692!

<0.001!

0.058!

0.586,!0.817!

3,754!

In!Table! A4.3! we!display!the! results! of!our!moving! violation-only! analysis!of!Black! and! White!

drivers!by!stop!location.!We!report!findings!by!year!for!stops!occurring!both!above!and!below!

Interstate!8.!The!data!show!that! i n!2014,!stops!occurring!above!I-8!involving!a!Black!driver!were!

more! likely! to! occur! during! daylight! hours,! when! driver! race/ethnicity! was! visible,! than! after!

dark,!when!it!was!not,!compared!to!Whites.!No!such!disparities!were!evident!in!either!2015!o r!

the!combined!2014/2015!data.!!

Conversely,! records! of! stops! initiated! in! those! divisions! located! below! Interstate! 8! in! 2014,!

2015,!and!2014/2015!combined!show!that!Black!drivers!were!more!likely!to!be!stopped!during!

daylight!hours!than!after!dark!than!were!Whites!stopped!under!similar!conditions.!

! !

107!

Table!A4.4.!

Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!drivers!will!be!stopped!for!a!

moving!violation,!above!and!below!Interstate!8!

Odds!

Ratio!

p-value!

Standard!

error!

95%!

Confidence!

Interval!

Number!

of!Stops!

2014!

!!!!!Above!Interstate!8!

1.089!

0.339!

0.097!

0.914,!1.297!

4,353!

!!!!!Below!Interstate!8!

0.721!

<0.001!

0.055!

0.620,!0.838!

4,485!

2015!

!!!!!Above!Interstate!8!

1.012!

0.909!

0.106!

0.823,!1.243!

3,390!

!!!!!Below!Interstate!8!

0.659!

<0.001!

0.060!

0.552,!0.787!

3,458!

Combined!

!!!!!Above!Interstate!8!

1.044!

0.515!

0.071!

0.915,!1.193!

7,743!

!!!!!Below!Interstate!8!

0.677!

<0.001!

0.039!

0.604,!0.759!

7,943!

Table! A4.4,! which! li sts! find in gs! of! our! location-b ased! analysis! of! moving! violati on! stops !

involving!Hispanic!and!White!drivers,!shows!a!similar!p attern.!We!find!no!statistical!d if ference!

between! Hispanic! and! White! drivers! stopped! for! a! moving! vi olati on ! above! I-8,! regardless! of!

stop!year.!!

These! data! show! evidence! across! stop! year! that! moving! violation! stops! involving! Hispanic!

drivers!were!less!likely!to!occur!during!daylight!hours!than!at!night,!when!compared!to!White!

drivers.!!

! !

108!

Appendix!5!

Limiting!the!veil!of!darkness!analysis!to!stops!involving!male!drivers!

Tables!A5.1!through!A5.4!show!results!of!our!application!of!the!veil!of!darkness!technique!to!a!

sub-sample! of! male! drivers! stopped! for! either! moving! or! equipment-related! violations.! The!

results!are!not!meaningfully!different!from!analysis!of!stops!in volvi ng!male!an d!female!drivers!

compared!under!similar!conditions.!

Table!A5.1.!

Modeling!the!effects!of!daylight!on!the!odds!that!Black!male!drivers!will!be!stopped!citywide!

for!either!a!moving!violation!or!equipment!violation!!

Odds!

Ratio!

p-value!

Standard!

error!

95%!

Confidence!

Interval!

Number!

of!Stops!

2014!

!!!!!Black!v.!White!

1.322!

<0.001!

0.089!

1.159,!1.509!

5,981!

!!!!!Young!Black!v.!Young!White!

1.487!

0.002!

0.193!

1.153,!1.918!

1,569!

2015!

!!!!!Black!v.!White!

0.844!

0.027!

0.064!

0.727,!0.981!

4,616!

!!!!!Young!Black!v.!Young!White!

0.695!

0.010!

0.098!

0.527,!0.917!

1,219!

Combined!

!!!!!Black!v.!White!

1.084!

0.108!

0.054!

0.982,!1.195!

10,597!

!!!!!Young!Black!v.!Young!White!

1.040!

0.675!

0.098!

0.865,!1.252!

2,788!

Table!A5.1!compares!citywide!stop!patterns!of!Black!and !White!male!drivers.!In!2014,!we!fi nd !

that! Black! men! were! more! li kely! to! be! stopped! during! daylight! hours! than! after! dark,! as!

compared!to!White!drivers.!In!2015,!the!exact!opposite!was!true.!Black!male!drivers!were!less!

likely!to!be!stopped!during!daylight!hours!than!they!were!after!dark,!compared!to!White!male!

drivers.!Analysis!of!the!2014/2015!combined!data!show!no!statistically!significant!dif ference!in !

the!stop!patterns!of!Black!and!White!male!drivers.!

109!

Table!A5.2.!

Modeling!the!effects!of!daylight!on!the!odds!that!Black!male!drivers!will!be!stopped!for!either!

a!moving!violation!or!equipment!violation,!above!and!below!Interstate!8!

Odds!

Ratio!

p-value!

Standard!

error!

95%!

Confidence!

Interval!

Number!

of!Stops!

2014!

!!!!!Above!Interstate!8!

1.368!

0.013!

0.172!

1.069,!1.749!

3,224!

!!!!!Below!Interstate!8!

0.998!

0.984!

0.104!

0.813,!1.225!

2,218!

2015!

!!!!!Above!Interstate!8!

1.142!

0.347!

0.162!

0.865,!1.508!

2,650!

!!!!!Below!Interstate!8!

0.645!

<0.001!

0.078!

0.509,!0.816!

1,553!

Combined!

!!!!!Above!Interstate!8!

1.254!

0.015!

0.117!

1.044,!1.506!

5,874!

!!!!!Below!Interstate!8!

0.806!

0.005!

0.063!

0.692,!0.938!

3,771!

In!Table!A5.2,!we!present!the!Black-White!comparative!analysis!by!stop!location.!Stops!of!Black!

male!drivers!initiated!above!I-8!were!more!likely!to!occur!during!daylight!hours!than!after!dark!

in!2014!and!2014/2015!combined,!but!not!2015,!when!compared!to!stops!of!White!men.!!

In!2015!and!2014/2015,!stops!of!Black!men!occurring!below!Interstate!8!were!less!likely!to!

occur!during!daylight!hours!than!after!dark,!compared!to!stops!involving!White!males.!

! !

110!

Table!A5.3.!!

Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!male!drivers!will!be!stopped!

citywide!for!either!a!moving!violation!or!equipment!violation!

Odds!

ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!

of!stops!

2014!

!!!!!Hispanic!v.!White!

1.088!

0.080!

0.053!

0.990,!1.197!

8,723!

!!!!!Young!Hispanic!v.!Young!White!

1.144!

0.173!

0.113!

0.943,!1.389!

2,119!

2015!

!!!!!Hispanic!v.!White!

0.827!

0.001!

0.046!

0.741,!0.923!

6,728!

!!!!!Young!Hispanic!v.!Young!White!

0.737!

0.005!

0.081!

0.595,!0.913!

1,822!

Combined!

!!!!!Hispanic!v.!White!

0.963!

0.297!

0.035!

0.896,!1.034!

15,451!

!!!!!Young!Hispanic!v.!Young!White!

0.928!

0.308!

0.068!

0.805,!1.071!

3,941!

Table!A5.3!displays!the!results!of!analysis!of!stop!patterns!of!Hispanic!and!Whi te!male!drivers,!

aggregated! at! the! city! level.! In! 2015,! Hispanic! males! were! less! likely! to! be! stopped! during!

daylight!than!they!were!after!dark,!compared!to!Whi te!male!drivers.!Analysis!of!the!2014!and!

2014/2015! combined! data! show! no! statistically! significant! difference! in! the! citywide! stop!

patterns!of!Hispanic!and!White!male!drivers.!

! !

111!

Table!A5.4.!

Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!male!drivers!will!be!stopped!for!

either!a!moving!violation!or!equipment!violation,!above!and!below!Interstate!8!

Odds!

Ratio!

p-value!

Standard!

error!

95%!

Confidence!

Interval!

Number!

of!Stops!

2014!

!!!!!Above!Interstate!8!

1.173!

0.078!

0.106!

0.982,!1.340!

3,712!

!!!!!Below!Interstate!8!

0.767!

0.001!

0.062!

0.655,!0.899!

4,292!

2015!

!!!!!Above!Interstate!8!

0.990!

0.920!

0.100!

0.812,!1.207!

3,061!

!!!!!Below!Interstate!8!

0.693!

<0.001!

0.068!

0.572,!0.840!

3,109!

Combined!

!!!!!Above!Interstate!8!

1.087!

0.214!

0.073!

0.953,!1.240!

6,773!

!!!!!Below!Interstate!8!

0.725!

<0.001!

0.045!

0.642,!0.819!

7,401!

Table! A5.4! shows! results! of! our! location-based! analysis! of! Hispanic! and! White! male! drivers!

stopped! for! either! an! equipment! or! moving! violation.! Analysis! of ! the! 2014,! 2015,! and!

2014/2015! combined! data! show! no! statistically! significant! difference! in! the! Above! I-8! stop!

patterns!of!Hispanic!and!White!male!drivers.!

As!was!the!case!with!Black!male!drivers,!stops!below!Interstate!8!involving!Hispanic!men!were!

less!likely!to!be!initiated!during!daylight!than!after!dark!than!were!stops!involving!White!male!

drivers.!

! !

112!

Appendix!6!

Division-level!traffic!stop!patterns,!by!year!

Tables!A6.1!through!A6.6!display!the!results!of!our!analysis!of!traffic!stop!patterns!in!the!nine!

SDPD!police!divisions,!broken!down!by!driver!race/ethnicity!and!stop!year.!!

Table!A6.1.!

Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!for!either!a!

moving!violation!or!an!equipment!violation!in!2014,!by!stop!location!!!

Odds!ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

Above!Interstate!8!

!!!!!Northern!

1.038!

0.878!

0.258!

0.638,!1.691!

1,343!

!!!!!Northeastern!

1.908!

0.002!

0.394!

1.273,!2.861!

1,204!

!!!!!Eastern!

1.018!

0.918!

0.182!

0.718,!1.445!

1,098!

!!!!!Western!

1.410!

0.057!

0.255!

0.989,!2.011!

1,416!

!!!!!Northwestern!

1.151!

0.681!

0.393!

0.590,!2.246!

594!

Sub-total$

1.253$

0.029$

0.129$

1.024,$1.534$

5,226$

Below!Interstate!8!

!!!!!Southeastern!

1.641!

0.030!

0.375!

1.048,!2.568!

740!

!!!!!Central!

0.724!

0.057!

0.123!

0.520,!1.010!

1,306!

!!!!!Southern!

0.952!

0.844!

0.236!

0.586,!1.548!

484!

!!!!!Mid-City!

0.977!

0.869!

0.140!

0.738,!1.292!

1,099!

Sub-total$

0.905$

0.238$

0.077$

0.766,$1.069$

3,402$

In!Table!A6.1,!we!list!the!odds!that!Black!drivers!will!be!stopped!for!a!moving!violation!or!an!

equipment! violation! in! daylight,! compared! to! White! drivers,! using! data! from! 2014.! In! the!

Northeastern! division,! Black! drivers! were! 90.8! percent! more! likely! to! be! stopped! during!

daylight!hours,!when!driver!race/ethnici ty!was!visible,!than!in!darkness!(p!=!0.002),!compared!

to! White! drivers.! Disparities! were! also! evident! in! data! from! the! Southeastern! division! (p! =!

0.030)!and!in!our!analysis!of!aggregate!data!from!the!five!divisions!located!above!Interstate!8!(p!

=! 0.029). We! found! no! statistically! significant! disparities! in! data! from! the! other! seven! patrol!

divisions,!or!in!the!aggregated!data!from!below!Interstate!8.!

113!

Table!A6.2.!

Modeling!the!effects!of!daylight!on!the!odds!that!Black!drivers!will!be!stopped!for!either!a!

moving!violation!or!an!equipment!violation!in!2015,!by!stop!location!!!

Odds!ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

Above!Interstate!8!

!!!!!Northern!

1.327!

0.277!

0.345!

0.797,!2.209!

1,211!

!!!!!Northeastern!

1.072!

0.749!

0.235!

0.699,!1.647!

1,087!

!!!!!Eastern!

1.281!

0.249!

0.275!

0.841,!1.952!

898!

!!!!!Western!

0.817!

0.375!

0.186!

0.522,!1.277!

904!

!!!!!Northwestern!

0.704!

0.403!

0.295!

0.309,!1.602!

392!

Sub-total$

1.067$

0.576$

0.124$

0.849,$1.341$

4,226$

Below!Interstate!8!

!!!!!Southeastern!

1.113!

0.716!

0.328!

0.625,!1.982!

456!

!!!!!Central!

0.650!

0.026!

0.125!

0.445,!0.949!

869!

!!!!!Southern!

1.208!

0.557!

0.389!

0.643,!2.272!

333!

!!!!!Mid-City!

0.978!

0.895!

0.163!

0.705,!1.358!

730!

Sub-total$

0.686$

<0.001$

0.069$

0.564,$0.834$

2,244$

Table! A6.2! reproduces! the! above! analysis! using! data! from! 2015.! We! find! no! statistically!

significant! evidence! of! Black-White! disparity! in! either! the! Northeastern! or! Southeastern!

divisions,! or! the! bel ow! I-8! aggregation.! In! 2015,! stops! in! the! Central! division! involving! Black!

drivers!were!less!l ikely!to!occur!d urin g!daylight!than!after!dark!(p!=!0.026),!compared!to!White!

drivers.!What!is!more,!our!analysis!of!the!aggregated!data!from!the!four!divisions!located!below!

Interstate! 8! revealed! a! similar! pattern:! White! drivers! were! more! likely! to! be! stopped! during!

daylight!hours!than!after!dark!(p!<0.001),!compared!to!Black!drivers.!!!!

We!found!no!statistically!significant!disparities!in!data!from!the!other!eight!patrol !divisions,!or!

in!the!aggregated!data!from!above!Interstate!8.!

114!

Table!A6.3.!

Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!drivers!will!be!stopped!for!either!a!

moving!violation!or!an!equipment!violation!in!2014,!by!stop!location!!!

Odds!ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

Above!Interstate!8!

!!!!!Northern!

0.870!

0.398!

0.143!

0.630,!1.202!

1,494!

!!!!!Northeastern!

1.250!

0.139!

0.188!

0.930,!1.679!

1,361!

!!!!!Eastern!

0.717!

0.026!

0.107!

0.536,!0.961!

1,227!

!!!!!Western!

1.240!

0.080!

0.152!

0.975,!1.576!

1,701!

!!!!!Northwestern!

1.519!

0.064!

0.976,!2.365!

679!

Sub-total$

1.084$

0.262$

0.078$

0.941,$1.249$

6,058$

Below!Interstate!8!

!!!!!Southeastern!

0.960!

0.850!

0.207!

0.629,!1.465!

916!

!!!!!Central!

0.595!

<0.001!

0.072!

0.469,!0.754!

1,718!

!!!!!Southern!

0.999!

0.991!

0.129!

0.775,!1.286!

2,766!

!!!!!Mid-City!

0.950!

0.682!

0.119!

0.743,!1.215!

1,418!

Sub-total$

0.755$

<0.001$

0.049$

0.665,$0.858$

6,382$

Table! A6.3! list! th e! results! of! our! application! of! the! veil! of! darkness! technique! to! stops!

conducted!in!2014!involving!Hispanic!and!White!drivers.!Stops!in!the!Eastern!(p!=!0.026)! and!

Central!(p!<!0.001)!divisions!i nvol ving!Hispanic!drivers!were!less$likely!to!occur!durin g!daylight!

hours!than!in!darkness,!compared!to!White!drivers.!Analysis!of!the!aggregated!data!from!the!

four!divisions!located!below!Interstate!8!produced!similar!ou tcomes:!White!drivers!were!more!

likely! to! be! stopped! during! periods! when! driver! race/ethnicity! was! visible,! compared! to!

Hispanic!drivers!(p!<!0.001).!!!

We!found!no!statistically!significant!disparities!in!data!from!the!other!seven!patrol!divisions,!or!

in!the!aggregated!data!from!above!Interstate!8.!

! !

115!

Table!A6.4.!

Modeling!the!effects!of!daylight!on!the!odds!that!Hispanic!drivers!will!be!stopped!for!either!a!

moving!violation!or!an!equipment!violation!in!2015,!by!stop!location!!!

Odds!ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

Above!Interstate!8!

!!!!!Northern!

1.033!

0.847!

0.177!

0.739,!1.445!

1,368!

!!!!!Northeastern!

1.241!

0.190!

0.204!

0.898,!1.713!

1,193!

!!!!!Eastern!

1.206!

0.284!

0.211!

0.856,!1.701!

1,016!

!!!!!Western!

0.711!

0.037!

0.116!

0.516,!0.979!

1,051!

!!!!!Northwestern!

1.030!

0.909!

0.263!

0.624,!1.698!

521!

Sub-total$

1.044$

0.607$

0.087$

0.887,$1.228$

4,835$

Below!Interstate!8!

!!!!!Southeastern!

1.191!

0.544!

0.343!

0.678,!2.093!

577!

!!!!!Central!

0.499!

<0.001!

0.070!

0.379,!0.657!

1,205!

!!!!!Southern!

0.983!

0.910!

0.149!

0.730,!1.323!

2,212!

!!!!!Mid-City!

0.807!

0.173!

0.127!

0.593,!1.098!

890!

Sub-total$

0.697$

<0.001$

0.055$

0.597,$0.815$

4,574$

Data!from!2015!reveal!similar!patterns.!Stops!conducted!in!the!Western!(p!=!0.037)!and!Central!

divisions! (p! <!0.001)!involving! Hispanic!drivers!were! less! likely!to! occur! during!daylight! hours!

than! after! dark,! compared! to! Whites.! Similarly,! in! the! aggregate,! Hispanics! stops! conducted!

below!I-8!were!less!likely!to!occur!in!daylight!than!after!dark!(p!<!0.001),!compared!to!Whites.!!

We!found!no!statistically!significant!disparities!in!data!from!the!other!seven!patrol!divisions,!or!

in!the!aggregated!data!from!above!Interstate!8.!

! !

116!

Table!A6.5.!

Modeling!the!effects!of!daylight!on!the!odds!that!Asian/Pacific!Islander!drivers!will!be!

stopped!for!either!a!moving!violation!or!an!equipment!violation!in!2014,!by!stop!location!

Odds!ratio!

p-value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

Above!Interstate!8!

!!!!!Northern!

0.722!

0.048!

0.119!

0.523,!0.996!

1,500!

!!!!!Northeastern!

1.274!

0.022!

0.134!

1.036,!1.566!

1,912!

!!!!!Eastern!

1.348!

0.050!

0.205!

1.000,!1.817!

1,216!

!!!!!Western!

1.074!

0.644!

0.168!

0.792,!1.459!

1,483!

!!!!!Northwestern!

0.811!

0.232!

0.142!

0.575,!1.144!

800!

Sub-total$

0.982$

0.784$

0.067$

0.859,$1.121$

6,349$

Below!Interstate!8!

!!!!!Southeastern!

1.110!

0.691!

0.293!

0.662,!1.862!

356!

!!!!!Central!

0.803!

0.202!

0.138!

0.516,!4.028!

1,305!

!!!!!Southern!

1.509!

0.104!

0.382!

0.919,!2.480!

499!

!!!!!Mid-City!

1.300!

0.133!

0.226!

0.923,!1.826!

860!

Sub-total$

1.007$

0.947$

0.104$

0.822,$1.233$

2,860$

Table!A6.5!lists!the! o dd s!that!API!drivers!will!be!stopped!for!a!moving! vi olati on !or!an!equipment!

violation!in!daylight,!compared!to!White!drivers,!using!data!from!2014.!In!the!Northeastern!(p !=!

0.022)! and! Eastern! (p! =! 0.050)! divisions,! API! drivers! were! more! likely! to! be! stopped! during!

daylight! hours,! when! driver! race/ethnicity! was! visible,! than! in! darkness,! compared! to! White!

drivers.!Data! from! the!Northern!division! reveal!th e! inverse:!API!drivers! were! less!likely!to! be!

stopped!during!daylight!hours!than!after!dark,!compared!to!Whites.!!

Statistically! significant! disparities! were! not! present! in! the! other! six! patrol! divisions,! or! in! the!

aggregated!data!from!above!and!below!Interstate!8.!

117!

Table!A6.6.!

Modeling!the!effects!of!daylight!on!the!odds!that!Asian/Pacific!Islander!drivers!will!be!

stopped!for!either!a!moving!violation!or!an!equipment!violation!in!2015,!by!stop!location!

Odds!

ratio!

p-value!

Standard!

error!

95%!confidence!

interva l!

Number!

of!stops!

Above!Interstate!8!

!!!!!Northern!

1.332!

0.095!

0.229!

0.951,!1.866!

1,368!

!!!!!Northeastern!

0.982!

0.869!

0.110!

0.787,!1.224!

1,682!

!!!!!Eastern!

1.065!

0.698!

0.172!

0.776,!1.460!

1,046!

!!!!!Western!

0.717!

0.111!

0.150!

0.476,!1.079!

937!

!!!!!Northwestern!

0.863!

0.430!

0.161!

0.599,!1.244!

662!

Sub-total$

0.905$

0.176$

0.066$

0.783,$1.046$

5,254$

Below!Interstate!8!

!!!!!Southeastern!

1.382!

0.391!

0.521!

0.660,!2.900!

166!

!!!!!Central!

1.468!

0.028!

0.256!

1.043,!2.067!

962!

!!!!!Southern!

1.388!

0.274!

0.416!

0.772,!2.498!

344!

!!!!!Mid-City!

0.846!

0.450!

0.187!

0.548,!1.305!

499!

Sub-total$

1.023$

0.849$

0.122$

0.809,$1.294$

1,839$

As! is! shown! in! Table! A6.6,! using! data! from! 2015,! we! find! evidence! showing! that! stops!

conducted!in!the!Central!division!involving!API!drivers!were!more!46.8!percent!likely!to!occur!

during!daylight!hours!than!after!dark!(p!=!0.028)!compared!to!White!driver!stops.!Statistical ly!

significant! disparities! were! not! present! in! any! of! the! other! eight! patrol! divisions,! or! in! the!

aggregated!data!from!above!and!below!Interstate!8.!

118!

Appendix!7!

Using!logistic!regression!to!model!post-stop!outcomes!

What!follows! are! the!results!of! our! analysis!of!post-stop! outcomes! using!multivariate!logistic!

regression.!This!technique!is!valuable!in!that!in!al lows!researchers!to!examine!the!relationship!

between! a! dichotomous! variable,! like! search/no! search,! and! several! other! variables.! The!

propensity! score! matching! technique! is! more! effective! at! isolating! the! effects! of! driver!

race/ethnicity!and ! thus!has!stronger!internal! validity!than!do!logistic!regression!models.!Logit!

models! allow! for! use! of! a! larger! su b-sampl e! of! the! traffic! stop! population! and! th us! have! a!

higher!degree!of!external!validity!than!do!the!results!of!the!matched!pairs!analysis.!!

Table!A7.1.!

Using!logistic!regression!to!model!the!likelihood!that!SDPD!officers!will!search!Black!drivers!!

Odds!ratio!

p-Value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

All!searches!

2.98!

<0.001!

0.091!

2.81,!3.17!

122,547!

!!!!Consent!!

3.63!

<0.001!

0.269!

3.14,!4.20!

116,745!

!!!!Fourth!waiver!!

4.48!

<0.001!

0.254!

4.01,!5.01!

116,745!

!!!!Inventory !!

1.99!

<0.001!

0.121!

1.77,!2.24!

116,745!

!!!!Incident!to !a rre s t!

1.38!

<0.001!

0.122!

1.17,!1.64!

116,745!

!!!!Other!(uncategorized)!

2.57!

<0.001!

0.171!

2.26,!2.93!

121,704!

The!results!shown!in!Table!A7.1!show!clearly!that!Black!drivers!are!more!likely!to!be!searched!

than! are! White! drivers! follo wing! discretionary! traffic! stops,! regardless! of! search! type.! Table!

A7.2!shows!similar!results!when!the!dataset!is!limited!to!Hispanic!and!White!drivers.!Hispanics!

drivers!were!more!likely!to!be!searched!than!are!White!drivers.!!

! !

119!

Table!A7.2.!

Using!logistic!regression!to!model!the!likelihood!that!SDPD!officers!will!search!Hispanic!

drivers!!

Odds!ratio!

p-Value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

All!searches!

1.93!

<0.001!

0.052!

1.83,!2.04!

163,897!

!!!!Consent!!

2.02!

<0.001!

0.140!

1.76,!2.31!

156,689!

!!!!Fourth!waiver!!

1.45!

<0.001!

0.086!

1.29,!1.63!

156,689!

!!!!Inventory !!

2.56!

<0.001!

0.118!

2.34,!2.81!

156,689!

!!!!Incident!to !a rre s t!

1.20!

0.008!

0.084!

1.05,!1.38!

156,689!

!!!!Other!(uncategorized)!

1.64!

<0.001!

0.097!

1.47,!1.85!

162,708!

Tables!A7.3!lists!the!results!of!four!logistic!regression!models!designed!to!estimate!the!effects!

of! race/ethnicity! on! the! discovery! of! contraband,! as! well! as! the! decision! to! issue! a! citation,!

initiate!a!field!interview,!and!make!an! arrest! following! the! discretionary! traffic! stops!of! Black!

and!White!drivers.!The!findings!are!in!line!with!the!results!of!our!matched!pairs!analysis:!Black!

drivers!were!less!likely!to!be!cited!than!Whites,!and!Blacks! were!also!less!likely!to!be!found!with!

contraband.! According! to! this! analysis,! Black! drivers! faced! a! greater! likelihood ! of! being!

subjected! to! a! field! interview! and! are! substantiall y! more! likely! to! be! arrested! compared! to!

White!drivers.!!!

Table!A7.3.!

Using!logistic!regression!to!model!post-stop!outcomes!for!Black!drivers!!

Odds!ratio!

p-Value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

Citation!

0.59!

<0.001!

0.009!

0.57,!0.60!

123,082!

Field!interview!

5.32!

<0.001!

0.204!

4.93,!5.73!

123,082!

Contraband*!

0.68!

<0.001!

0.071!

0.55,!0.83!

122,547!

Arrest!

1.37!

<0.001!

0.081!

1.22,!1.54!

123,082!

*!Includes!statistical!controls!for!police!search!

Table!A7.4!lists! th e!results!of!four!logistic!regression!models!evaluating!the!post-stop!outcomes!

of!Hispanic!and!White!drivers.!These!findings!reflect!the!results!of!our!matched!pairs!analysis.!

Hispanic! drivers! were! less! likely! than! White! drivers! to! be! found! with! contraband! following! a!

120!

search! and! were! more! likely! to! be! the! subject! of! a! field! interview.! We! found! no! statistical!

difference!in!either!the!arrest!or!citation!rates!of!Hispanic!and!White!drivers.!!

Table!A7.4.!!

Using!logistic!regression!to!model!post-stop!outcomes!for!Hispanic!drivers!!

Odds!ratio!

p-Value!

Standard!

error!

95%!

confidence!

interva l!

Number!of!

stops!

Citation!

0.99!

0.320!

0.011!

0.97,!1.01!

164,635!

Field!interview!

1.94!

<0.001!

0.075!

1.80,!2.09!

164,635!

Contraband*!

0.58!

<0.001!

0.054!

0.48,!0.70!

163,897!

Arrest!

1.17!

0.081!

0.103!

0.98,!1.39!

164,635!

*!Includes!statistical!controls!for!police!search!

In! each! case,! the! results! generated! by! our! multiple! logistic! regression! models! are! consistent!

with!the!findings!produced!by!the!propensity!score!matching!analysis!described! in!Chapter!5.!

Taken! together,! these! two! sets! of! results! suggest! that! across! most! post-stop ! outcomes,!

including! search,! contraband! discovery,! and! field! interviews,! Black! and ! Hispanic! drivers! are!

subject!to!disparate!levels!of!scrutiny.!!!

! !

121!

Appendix!8!!

Describing!matched!and!unmatched!drivers!

Table! A8.1! lists! by! race/ethnicity! the! outcome! of! this! matching! process! for! Black! and! White!

drivers! across! eight! stop!characteristics! upon! which! the! match! was! based.! These! include!the!

reason!for!and!location!(police!district)!of!the!stop,!the!day!of!the!week,!month,!and!time!of!day!

during!which!the!stop!occurred,!and!the!driver’s!age,!gender,!and!residency!status.!!

The!Matched!Black! Drivers!column!lists!by!percentage!the!distribution!of!19,948!stops!involving!

matched!Black!drivers:!66.0!percent!were!stopped!for!moving!violations,!9.0!were!stopped!in!

the!Northern!patrol! division,! 10.1!percent! were!stopped! between!noon! and!3:00! PM,!and! so!

on.!The!Matched!White!Drivers!column!lists!similar!information!fo r!the!19,948!matched!White!

drivers.! The! Unmatched! Black! Drivers! column! describes! the! 4,150! Black! drivers! for! which! a!

suitable!match! co ul d!not!be!found.!The!rightmost!column,!Unmatched!White!Drivers,!describes!

the!74,017!White!drivers!that!we!could!not!appropriately!match.!Table!A8.2!lists!the!same!data!

for!Hispanic!drivers!and!their!matched!(and!unmatched)!White!counterparts.!

Table!A8.1.!!

Describing!matched!and!unmatched!Black!and!White!drivers!!

Matched!Black!

drivers!

(n=19,948 )!

Matched!White!

drivers!

(n=19,948 )!

Unmatched!

Black!drivers!

(n=4,088)!

Unmatched!

White!drivers!

(n=73,979 )!

Reason!for!stop!

Moving!violation!

66.0!!

64.6!!

31.3!!

80.6!

Equipment!violation!

32.3!

33.4!

66.2!

18.2!

Code!violation !

0.7!!!

0.7!

1.1!

0.4!

Radio!call/citizen!contact!

0.6!!

0.7!!

0.5!!

Observation/knowledge!

0.3!!

0.5!

0.1!!

Suspect!information!

0.2!!

0.3!

0.5!!

0.1!!

Other!

<0.1!

0.1!

0.0!

0.1!

122!

Table!A8.1.!Describing!matched!and!unmatched!Black!and!White!drivers,!cont.!

Stop!location!

Northwestern!

3.1!!

3.5!!

0.0!!

9.4!!

Northern!

9.0!

9.1!

0.0!

25.2!

Northeastern!

9.2!!

0.0!!

15.7!

Eastern!

14.2!!

0.0!!

15.2!!

Southeastern!

8.4!!

7.8!!

82.5!!

0.0!!

Central!

17.1!!

17.4!

0.4!

9.0!!

Western!

11.4!!

10.8!

0.0!

19.0!

Southern!

!4.7!

!5.3!

0.1!

2.7!

Mid-City!

22.5!

22.7!!

17.0!

3.8!!

Stop!time!

12:00–3:00!a.m.!

13.3!!

13.0!

14.6!

8.0!

3:00–6:00!a.m.!

3.7!!

4.0!!

4.1!

1.9!

6:00–9:00!a.m.!

11.7!!

11.1!

8.6!

13.7!

9:00!a.m.–12:00!p.m.!

17.4!!

17.0!

12.4!

23.7!

12:00–3:00!p.m.!

10.1!

10.3!

4.6!

15.5!

3:00–6:00!p.m.!

15.5!

16.2!!

24.8!

15.4!!

6:00–9:00!p.m.!

10.7!!

11.5!

14.7!

9.4!

9:00!p.m.–12:00!a.m.!

17.6!!

17.1!

16.3!!

12.3!

Stop!day!

Monday!

12.4!!

13.0!

15.4!

12.2!

Tuesday!

16.9!!

16.5!

12.6!

19.2!

Wednesday!

15.6!!

15.8!!

11.6!!

19.5!!

Thursday!

16.0!

15.7!

14.5!

17.6!

Friday!

15.1!!

14.6!!

16.5!!

13.3!!

Saturday!

13.5!

13.6!

15.1!

10.3!

Sunday!

10.4!

10.8!

14.4!

8.0!!

123!

Table!A8.1.!Describing!matched!and!unmatched!Black!and!White!drivers,!cont.!

Stop!month!

January!!

8.9!

9.4!

10.5!

8.7!

February!

10.5!

11.6!

10.0!

March!

9.4!

9.6!

8.1!

9.0!

April!

9.6!

9.4!

9.3!

10.0!

May!!

8.6!

8.8!

7.4!

8.9!

June!!

7.8!

8.1!

8.3!

July!

7.5!

8.5!

August!!

8.9!

8.6!

9.5!

7.9!

September!

7.5!

6.8!

6.9!

October!

6.9!

6.7!

7.2!

7.3!

November!

7.6!

6.3!

7.8!

December!

6.7!

7.0!

6.9!

6.8!

Driver!age!

Under!18!

0.5!!

0.7!!

0.3!!

1.5!

18-25!

24.5!

29.2!

18.6!

26-35!

32.4!!

31.3!

30.7!

26.2!

36-45!

17.9!!

18.3!

17.1!

18.0!

46!and!over!

24.7!!

24.3!!

19.9!

34.3!!

Driver!gender!

Male!

70.0!!

69.6!

77.8!

59.5!!

Female!

30.0!!

30.4!

22.2!

40.5!!

Driver!residency!status!

Resident!

77.7!!

77.6!

90.1!

73.3!

Non-resident!

22.3!!

22.4!

9.9!

36.7!

! !

124!

Table!A8.2.!!

Describing!matched!and!unmatched!Hispanic!and!White!drivers!!

Matched!Hispanic!

drivers!

(n=39,252 )!

Matched!

White!drivers!

(n=39,252 )!

Unmatched!

Hispanic!

drivers!

(n=24,928 )!

Unmatched!

White!drivers!!

(n=54,675 )!

Reason!for!stop!

Moving!violation!

69.5!

71.1!

61.3!

82.1!

Equipment!violation!

29.0!

27.7!

37.6!

16.5!

Code!violation !

0.4!

0.3!

0.6!

Radio!call/citizen!contact!

0.6!

0.4!

0.3!

0.6!

Observation/knowledge!

0.2!

0.1!

Suspect!information!

0.2!

<0.1!

Other!

0.1!

<0.1!

0.1!

Stop!location!

Northwestern!

6.2!

5.5!

0.0!

10.0!

Northern!

12.4!

12.7!

0.0!

28.3!

Northeastern!

10.3!

9.9!

0.0!

17.5!

Eastern!

13.4!

13.9!

<0.1!

15.9!

Southeastern!

4.5!

4.2!

22.0!

0.0!

Central!

17.7!

17.0!

3.2!

6.2!

Western!

13.6!

13.5!

0.0!

20.1!

Southern!

7.0!

7.6!

64.5!

0.0!

Mid-City!

15.0!

15.8!

10.3!

2.0!

Stop!time!

12:00–3:00!a.m.!

10.8!

10.4!

8.3!

3:00–6:00!a.m.!

3.5!

3.2!

3.0!

1.6!

6:00–9:00!a.m.!

13.8!

13.4!

13.0!

13.3!

9:00!a.m.–12:00!p.m.!

19.3!

20.7!

19.1!

23.6!

12:00–3:00!p.m.!

11.6!

11.8!

10.5!

16.2!

3:00–6:00!p.m.!

15.1!

15.5!

23.0!

15.4!

125!

Table!A8.2.!Describing!matched!and!unmatched!Hispanic!and!White!drivers,!cont.!

6:00–9:00!p.m.!

10.6!

10.0!

11.8!

9.3!

9:00!p.m.–12:00!a.m.!

15.3!

14.9!

11.4!

12.4!

Stop!day!

Monday!

12.7!!

12.4!

13.9!

12.1!

Tuesday!

17.5!!

18.0!

15.6!!

19.2!!

Wednesday!

17.3!!

17.6!

15.0!

19.5!

Thursday!

16.4!!

16.7!

15.2!

17.7!

Friday!

14.4!

14.3!!

16.2!

13.2!

Saturday!

12.2!!

12.0!!

12.9!

10.2!

Sunday!

9.5!

9.1!

11.3!

8.1!

Stop!month!

January!!

8.8!

8.5!

8.6!

8.9!

February!

10.2!

10.5!

10.3!

9.8!

March!

9.2!

9.1!

9.4!

9.0!

April!

9.8!

9.1!

10.1!

May!!

8.9!

8.7!

8.4!

8.9!

June!!

8.2!

7.9!

8.5!

8.3!

July!

7.6!

7.8!

9.0!

8.6!

August!!

8.2!

8.3!

8.6!

8.0!

September!

7.1!

7.3!

7.0!

October!

7.4!

7.3!

7.1!

November!

7.8!

8.0!

6.9!

7.6!

December!

6.9!

7.1!

6.8!

6.6!

126!

Table!A8.2.!Describing!matched!and!unmatched!Hispanic!and!White!drivers,!cont.!

Driver!Age!

Under!18!

0.9!

0.6!

0.5!

1.9!

18-25!

24.9!

25.0!

29.8!

16.1!

26-35!

30.4!

30.7!

27.5!

25.2!

36-45!

20.5!

20.0!

19.9!

16.5!

46!and!under!

23.4!

23.7!

22.2!

40.3!

Driver!gender!

Male!

66.4!

67.3!

68.2!

57.7!

Female!

33.6!

32.7!

31.9!

42.3!

Driver!residency!status!

Resident!

70.8!

70.7!

69.0!

76.8!

Non-resident!

29.2!

29.3!

31.0!

23.2!

! !

127!

Appendix!9!

Modeling!driver!hit!rates!after!dropping!missing!contraband!cases!!

As! we! note! in! Chapter! 3,! 93! percent! of! stops! recorded! in! 2014! and! 2015! were! missing!

information! abo ut! the! discovery! of! contraband.! In! the! analysis! discussed! in! Chapter! 5,! we!

interpreted! these! missing! data! to ! mean! that! no! contraband! was! found.! To! account! for! the!

possibility!that!this!assumption!affected!the!accuracy!of!our!analysis,!we!dropped!the!missing!

data! and! re-matched!Black! and! Hispanic! drivers! with!White! drivers.! Though!the! sample!sizes!

were!significantly!smaller,!the!results!are!consistent!with!the!previous!‘hit!rate’!findings,!as!is!

shown!in!Tables!A9.1!and!A9.2.!

Table!A9.1.!

Comparing!hit!rates!among!matched!Black!and!White!drivers!after!dropping!missing!and!null!

cases!!!!

Matched!Black!

drivers!(%)!

Matched!White!

drivers!(%)!

Difference!

(%)!

p-Value!

All!searches!

10.7!

17.9!

-50.71!!

<0.001!!

Consent!!

9.9!

19.7!

-66.25!

<0.001!

Fourth!waiver!!

6.9!

22.6!

-106.06!

<0.001!

Inventory !!

19.8!

18.6!

6.17!

0.024!

Incident!to !a rre s t!

4.1!

9.0!

-74.52!

0.810!

Other!(uncategorized)!

25.5!

39.7!

!-43.55!

0.055!!

Note:!The!analysis!is!based!on!a!total!of!1,998!Black!drivers!and!1,998!matched!White!drivers.!Missing!and!null!cases!dropped.!!

Table!A9.2.!

Comparing!hit!rates!among!matched!Hispanic!and!White!drivers!after!dropping!missing!and!

null!cases!!!!

Matched!Hispanic!

drivers!(%)!

Matched!White!

drivers!(%)!

Difference!(%)!

p-Value!

All!searches!

9.8!

17.1!

54.36!

<0.001!

Consent!!

9.6!

22.2!

79.43!

<0.001!

Fourth!waiver!!

13.6!

16.9!

22.20!

0.258!

Inventory !!

3.9!

5.5!

33.80!

0.222!

Incident!to !a rre s t!

11.0!

18.5!

51.01!

0.021!

Other!(uncategorized)!

35.2!

46.1!

26.77!

0.097!

Note:!The!analysis!is!based!on!a!total!of!3,038!Hispanic!drivers!and!3,038!matched!White!drivers.!Missing!and!null!cases!

dropped.!! !

128!

Appendix!10!

Modeling!driver!hit!rates!after!dropping!missing!contraband!cases!

The!analysis!of!citation!rates!discussed!in!Chapter!5!was!based!on!the!assumption!that!missing!

and! null! cases! indicated! th at! no! citation! was! issued.! To! address! the! possibility! that! these!

findings! were! skewed! by! the! incorporation! of! ambiguous! data,! we! re-matched! drivers! after!

dropping!from!the! sample!stop!records!that!includ ed! either!missing!or!null!citation!data.! The!

results!are!shown!in!Table! A10.1! and! A10.2.! The!results!were!substantively!unchanged:!Black!

drivers!remain!less! likely!to!receive!a!citation!than! White!drivers,!while!Hispanics!and!Whites!

are!ticketed!at!nearly!identical!rates.!

Table!A10.1.!!

Comparing!citation!rates!for!matched!Black!and!White!drivers!after!dropping!missing!

contraband!cases!

Matched!

Black!drivers!

(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-Value!

Matched!

pairs!

Searched!drivers!includ ed!

54.6!

60.4!

-!5.1!

<0.001!

19,103!

Searched!drivers!excluded!

54.4!

60.5!

-!6.1!

<0.001!

18,504!

!Note:!Missing!and!null!cases!dropped.!

Table!A10.2.!!

Comparing!citation!rates!for!matched!Hispanic!and!White!drivers!after!dropping!missing!

contraband!cases!

Matched!

Hispanic!

drivers!(%)!

Matched!

White!

drivers!(%)!

Difference!

(%)!

p-Value!

Matched!

pairs!

Searched!drivers!includ ed!

63.7!

62.7!

0.9$

0.003!

38,059!

Searched!drivers!excluded!

63.7!

62.9!

0.8$

0.011!

37,203!

Note:!Missing!and!null!cases!dropped.!

! !

129!

Appendix!11!

SDPD!officer!training!!

On!November!4,!2016,!we!received!the!following!statement!from!the!San!Diego!Police!

Department!regarding!their!current!officer!training!requirements:!

SDPD$is$a$recognized$leader$in$officer$training.$$The$concepts$of$de-escalation,$non-

biased$policing,$community$policing$and$diversity$are$embedded$in$all$training$at$the$

academy,$and$all$sworn$ranks$receive$ongoing$training$in$these$areas.$$The$following$

highlights$specific$training$courses$offered$in$the$past$few$years.$

• Academy$Training$for$New$Recruits:$

§ People$with$Disabilities$&$Mental$Illness—15$hours$

§ Policing$in$the$Community—24$hours$(POST$only$requires$18$hours)$

Includes$Community$Policing,$Media$Sensitivity,$Community$Mobilization,$

Community$Partnerships,$Resource$Development,$Crime$Prevention,$etc.$$

§ Cultural$Diversity/Discrimination—46$hours$(POST$only$requires$16$hours)$

Includes$EEO,$Cultural$Diversity,$Racial$Profiling,$Spanish,$LGBT,$Hate$crimes$

§ Victimology$and$Victim$Assistance—6$hours$

• New$Officer$Phase$Training$after$Academy—increased$by$5$weeks$in$2015:$

§ Agency-Specific$Training—immediately$follows$academy$graduation$

Includes$family$wellness$day$(added$in$2012)$and$one-day$bus$tour$(added$in$

spring$2015)$

§ Observation/Community$Engagement$Phase—one$month,$provided$prior$to$field$

training$phases$(added$in$summer$2015)$

§ Crisis$Response$Team$Training$(CRT)—40$hours,$provided$to$all$new$officers$

(added$in$2015)$

Includes$de-escalation,$dealing$with$the$mentally$ill,$slowing$down$responses,$

awaiting$adequate$cover,$and$supervisory$oversight$

§ Emotional$Intelligence/Effective$Interactions—16$hours,$after$completion$of$

fourth$field$training$phase,$just$prior$to$being$released$on$their$own$(added$fall$

2015)$

• Advanced$Officer$Training$(AOT)$required$for$all$officers$and$sergeants$every$two$

years—40$hours$

§ 2015-2016$agenda$includes$the$following$topics:$

• Non$Biased$Based$Policing—3.5$hours$

• Tactical$Communication—2$hours$

• Defensive$Tactics/Use$of$Force$(including$de-escalation)—4.5$hours$

• Civil$Liabilities—2$hours$

130!

• Wellness$(including$emotional$intelligence)—2$hours$

§ 2017-2018$planned$agenda$includes$the$following$topics:$

• Non$Biased$Based$Policing—3$hours$

• Tactical$Communication—2$hours$

• Defensive$Tactics/Use$of$Force$(including$de-escalation)—5$hours$

• Emotional$Intelligence—5$hours$

• Command$Training$required$for$all$sergeants,$lieutenants$and$captains—40$hours$

(added$in$summer$2015)$

§ 2015$agenda$included$the$following$topics:$

• PERF$Report$and$Recommendation$Implementation$Plan—1.5$hours$

• Emotional$Intelligence$Model—2$hours$

• Procedural$Justice$Model—2$hours$

• Tactical$De-escalation—1$hour$

• Crucial$Conversations/$Practical$Application$of$Emotional$Intelligence—2$

hours$

• Employee$Wellness/Self$Care—1$hour$

• Mitigating$Liabilities—2$hours$

• Captain’s$Discussion—3$hours$

• Non-Bias$Based$Policing—1.5$hours$

• Body$Worn$Camera$Panel$(how$to$enhance$accountability,$transparency$and$

reduce$liability)—2$hours$

• Leadership—4$hours$

§ 2016$agenda$included$the$following$topics:$

• Leadership—2$hours$

• Critical$Incident$Debrief$(lessons$learned)—2$hours$

• Demonstration$Management—1$hour$

• Tactical$Scenario$Training—4$hours$

• Fall$2015$Field$Training$Officer$Refresher—all$Field$Training$Officers,$included$the$

following:$

§ Procedural$Justice!

§ Emotional$Intelligence$