RACIAL DISPARITIES IN FLORIDA
SAFETY BELT LAW ENFORCEMENT

A REPORT BY
THE AMERICAN CIVIL LIBERTIES UNION RACIAL JUSTICE PROGRAM
AND
THE AMERICAN CIVIL LIBERTIES UNION OF FLORIDA

JANUARY 2016

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

TABLE OF CONTENTS
SUMMARY ............................................................................................................................ 1
I. INTRODUCTION ................................................................................................................. 5
II. FINDINGS ......................................................................................................................... 8
Finding #1: In 2014, Black People Were Stopped and Issued Seatbelt Tickets by
Law Enforcement Across Florida Far Out of Proportion to Their Estimated
Representation Among Drivers ............................................................................. 8
Finding #2: In 2014, Law Enforcement Officers Across Florida Stopped and
Ticketed Black People for Seatbelt Violations Nearly Twice as Often as
White People ......................................................................................................... 9
Finding #3: Racial Disparities in Florida Safety Belt Law Enforcement are
Persistent............................................................................................................ 10
Finding #4: Any Differences in Seatbelt Use by White and Black People Do Not
Explain Why Florida Law Enforcement Officers Stopped and Ticketed
Black People for Seatbelt Violations at Nearly Twice the Rate of White
People in 2014 ..................................................................................................... 11
Finding #5: Certain Law Enforcement Agencies’ Enforcement of the Florida
Safety Belt Law Has Resulted in Racial Disparities That Exceed Even
the Statewide Average Racial Disparity .............................................................. 13
Finding #6: Law Enforcement Agencies’ Failure to Report Seatbelt Citation
Data to State Authorities as Required by Law Underscores Concern
About Racial Disparities in Safety Belt Law Enforcement ................................... 14
III. RECOMMENDATIONS .................................................................................................... 17
Recommendation #1: Investigation ................................................................................ 19
Recommendation #2: Enforcement ................................................................................ 19
Recommendation #3: Transparency............................................................................... 20
Recommendation #4: Supervision .................................................................................. 21
IV. METHODOLOGY, DATA, & SOURCES .............................................................................. 22
Description Versus Causal Analysis ............................................................................... 22
Methodology ................................................................................................................... 23
Data and Sources ............................................................................................................ 24
Survey Weighting and Error Estimation in 1-Year ACS/PUMS Data ............................... 26
Lack of Adequate Data on Latino Drivers ........................................................................ 29
V. CALCULATIONS .............................................................................................................. 31

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

SUMMARY
To help inform public debate about the prevalence of racial profiling, since 2005, the Florida
Safety Belt Law has required that every Florida law enforcement agency annually report the
race and ethnicity of seatbelt citation recipients to the Governor, President of the Senate, and
Speaker of the House of Representatives. See Fla. Stat. Ann. § 316.614(9). This requirement is
carried out through annual reports to the Florida Department of Highway Safety and Motor
Vehicles (FDHSMV).
American Civil Liberties Union (ACLU) analysis of the most recent seatbelt citation data
confirms that the Florida Safety Belt Law has been applied more often to Black motorists than
white motorists. Across Florida, and in specific counties, Blacks are stopped and issued
seatbelt citations far out of proportion to their estimated representation among Florida
drivers, despite the fact that Black and white people in Florida use seatbelts at closely
comparable rates. These findings suggest that biased policing impacts seatbelt enforcement.

In 2014, Florida law
enforcement officers stopped
and ticketed Black motorists
for seatbelt violations 1.9
times more often than white
motorists.
This racial disparity is
persistent.
In 2011, Black motorists were
stopped and ticketed 2.1
times more often than white
motorists.
In 2014, Black people made up only 13.5% of the estimated Florida resident driver population,
but made up 21.96% of recipients of all seatbelt citations reported to state authorities.

1

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

If Black people had been stopped and ticketed for seatbelt violations in
proportion to their estimated representation among Florida drivers, they
would have received 20,296 fewer seatbelt citations in 2014.

Certain agencies’ enforcement of the Florida Safety Belt Law resulted in
disparities that met or exceeded the statewide racial disparity.
Black motorists were stopped and cited for seatbelt offenses:
4 times more often than white motorists by the Escambia County Sheriff’s Office in 2011.*
3 times more often than white motorists by the Palm Beach County Sheriff’s Office in 2014.
2.8 times more often than white motorists by the Orange County Sheriff’s Office in 2014.
1.9 times more often than white motorists by the Broward County Sheriff’s Office in 2014.
* 2011 is the most recent year that this agency reported seatbelt citation data to the FDHSMV.

These racial disparities raise serious concerns that law enforcement engage in
racial profiling when enforcing the Florida Safety Belt Law.

2

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

Differences in seatbeltwearing behavior
between Black and white
people in Florida, as
documented by statewide
and national studies, DO
NOT explain the racial
disparities in seatbelt
citation rates identified in
this report.

A number of law enforcement agencies fail to collect and report seatbelt
citation data to state officials as required by the Florida Safety Belt Law,
impeding the law’s goal of helping to identify agencies that may engage in
racial profiling.

3

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

RECOMMENDATIONS
1. INVESTIGATION
County Commissions charged with oversight responsibility and the Florida Attorney
General’s Office of Civil Rights must investigate law enforcement agencies whose
implementation of the Florida Safety Belt Law has led to large racial disparities, including the
Sheriff’s Offices for Escambia, Palm Beach, Orange, and Broward Counties to determine the
cause and what measures should be taken to promote fair and impartial policing.

2. ENFORCEMENT
Not all law enforcement agencies comply with the Florida Safety Belt Law’s clear requirement
to report annually to state officials the race and ethnicity of people stopped and cited for
seatbelt offenses. The Florida Legislature should enact legislation to create consequences
for law enforcement agencies that fail to comply with the statute’s reporting requirement,
including the City of Miami Police Department, the City of Tampa Police Department, the
Jacksonville Sheriff’s Office, and the Escambia County Sheriff’s Office.

3. TRANSPARENCY
Law enforcement agencies should collect, make available to the public, and analyze data on
the race and ethnicity of civilians and the basis and outcome of encounters for all traffic and
pedestrian stops, frisks, searches, citations, summons, and arrests. This transparency
measure will help identify and help eliminate racial profiling in all police-civilian encounters—
not just those involving Florida Safety Belt Law enforcement.

4. SUPERVISION
Law enforcement agencies should analyze the seatbelt citation data they collect and report to
the Florida Department of Highway Safety and Motor Vehicles and ensure that supervising
officers monitor racial disparities in their supervisees’ enforcement of the Florida Safety Belt
Law. They should also provide testing on implicit bias and training on bias-free policing to all
officers, and retrain officers whose stop and citation practices result in large racial disparities.

4

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

I. INTRODUCTION
This report analyzes data on Florida Safety Belt Law enforcement, which is collected by law
enforcement agencies across the state and provided to the Florida Department of Highway
Safety and Motor Vehicles for publication in an annual report. 1 It identifies sizable and
statistically significant disparities in the rates at which white and Black motorists across the
state are stopped and issued seatbelt citations. 2 It also identifies specific law enforcement
agencies whose enforcement of the Safety Belt Law has created racial disparities that exceed
the average statewide racial disparity, as well as agencies that violate the law’s data collection
and reporting requirement.
These findings contribute to concerns about racial profiling—the targeting of people for
humiliating and often frightening detentions, interrogations, and searches without evidence of
criminal activity and based on the individuals’ perceived race, ethnicity, nationality or religion.
This report’s findings underscore the need for policy reforms to help identify and address
racial profiling, including expanded data collection and publication as well as enhanced
training and supervision on bias-free policing.
Recent events have underscored the concern of communities across Florida that racial
profiling is a problem in our state. In 2013 and 2014, media disclosed shocking reports that
the Miami Gardens Police Department conducted thousands of illegal stops, frisks, searches,
and arrests between 2008 and 2013 under a racially-motivated quota system that directed
officers to target Black men between the ages of fifteen and thirty. 3 Those reports led to a civil
rights lawsuit in federal court and disclosed that Miami Gardens police officers had even
subjected Black children to police-civilian encounters. 4 Also in 2013, an ACLU report analyzing
2010 data from the Federal Bureau of Investigation’s Uniform Crime Reporting Program and
the United States Census Bureau revealed that Black people in Florida were 4.2 times more
likely to be arrested for marijuana possession than white people—despite the fact that the
federal government has documented that Black and white people use marijuana at
comparable rates. 5
Media reports have raised additional public concerns about the ticketing of Black people for
low-level offenses at rates far out of proportion to their population figures. An investigation
into the Fort Lauderdale Police Department’s enforcement of a bicycle registration law from
2010 to 2013 found that 86% of citations were issued to Black people, 6 who constituted only
29% of the Fort Lauderdale population in 2013. 7 In April 2015, a Tampa Bay Times
investigation revealed that 79% of Tampa Police Department bicycle citations issued in the
previous twelve years were given to Black people, who comprised only about a quarter of the
Tampa population during the time. 8 A CBS Miami report examined 44,860 marijuana cases
5

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

closed between 2010 and 2014 in Miami-Dade County and found that 55% of the cases involved
Black defendants, even though Black people made up less than 20% of the county’s
population. 9
Public outcry concerning the ticketing of Black people for low-level offenses at rates far out of
proportion to their population figures stems from the possibility that racial profiling by police
who conduct pedestrian and traffic stops is one cause of these racial disparities. This concern
is particularly heightened when evidence shows that white and Black people engage in the
prohibited conduct at closely comparable rates, as is the case with marijuana possession. 10

THE FLORIDA SAFETY BELT LAW
The Florida Safety Belt Law was enacted in 1986 to grant law enforcement officers the
authority to stop and cite a driver for a violation of Florida’s safety belt requirements. The
statute was initially a “secondary enforcement” law, which permitted a law enforcement
officer to stop and cite a motorist only after observing another legal violation. 11
Concerns about racial profiling in traffic enforcement led to the amendment of the Florida
Safety Belt Law in 2005 to require “each law enforcement agency” to “adopt departmental
policies to prohibit the practice of racial profiling” and to record the race and ethnicity of all
individuals cited pursuant to the statute. 12 The law continues to require that all law
enforcement agencies maintain such policies and provide, on an annual basis, information
about the race and ethnicity of people issued seatbelt citations to the Governor, President of
the Senate, and Speaker of the House of Representatives. This requirement is carried out
through annual reports to the Florida Department of Highway Safety and Motor Vehicles
(FDHSMV). 13
In 2009, the Florida Safety Belt Law was further amended to permit “primary enforcement,”
which allowed law enforcement officers to stop and cite drivers solely for violating safety belt
requirements, in order to promote safety by deterring people from failing to wear seatbelts.
The amendment was passed despite concerns expressed by some legislators, including the
Florida Conference of Black State Legislators and the ACLU of Florida, that primary
enforcement of seatbelt requirements would further open the door to racially motivated traffic
stops.14 In enacting the amendment over these objections, the Florida Legislature appeared to
rely on the Florida Safety Belt Law’s pre-existing requirements for law enforcement agencies
to adopt policies prohibiting racial profiling and to annually collect and report race and
ethnicity data concerning seatbelt citations to state authorities.

6

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

Since 2005, the FDHSMV has published annual reports compiling the seatbelt citation data
submitted by law enforcement agencies across the state pursuant to the Florida Safety Belt
Law. The reports are publicly available on the FDHSMV website to “provide lawmakers,
partners and stakeholders, the media, and citizens with important facts and valuable
information related to public safety and motor services.” 15 The data may be analyzed to
identify racial disparities in the enforcement of Florida’s Safety Belt Law against white and
Black motorists, and thereby help to inform the public debate about racial profiling by Florida
law enforcement.

7

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

II. FINDINGS
Finding #1:

IN 2014, BLACK PEOPLE WERE STOPPED AND ISSUED SEATBELT
TICKETS BY LAW ENFORCEMENT ACROSS FLORIDA FAR OUT OF
PROPORTION TO THEIR ESTIMATED REPRESENTATION AMONG
DRIVERS.
Non-Hispanic Black people made up 21.9% of all seatbelt citation recipients in 2014, based on
data reported by the 147 law enforcement agencies that submitted seatbelt citation data to the
Florida Department of Highway Safety and Motor Vehicles. 36,838 out of a total of 168,199
seatbelt citations issued by county sheriff’s offices, municipal police departments, state law
enforcement agencies, and state university law enforcement agencies across Florida, were
issued to people whom law enforcement officers identified as non-Hispanic Blacks.16
A reasonable estimate of the racial composition of drivers on Florida roads, however, suggests
that significantly fewer seatbelt citations should have been issued to non-Hispanic Blacks.
According to data from the U.S. Census Bureau’s American Community Survey (ACS) Public
Use Microdata Sample (PUMS), in 2014, non-Hispanic Black people made up only 13.5% of the
estimated statewide population of driving age (15 to 85 years old) living in a household with
access to at least one vehicle. 17 While there is no public data that records the exact racial and
ethnic breakdown of all drivers on Florida roads in any given year, ACS/PUMS data provides a
reasonable approximation.
If non-Hispanic Black people had been cited for seatbelt violations in proportion to their share
of the estimated Florida driving age population with access to a vehicle in the household, they
would have received 20,296 fewer citations in 2014.

8

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

Finding #2:

IN 2014, LAW ENFORCEMENT OFFICERS ACROSS FLORIDA STOPPED
AND TICKETED BLACK MOTORISTS FOR SEATBELT VIOLATIONS
NEARLY TWICE AS OFTEN AS WHITE MOTORISTS.
The seatbelt citation data reported annually by law enforcement agencies to the Florida
Department of Highway Safety and Motor Vehicles can be used to calculate race-specific
seatbelt citation rates across the state of Florida. This analysis involves dividing the total
number of seatbelt citations issued to a racial group in a given year, as reported by the
FDHSMV, by an estimate of the race-specific resident driving population of Florida based on
U.S. Census ACS/PUMS data. See infra Section IV. Methodology, Data and Sources, 23.
This analysis reveals that in 2014, non-Hispanic Black people were issued seatbelt citations at
a rate of 1,821 per 100,000 resident drivers, while the citation rate for non-Hispanic white
people was 970 per 100,000 resident drivers. See infra Section V. Calculations, 32.
As a result, in 2014, law enforcement officers across Florida stopped and cited Black
motorists for seatbelt violations 1.9 times more often than white motorists. See infra
Section V. Calculations, 34.
9

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

What does this mean?
In layman’s terms, this means that after factoring each racial group’s share of the estimated
Florida resident population of driving age with access to at least one vehicle in their
household, Black motorists were NEARLY TWICE as likely to be stopped and issued a
seatbelt ticket than white motorists in 2014.
This racial disparity is highly significant in statistical terms. Even after taking into account
sampling error in the U.S. Census ACS/PUMS data used to calculate seatbelt citation rates,
the chances are less than 1 out of 1,000 that the ratio of the Black seatbelt citation rate to
the white seatbelt citation rate is less than 1.85 to 1. See infra Section V. Calculations, 34.

Finding #3:

RACIAL DISPARITIES IN FLORIDA SAFETY BELT LAW ENFORCEMENT
ARE PERSISTENT.
Racial disparities in seatbelt citation rates across Florida are not a new phenomenon. 199 law
enforcement agencies provided 2011 seatbelt citation data to the FDHSMV.18 Analysis of the
data shows that law enforcement officers across Florida stopped and issued seatbelt citations
10

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

to non-Hispanic Black people at a rate of 3,172 per 100,000 resident drivers, and to nonHispanic white people at a rate of 1,543 per 100,000 resident drivers. See infra Section V.
Calculations, 33.
As a result, in 2011, law enforcement officers across Florida stopped and cited Black
motorists for seatbelt violations 2.1 times more often than white motorists. See infra
Section V. Calculations, 34.
Even after taking the sampling error associated with the estimate of resident drivers of each
race into account, in 2011 the statewide Black citation rate was MORE THAN TWICE the
statewide white citation rate. See infra Section V. Calculations, 33.
This finding shows that sizable racial disparities in Florida seatbelt citation rates are a
persistent problem.

Finding #4:

ANY DIFFERENCES IN SEATBELT USE BY WHITE AND BLACK PEOPLE
DO NOT EXPLAIN WHY FLORIDA LAW ENFORCEMENT OFFICERS
STOPPED AND TICKETED BLACK MOTORISTS FOR SEATBELT
VIOLATIONS AT NEARLY TWICE THE RATE OF WHITE PEOPLE IN 2014.
Statewide studies document relatively minor differences in the failure to wear seatbelts
between white and Black people in Florida. The Florida Department of Transportation (FDOT)
conducts an annual observational study on seat belt use in Florida. The 2014 FDOT study
found that Black people wore seat belts at a rate approximately five to six percentage points
below that of white people, although usage rates in both groups increased following a
campaign to encourage seatbelt use. 19 Accordingly, 91.5% of white people and 85.8% of Black
people were observed to wear seat belts in the study.20
National studies similarly document disparities in the seat belt usage rates of Black and white
people in the range of three to nine percentage points. The National Highway Traffic Safety
Administration (NHTSA) has conducted two major surveys of seat belt use nationwide: the
National Occupant Protection Use Survey (NOPUS), and the Motor Vehicle Occupant Safety
Survey (MVOSS). The 2013 NOPUS, an observational study, found that the rate of seat belt use
by Black people fluctuated from five to nine percentage points lower than that of white
people.21 The 2007 MVOSS, a telephone survey administered to a random sample of
approximately 6,000 people aged 16 years and older, found a disparity in the rate of seat belt
use of approximately three to five percentage points between Black and white respondents. 22
11

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

Studies from other states similarly document a five to six percentage point gap in seat belt
usage by white and Black people.23
Observational and survey data on seat belt use are both susceptible to error. 24 But even if the
2014 FDOT study accurately measured that Black people in Florida wear seat belts at a rate
five to six percentage points lower than whites, that disparity does not explain why law
enforcement officers across Florida stopped and cited Black motorists for seat belt offenses at
NEARLY DOUBLE the rate of white motorists. Nor do the findings of the national studies
discussed above explain that disparity
For differences in seatbelt use to explain the statistically significant racial disparities in
citation rates documented by this report, Black people in Florida would have to wear seatbelts
at a rate nearly forty-five percentage points lower than white people. No evidence supports
such a finding.

12

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

Finding #5:

CERTAIN LAW ENFORCEMENT AGENCIES’ ENFORCEMENT OF THE
FLORIDA SAFETY BELT LAW HAS RESULTED IN RACIAL DISPARITIES
THAT EXCEED EVEN THE STATEWIDE AVERAGE RACIAL DISPARITY.
Law enforcement agencies across Florida collectively stopped and cited Black motorists 1.9
times more often than white motorists in 2014, and 2.1 times more often than white motorists
in 2011. We calculated citation rates for a sample of specific law enforcement agencies to
determine whether these statewide disparities were consistent throughout Florida. To
conduct this analysis, we used the same methods employed to calculate the statewide racial
disparity in seatbelt citation rates. See infra Section IV: Methodology, Data, and Sources, 22–
30.
The law enforcement agencies identified in the chart below stopped and ticketed non-Hispanic
whites and non-Hispanic Black motorists at rates that resulted in racial disparities that met
or exceeded the statewide racial disparity for the year the data was collected.

Law Enforcement Agencies that Met or Exceeded the Statewide Racial Disparity in
Seatbelt Citations
In 2011, Black motorists were Stopped and Cited for Seatbelt Offenses:

4.0 times more often than white motorists by the Escambia County Sheriff’s Office—
a disparity ALMOST DOUBLE the 2011 statewide racial disparity of 2.1.
In 2014, Black motorists were Stopped and Cited for Seatbelt Offenses:

3.0 times more often than white motorists by the Palm Beach County Sheriff’s Office;
2.8 times more often than white motorists by the Orange County Sheriff’s Office;
1.9 times more often than white motorists by the Broward County Sheriff’s Office;
as compared to the statewide racial disparity of 1.9.

The sheriff’s offices serving the counties of Escambia, Palm Beach, Orange, and Broward span
all corners of the state of Florida. Each enforces seatbelt requirements in a manner that
resulted in sizable and statistically significant racial disparity that is even larger than, or at
least equal to, the relevant statewide racial disparity in seatbelt citations. This finding calls
into question whether these agencies apply the Florida Safety Belt Law fairly, and necessitates
further investigation into these agencies’ enforcement practices.
13

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

Finding #6:

LAW ENFORCEMENT AGENCIES’ FAILURE TO REPORT SEATBELT
CITATION DATA TO STATE AUTHORITIES AS REQUIRED BY LAW
UNDERSCORES CONCERN ABOUT RACIAL DISPARITIES IN SAFETY
BELT LAW ENFORCEMENT.
Since 2005, the Florida Safety Belt Law has required that all law enforcement agencies collect
and report the race and ethnicity of seatbelt citation recipients to the Governor, President of
the Senate, and Speaker of the House of Representatives. This requirement is carried out
through annual reports to the FDHSMV. 25 A number of law enforcement agencies regularly fail
to follow this clear directive. In 2007, 293 law enforcement agencies reported data to the
FDHSMV, marking the highest rate of compliance since the Florida Safety Belt Law’s data
collection and reporting provision went into effect. In contrast, by 2014, the number of
reporting agencies had fallen to an all-time low of 147. 26

Among the law enforcement agencies that violate the Florida Safety Belt Law’s requirement to
collect and report seatbelt citation data are those serving Florida’s second (Miami) and third
(Tampa) largest cities.
14

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

TAMPA POLICE DEPARTMENT (TPD): TPD seatbelt citation data appears in only the FDHSMV
annual report reporting data from across Florida for 2005—the year the Florida Safety Belt
Law’s data collection and reporting requirement went into effect. This shows that TPD has
consistently failed to comply with the Florida Safety Belt Law’s data collection and reporting
requirements. 27
In response to an ACLU public records request for all records and data maintained or shared
with the FDHSMV pursuant to the Florida Safety Belt Law, 28 the Tampa Police Department
disclosed only the number of seatbelt citations it issued in 2014 to people identified as white,
Black, Hispanic, or “other” race. 29
In 2014, Tampa police officers issued 575 seatbelt citations to Black motorists and 549
seatbelt citations to white motorists. TPD officers stopped and issued more seatbelt citations
to Black motorists than white motorists despite the fact that Black people made up only 23% of
the Tampa population in 2014. 30
Although dataset limitations prevent a reliable calculation of the racial disparity in 2014 TPD
seatbelt citation rates, 31 even the raw numbers disclosed by TPD raises serious concerns.
Reported numbers show that Black motorists are stopped and cited far out of proportion to
their basic population figures, despite the lack of any evidence suggesting that Black people
make up a greater proportion of Tampa drivers or that Black people in Tampa commit seatbelt
violations at significantly higher rates than white people.
The TPD’s own data therefore raises questions about the source of the racial disparity in 2014
TPD seatbelt citations and why TPD has failed to report seatbelt citation data to state
authorities as required by law since 2005.
The City of Miami Police Department (MPD) failed to report data to the FDHSMV in 2005, 2009,
2012, 2013, and 2014. In response to an ACLU public records request for MPD seatbelt citation
data for 2005, 2009, 2012, and 2013. MPD responded that it has “no record pertaining to annual
reports on Seat Belt Violation Data Collection.” 32 The FDHSMV 2014 Report on seatbelt citation
data does not include any information from MPD. Nor did the City of Miami disclose any 2014
data to the ACLU.
MPD’s response to the ACLU’s public records request raises concerns about whether data on
the agency’s 2014 enforcement of the Florida Safety Belt Law would show racial disparities in
citation rates that are larger than the estimated statewide disparity, and why MPD has failed to
comply with the law’s data collection and reporting requirements.
15

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

Similarly, the Escambia County Sheriff’s Office and the Jacksonville Sheriff’s Office, which
serves Duval County, have failed to report seatbelt citation data to the FDHSMV since 2011.
Yet, the 2011 data reported by the Escambia County Sheriff’s Office reveals that the agency’s
stopped and issued seatbelt citations to Black people FOUR TIMES as often as white people—a
racial disparity that is ALMOST DOUBLE the statewide racial disparity of 2.1 for 2011.
Although dataset limitations prevent a reliable calculation of the racial disparity in the
Jacksonville Sheriff’s Office’s 2011 seatbelt citation rates, even the raw data reported raises
concerns. 33 In 2011, the Jacksonville Sheriff’s Office issued 2,855 seatbelt citations to Black
people—far more than the 2,307 seatbelt citations issued to white people—despite the fact
that Black people made up only 29% of the Duval County population in 2011. 34
The failure of each of these law enforcement agencies to collect and report seatbelt citation
data is a violation of the Florida Safety Belt Law. When that violation is accompanied by the
fact that analysis of the data that is disclosed reveals dramatic racial disparities in the rates at
which these agencies stop and ticket Black and white people for seatbelt offenses, there is an
even greater need for investigation.

16

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

III. RECOMMENDATIONS
Racial disparities in citation rates do not prove, without additional information, that law
enforcement officers have engaged in racial profiling. Nevertheless, the racial disparities in
Florida’s statewide seatbelt citation rates and the rates of specific law enforcement agencies
documented in this report require investigation and action for three reasons.
#1
Racial disparities in seatbelt citation rates may result from racial profiling,
particularly when white and Black people engage in the prohibited conduct at closely
comparable rates.
As discussed above, the Florida Department of Transportation’s 2014 study of seatbelt use
found only a five to six percentage point difference in the rates at which white and Black people
wear seat belts. That disparity in seatbelt use does not explain what analysis of 2014 statewide
seatbelt citation data reveals: that Black motorists were stopped and ticketed for seatbelt
offenses nearly twice as often as white motorists, even when accounting for their estimated
representation in the driving-age population living in a household with access to a vehicle.
The existence of such a large and statistically significant racial disparity in statewide seatbelt
citation rates—and the more dramatic racial disparities in specific law enforcement agencies’
seatbelt citation rates—demonstrates the need for investigation into police practices to
determine whether officers are targeting Black motorists for the encounters that result in
seatbelt citations in part, or even primarily, because of their race.
#2
Racial disparities in seatbelt citation rates raise concerns about the impact of
disproportionate police encounters, ticketing, and fines on people of color.
Fostering public safety by promoting seatbelt use is important. But stopping and ticketing
Black motorists for seatbelt citations at rates that far exceed that of white motorists and that
are not explained by any documented differences in seat belt use, suggests that communities
of color may disproportionately and unfairly bear the stigmatic harm and financial cost of
Florida Safety Belt Law enforcement.
A federal judge recognized the stigmatic harm stemming from high rates of stops of Blacks
and Latinos when addressing the New York Police Department’s stop-and-frisk program,
which disproportionately impacted these communities: “While it is true that any one stop is a
limited intrusion in duration and deprivation of liberty, each stop is also a demeaning and

17

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

humiliating experience. No one should live in fear of being stopped whenever he leaves his
home to go about the activities of daily life.” 35
There are also real financial costs to the disproportionate enforcement of seatbelt
requirements against Black motorists. The cost of a safety belt violation in Florida can be as
high as $60. 36 For those unable to pay, a seatbelt citation can lead to additional penalties,
including a driver’s license suspension pursuant to Florida law. 37
The financial burden of paying for fines associated with seatbelt citations hits communities of
color particularly hard because of the gap in white and Black household wealth—the
difference between the net worth (assets minus debts) of a typical white household and a
typical Black household. The racial wealth gap is a longstanding problem, and has even been
widening in recent years. In 2014, a Pew Research Center study found that the median wealth
of white households was 13 times the median wealth of Black households in 2013—the highest
racial wealth gap documented since 1989. 38
The disproportionate stopping and ticketing of Black motorists for seatbelt citations—as well
as other traffic and low-level offenses—exacerbates the already widening racial wealth gap by
disproportionately imposing fines on Black households in comparison to white households.
Moreover, because of the racial wealth gap, a greater proportion of Black motorists, in
comparison to white motorists, may have difficulty paying seatbelt fines, and thereby face a
higher risk of driver’s license suspension for nonpayment, additional tickets and fines for
driving on a suspended license, and a resulting cycle of tickets, debt and jail from which it may
be difficult to break free.
#3
Even routine police stops of civilians can dramatically escalate, subjecting those
disproportionately targeted for Florida Safety Belt Law enforcement to the risk of further
harm.
Tragic events in Florida and throughout the country illustrate that police-civilian encounters—
even traffic stops concerning minor infractions—carry a grave risk of escalation.
In 2014, Arthur Green Jr., a diabetic man, died in the custody of Tampa Police officers following
a traffic stop in which he was handcuffed. 39 In 2015, Walter Scott and Samuel Dubose were
shot and killed by police officers in South Carolina and Cincinnati following traffic stops for
minor traffic infractions (driving a car with a broken taillight, driving a car missing a front
license plate).40 Also in 2015, Sandra Bland was pulled over in Prairie View, Texas for a minor
traffic violation that escalated, leading to her arrest, jailing, and subsequent death.41

18

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

The Washington Post’s compilation of a database tracking deadly police shootings reveals that
traffic stops were one of the most common precursors to a fatal police shooting of a Black
person in 2015. 42
Racial disparities in seatbelt citation rates clearly call for serious response and policy reforms
to determine whether racial profiling is taking place and, if so, to eliminate it. In light of the
sizable racial disparities in Florida Safety Belt Law enforcement identified in this report, the
ACLU and the ACLU of Florida issue the following recommendations.

RECOMMENDATION #1:
INVESTIGATION
This report documents sizable and statistically significant disparities in the rates at which the
Sheriff’s Offices for Escambia, Palm Beach, Orange and Broward Counties issued seatbelt
citations to non-Hispanic white and non-Hispanic Black motorists —rates that met or
exceeded the already sizable statewide racial disparity in seatbelt citation rates of Florida law
enforcement agencies overall. These county-level disparities raise particular concern about
whether these sheriff’s offices engage in racial profiling because any differences in seatbelt
usage rates as documented by the state and national studies discussed above fail to explain
the size of the racial disparities in citation rates identified in this report.
The relevant County and City Commissions charged with oversight responsibility and the
Florida Attorney General’s Office of Civil Rights must investigate law enforcement agencies
whose implementation of the Florida Safety Belt Law has led to large racial disparities,
including the Sheriff’s Offices for Escambia, Palm Beach, Orange, and Broward Counties.
Their investigations should determine whether officers engage in racial profiling in traffic
enforcement, whether stops conducted to enforce seatbelt requirements are adequately
supported by individualized and articulable evidence to support reasonable suspicion that an
offense has occurred, whether testing on implicit bias and training on bias-free policing is
adequate, and whether the agencies adequately monitor and address racial disparities in the
practices of specific officers.

RECOMMENDATION #2:
ENFORCEMENT
All law enforcement agencies must comply with the Florida Safety Belt Law’s requirement to
collect and report to state authorities the race and ethnicity of people ticketed for safety belt
19

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

violations. This requirement was instituted to inform the public by identifying racial disparities
in traffic enforcement that may signal the existence of racial profiling. The data that has been
collected and reported in 2011 and 2014 demonstrate sizable, statistically significant, and
persistent racial disparities in seatbelt law enforcement. Yet, 52 fewer agencies reported
seatbelt citation data to state authorities in 2014 than in 2011, undermining the Florida Safety
Belt Law’s data reporting provision, which seeks to help inform the public about potential
indicators of racial profiling in traffic enforcement.
The Florida Legislature should enact legislation to create consequences for law enforcement
agencies that fail to comply with the Florida Safety Belt Law’s data collection and reporting
requirement. This includes the police departments that serve Florida’s second and third
largest cities—the City of Miami Police Department and the Tampa Police Department—as
well as the Escambia County Sheriff’s Office and the Jacksonville Sheriff’s Office.

RECOMMENDATION #3:
TRANSPARENCY
Law enforcement agencies should collect, report, and analyze race and ethnicity data for ALL
traffic and pedestrian stops, frisks, searches, citations, summons, and arrests to identify and
help eliminate racial profiling in all police-civilian encounters—not just those involving
enforcement of Florida’s Safety Belt Law.
The disclosure of seatbelt citation data pursuant to the Florida Safety Belt Law has already
helped inform the debate about racial profiling in Florida by permitting analysis that identifies
troubling racial disparities in seatbelt law enforcement. The disclosure of data more broadly
on police-civilian encounters, including traffic and pedestrian stops, frisks, and searches, will
help identify the causes of these racial disparities, as has the disclosure of such data by law
enforcement agencies across the country.
For example, the collection and publication of New York Police Department (NYPD) data on all
stops, frisks, and searches conducted between 2004 and 2009 permitted expert analysis, which
determined that the racial composition of New York neighborhoods was the main factor
determining stop rates, exceeding the role of crime, social conditions, and the allocation of
police resources.43 Expert analysis also found that although Blacks and Hispanics were
disproportionately stopped, police seized weapons or contraband more often in stops of whites
than in stops of Blacks or Hispanics.44 These findings helped inform a federal court’s 2013
ruling that the NYPD engaged in a pattern and practice of racial profiling and unconstitutional
stops, and decision to order policy reforms.45
20

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

Similarly, in 2014 and 2015, expert analysis of Boston Police Department (BPD) data on 20072010 police-civilian encounters revealed that even after controlling for crime and other social
factors, BPD officers were more likely to initiate encounters in neighborhoods with higher
concentrations of Black people.46 As a result of public debate and advocacy informed by these
disclosures, BPD announced reforms in 2015 to improve training on bias-free policing, to pilot
the use of body-worn cameras, and to make police-civilian encounter data publicly-available
on a regular basis.47 In early 2016, BPD disclosed data on 2011-2015 police-civilian
encounters. 48
In 2015, review of 250 randomly selected Chicago Police Department (CPD) field contact cards
for stops conducted in 2012 and 2013 revealed that Black Chicagoans were subjected to 72% of
all stops, yet constituted just 32% of the city’s population at the time. 49 Advocacy supported by
a report analyzing CPD stop data brought about a settlement to reform CPD stop-and-frisk
practices, including by requiring an independent evaluation of police practices and procedures,
improved data collection, and additional training for officers. 50
Time and time again, the disclosure of information about police-civilian encounters, including
the race and ethnicity of the civilian, the reason for the encounter, and the outcome of the
encounter, help to identify racial profiling and unsupported stops, frisks, and searches. The
existence of sizable and statistically significant racial disparities in Florida seatbelt citation
rates calls for the collection and publication of such data by Florida law enforcement agencies.

RECOMMENDATION #4:
SUPERVISION
The Florida Legislature enacted the Florida Safety Belt Law’s data collection and reporting
requirement to help inform the public about which law enforcement agencies may engage in
racial profiling in traffic enforcement. Yet, there is no indication that any of the law
enforcement agencies that reported data to the FDHSMV, or even FDHSMV itself, analyzed
seatbelt citation data to determine whether enforcement of seatbelt requirements has led to
racial disparities.
Law enforcement agencies should analyze the seatbelt citation data they collect and report to
FDHSMV and ensure that supervising officers monitor racial disparities in their supervisees’
enforcement of the Florida Safety Belt Law. Agencies should provide testing on implicit bias
and training on bias-free policing to all officers, and retrain officers whose stop and citation
practices result in large racial disparities.
21

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

IV. METHODOLOGY, DATA, & SOURCES
This report calculates disparities in the rates at which all Florida law enforcement agencies
combined issued seatbelt citations to white and Black motorists in 2014 and 2011. It also
calculates racial disparities in seatbelt citation rates for a sample of large law enforcement
agencies from across the state in order to determine whether the 2014 statewide disparity is
consistent across Florida.
We selected a sample of county sheriff’s offices for analysis because resource constraints
prohibited calculating race-specific citation rates for every law enforcement agency that
reported data in 2014. We focused on county sheriff’s offices because the U.S. Census
Bureau’s ACS/PUMS data provide a reasonable estimate of the resident driver population by
race for certain counties—the territories over which sheriff’s offices exercise jurisdiction—
thereby permitting the calculation of seatbelt citation rates for white and Black motorists by
those county sheriff’s offices. Through this method of selection, we decided to analyze
seatbelt citation rates for the following law enforcement agencies based on the most recent
data provided to the Florida Department of Highway Safety and Motor Vehicles: the Escambia
County Sheriff’s Office (2011 data); the Palm Beach County Sheriff’s Office (2014 data), the
Orange County Sheriff’s Office (2014 data), and the Broward County Sheriff’s Office (2014 data).
Dr. Ralph B. Taylor assisted in programming, calculations, and preparing sections of the
report. 51

DESCRIPTION VERSUS CAUSAL ANALYSIS
This report documents sizable and statistically significant disparities in the rates at which
Florida law enforcement agencies overall and specific law enforcement agencies issued
seatbelt citations to non-Hispanic white and non-Hispanic Black motorists in 2011 and 2014. It
raises questions about the cause(s) of the disparities in light of the finding of state and national
studies that document only a five to nine percentage point difference in the rate at which white
and Black people use seatbelts. This report does not, however, identify or isolate the cause(s)
of racial disparities in Florida seatbelt citation rates.
A causal analysis would require, at the very least, consideration of variables other than race
and seatbelt use that might impact seatbelt citation rates, such as driving quantity, driving
quality, and driving location. 52 This kind of empirical analysis is generally conducted through
use of multivariate regression analysis, which allows researchers to control for many factors
that may contribute to seatbelt citation rates and to estimate the individual impact of each.53
22

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

Further, in many instances causal analysis is likely to require a different type of research
design that goes beyond acquiring cross-sectional data. The racial disparities identified in this
report invite further analysis through use of such methods.

METHODOLOGY
In order to calculate racial disparities in race-specific Florida statewide seatbelt citation rates
and the race-specific seatbelt citation rates for particular law enforcement agencies, we first
calculated seatbelt citation rates for each of the jurisdictions of interest.
The “seatbelt citation rate” is calculated by dividing the number of seatbelt citations issued to
a racial group (e.g., non-Hispanic whites) in a given year and in a given jurisdiction by the
resident driving population of that racial group in the same geographic area. The latter is
divided by 100,000 to create citation rates per 100,000 resident drivers.
For example, in order to calculate the Palm Beach County Sheriff’s Office’s (PBCSO) seatbelt
citation rates for whites and Blacks, we used the following equations:
2014 PBCSO Seatbelt Citation
No. of seatbelt citations issued by
Rate for non-Hispanic whites = PBCSO in 2014 to non-Hispanic whites x 100,000
residents
2014 ACS/PUMS 1-year estimate of
number of non-Hispanic whites in PBC
of 15 to 85 years of age, living in a household
with access to at least one vehicle
2014 PBCSO Seatbelt Citation
No. seatbelt citations issued by
Rate for non-Hispanic Blacks = PBCSO in 2014 to non-Hispanic Blacks x 100,000
residents
2014 ACS/PUMS 1-year estimate of
number of non-Hispanic Blacks in PBC
of 15 to 85 years of age, living in a household
with access to at least one vehicle
The rate calculations thus compare the race-specific counts of seatbelt citations by law
enforcement agencies against a benchmark consisting of an estimate of the race-specific
number of residents of driving age living in a household with access to at least one vehicle in
the area of interest based on 1-year ACS/PUMS data. This benchmark was selected in order to
help develop a racial profile of the people who should be at risk of being issued a seatbelt
citation by police in a given jurisdiction, assuming no bias. Use of this benchmark permits a
higher quality estimation of race-specific citation rates (and disparities between those rates)
23

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

than rates derived from race-specific population figures in the geographic areas of interest. 54
This is because not all members of the non-Hispanic white or non-Hispanic Black resident
population are among the driving population that may be stopped and ticketed for a seatbelt
citation in a given jurisdiction. Using a benchmark of the race-specific population of driving
age (15 to 85 years old) living in a household with access to at least one vehicle, as estimated
by 1-year ACS/PUMS data, provides a better estimate of the size of the race-specific driving
population that may be stopped and ticketed in the geographic area of interest.
For example, perhaps officers of the Palm Beach County Sheriff’s Office stop and cite nonHispanic Blacks for seatbelt violations at a higher rate than non-Hispanic whites in part
because there are more non-Hispanic Blacks than non-Hispanic whites driving on Palm Beach
County roads—a factor that is influenced by the size of each population that is of driving age
and that has access to a vehicle to drive. If non-Hispanic Blacks are over-represented in the
population of drivers on Palm Beach County roads, they more likely to be stopped and issued a
seatbelt citation. Using ACS/PUMS data to identify the population figures for non-Hispanic
Blacks and non-Hispanic whites of driving age (15 to 85 years old) who report living in a
household with access to at least one vehicle in each of the geographic areas studied helps
account for these factors. 55
After obtaining the race-specific seatbelt citation rates for each of the law enforcement
agencies of interest—Florida statewide, Escambia County Sheriff’s Office, Palm Beach County
Sheriff’s Office, Orange County Sheriff’s Office, and Broward County Sheriff’s Office—we
calculated the racial disparity by dividing the higher citation rate by the lower rate. In each
case, the Black citation rate was higher than the white citation rate. The result of the
calculation is a number that represents the number of times more often the relevant law
enforcement agency stopped and issued seatbelt citations to Blacks as compared to whites.
For example:
Racial Disparity in PBCSO
=
2014 Seatbelt Citation Rates

PBCSO 2014 Seatbelt Citation Rate for non-Hispanic Blacks
PBCSO 2014 Seatbelt Citation Rate for non-Hispanic whites

DATA AND SOURCES
A. Florida Department of Highway Safety and Motor Vehicles Seat Belt Violation Reports
The numerators for the seatbelt citation rate calculations were drawn from data published in
2014 and 2011 reports of the Florida Department of Highway Safety and Motor Vehicles, which
are available on the FDHSMV website and compile seatbelt citation data reported by law
enforcement across the state. 56 In 2014, 147 law enforcement agencies reported seatbelt
24

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

citation data to FDHSMV. 57 In 2011, 199 law enforcement agencies reported data to the state
agency. 58 The numerators for the rate calculations for the Sheriff’s Offices for Palm Beach,
Broward, and Orange Counties are drawn from data published in the FDHSMV 2014 Report.
The numerator for the rate calculations for the Escambia County Sheriff’s Office was drawn
from data published in the FDHSMV 2011 Report because the Escambia County Sheriff’s Office
failed to reported seatbelt citation data to the FDHSMV after that year.
B. 1-Year American Community Survey Public Use Microdata Sample Data Files
The denominators for the rate calculations are estimates of the race-specific resident driver
populations in the areas served by the law enforcement agencies studied, and were derived
through use of the U.S. Census Bureau’s American Community Survey (ACS) one-year Public
Use Microdata Sample (PUMS) data files.
The PUMS data files are a subset of the American Community Survey, which is an ongoing
survey that provides population information on a yearly basis. 59 The U.S. Census Bureau
produces the ACS/PUMS files so that data users can create custom tables containing
information about individuals or housing units sampled in the American Community Survey
that are not available through pretabulated (or summary) ACS data products. 60 ACS/PUMS
data files include records with individual response information, including responses to a
question about household vehicle access. The 1-year ACS/PUMS files are accessible via
American FactFinder, the Census Bureau’s FTP site, and DataFerrett. 61
To derive estimates of the race-specific resident driver population in each of the areas of
interest, we downloaded 2014 and 2011 ACS 1-year PUMS files containing data on individuals
and households in Florida through the American FactFinder website. 62 The data files were
extracted and imported into Stata for analysis.
We separately analyzed 2014 and 2011 data using Stata (v. 14). We first merged the person
and household data files for 2014 ACS 1-year PUMS data for Florida into one file. In the
merged file, each record corresponds to an individual and contains information about the
individual’s household, including the number of vehicles available for the household. Within
this merged data file, we selected the appropriate geography (e.g., the entire state of Florida),
and asked the program to provide, after taking survey design and sampling error into account,
the total number of white non-Hispanic residents of driving age (15 to 85 years old) with access
to at least one vehicle in the household. The same was done for non-Hispanic Blacks in the
same geographic area. We followed the same methodology to secure denominators for white
and Black 2011 statewide seatbelt citation rate calculations.

25

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

In order to obtain the denominators for the rate calculations for the Sheriff’s Offices for Palm
Beach County, Orange County, and Broward County, we used 2014 1-year ACS/PUMS data files
to identify the number of non-Hispanic whites of driving age with access to at least one vehicle
in the household, and the number of non-Hispanic Blacks of driving age with access to at least
one vehicle in the household. We were able to do this by selecting the records in the data file
corresponding to particular Public Use Microdata Areas (PUMAs) that fall within each county.
The U.S. Census Bureau uses Public Use Microdata Areas (PUMAs) as the intra-state
geographic units for collecting PUMS data.63 PUMAs cover the entire territory of the United
States (and U.S. territories), and do not cross state lines. We used the U.S. Census Bureau’s
TIGERweb mapping utility to identify PUMAs that align with the territorial boundaries of each of
the counties of interest—Escambia, Palm Beach, Orange and Broward. 64 For each county, we
were able to identify the specific PUMAs that together covered the entire population contained
within the county’s political boundaries. This meant that 1-year ACS/PUMS data could be used
to estimate the number of race-specific residents of driving age living in a household with
access to at least one vehicle in each county of interest. 65

SURVEY WEIGHTING AND ERROR ESTIMATION IN 1-YEAR ACS/PUMS DATA
A. Survey Weighting
The ACS/PUMS microdata sets are individual-level records that are in effect a sample of a
sample. The sample of households from which PUMS households are sampled are the
American Community Survey. All individuals in an ACS-sampled household which is sampled
for inclusion in the PUMS are reported as person level records in the PUMS data.
Because the PUMS uses a complex, multi-stage sampling design, individual PUMS records
must be re-weighted to counterbalance sampling decisions in order to obtain accurate
population estimates. 66 Person-level estimates from ACS/PUMS data will be representative of
the population from which they were sampled (putting aside non-sampling errors), if the
estimates are weighted using the weight variable “PWGTP: Person's weight for generating
statistics on individuals (such as age).”
B. Sampling Error
All samples contain sampling error. In a properly executed survey sampling design, those
errors will be random and will not bias the estimates derived from the survey. But sampling
error creates uncertainty when the researcher wishes to make inferences back to the
26

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

population from which the surveyed individuals/households were sampled. In other words, the
uncertainty around a population parameter obtained from a representative sample, like the
ACS/PUMS, must be captured. One way to do so is by determining a confidence interval, which
is a calculation that expresses the degree of uncertainty represented in a data sample.
Because 1-year ACS/PUMS data were generated from a complex sampling design, estimation
of the uncertainty arising from sampling errors associated with these estimates requires
special treatment. The U.S. Census Bureau recommends approximating direct standard
errors for population estimates obtained through use of 1-year ACS/PUMS data files by using
their replicate weights. 67 To facilitate this approximation, eighty replicate weights are attached
to each record in a 1-year ACS/PUMS data file. Each of these eighty sets of weights represents
an independent estimate of each person’s weight using a balanced repeated replication (BRR)
technique. 68
Stata (v. 14), the analysis program used here, permits using these replicate weights to make
appropriate survey-based estimates and the sampling error around those estimates, by survey
setting the data first—in essence telling the program about the survey sampling design—and
second, reminding the program that survey-adjusted estimates are requested when specific
commands are invoked. 69 The eighty replicate weights are used to make 80 estimates of the
sample statistic requested, and the standard error based on those 80 estimates is reported.
Estimates of standard errors and confidence intervals which do not take the survey design into
account can under-estimate the standard errors associated with a sample statistic.
To generate the race-specific population totals for residents of driving age living in a
household with access to at least one vehicle in each of the geographic areas of interest (the
denominator for seatbelt citation rate calculations), the Stata command “svy: total” was used.
This command specifies the probability level for the confidence intervals at 99.9%.
This means, assuming no non-sampling error, that the population number sought is likely to
lie within the interval provided 999 times out of 1,000. The chance that the population number
is outside the confidence interval provided (again assuming no non-sampling error) is less
than one in a thousand.
The confidence intervals associated with the denominator derived from 1-year ACS/PUMS data
files are simply multiples of the standard errors reported. For example:
The upper limit of the 99.9% confidence interval (99.9% UCL) =
Driving-age Population Estimate + 3.2905*(standard error of estimate).

27

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

The lower limit of the 99.9% confidence level (99.9%LCL) =
Driving-age Population Estimate – 3.2905*(standard error of estimate).
C. Non-Sampling Error Concerns at the State and County Levels
As discussed, the denominators used for the seatbelt citation rate calculations are estimates
of the race-specific resident population of driving age living in households with access to at
least one vehicle. At the state level, use of this denominator has two limitations that are not
related to possible sampling error in the 1-year ACS/PUMS data from which these estimates
are derived. For the reasons explained below, the first limitation suggests that this report
might underestimate racial disparities in seatbelt citation rates, while the impact of the second
is unquantifiable based on the available information.
1. Vehicles Per Household
First, the denominator used assumes that exposure to risk of receiving a seatbelt violation is
the same for all members of the non-seatbelt-using driving population regardless of the
number of vehicles to which each member of that group has access in his/her household. It
seems plausible, all else equal, that those members of the non-seatbelt-using driving
population with access to more vehicles in their respective households will drive more miles
per year, and thus be at greater risk of receiving a seatbelt violation from a law enforcement
officer.
The assumption described above is not problematic, however, as long as violations of that
assumption are race neutral.
The violations of that assumption, however, may not be race neutral. The fraction of white
people in the group of interest with access to more than one vehicle in their household might
be higher than it is for Black people in the group of interest. If, net of other factors among the
driving and non-seatbelt-using population, such a difference in vehicle access exists and
translates to exposure to the risk of receiving a citation, the race differences seen in this
report might underestimate racial disproportionality in seatbelt citation rates.
Differences by race in the number of vehicles accessible within driving households do appear
in the dataset. In 2014, among the black non-Hispanic resident driving population, the percent
with access to only one car was 36.7 percent (survey weighted). That same year, among the
white non-Hispanic resident driving population, the comparable figure was 30.9 percent
(survey weighted). This means that the fraction of resident white non-Hispanics of driving age
with access to more than one vehicle was larger (69.1 percent) than the comparable fraction of
28

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

resident Black non-Hispanics of driving age (63.3 percent). Roughly the same differences
were seen using the 2011 data.
If this same race difference applies to non-seatbelt-using drivers, and if other determinants of
miles driven besides numbers of vehicles are racially neutral for these same resident
drivers—an assumption that cannot be tested with these data—this difference might mean that
the racial discrepancies noted in this report are low estimates.
2. Racial Mix of Non-Resident Drivers
For each jurisdiction of interest—the state of Florida, Escambia County, Palm Beach County,
Orange County, and Broward County—the denominator is based on resident race-specific
driving populations. These denominators do not capture the impact of non-resident drivers
who receive seatbelt citations while traveling in each area of interest.
This limitation becomes problematic only to the extent that the racial mix of non-resident
drivers receiving seatbelt citations while driving in the jurisdiction of interest differs markedly
from the racial mix of drivers receiving seatbelt citations who reside within the jurisdiction of
interest. A marked discrepancy between the racial mix of out-of-jurisdiction, non-seatbeltwearing drivers and the in-jurisdiction racial mix means that the two race-specific residentbased denominators used for seatbelt citation rate calculations may differ in the adequacy with
which they capture the population at risk.
There is no way to quantify the impact of this limitation of the data used for the denominator in
the seatbelt citation rate calculations, or the extent to which it is problematic, given the data
sources used here. Nevertheless, this limitation is potentially more problematic at the level of
seatbelt citation rate calculations for county sheriff’s offices than for the state of Florida as a
whole.

LACK OF ADEQUATE DATA ON LATINO DRIVERS
This report addresses disparities in the rate at which law enforcement agencies issued
seatbelt citations to people identified as non-Hispanic white and non-Hispanic Black by law
enforcement officers. It does not analyze seatbelt citation data regarding Latinos, who are
referred to as “Hispanic” in the seatbelt citation data reported by law enforcement agencies to
the FDHSMV pursuant to the Florida Safety Belt Law.
Data concerning the race and ethnicity of recipients of seat belt citations rely on the
observation of law enforcement officers who issue the citations, rather than self-identification
29

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

by the individuals cited. As a result, there is a risk that law enforcement officers improperly
identify the race or ethnicity of the person cited for a seatbelt violation. This risk is particularly
high with respect to Latinos, who are often misidentified as non-Hispanic white or nonHispanic Black. The likely underreporting of citations issued to Latinos makes it difficult to
calculate seatbelt citation rates for Latinos and to assess any disparities between that rate and
the seatbelt citation rate for non-Hispanic white people.
Accordingly, this report focused on analyzing seatbelt citation rates for motorists identified by
law enforcement officers in their own data as non-Hispanic white and non-Hispanic Black.
Nevertheless, the recommendations offered to further investigate and address the sizable and
statistically significant racial disparities in white/Black seatbelt citation rates across Florida
and by specific law enforcement agencies would benefit all communities, including Latinos, as
well as the public at large.

30

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

V. CALCULATIONS
2014 Percentage of Non-Hispanic Blacks Among Florida Resident Drivers

Number of non-Hispanic Black Florida
residents aged 15 to 85 years old in
household with access to 1+ vehicle in
2014*

2,022,816

Total Number of Florida residents (of
any race) aged 15 to 85 years old in
household with access to 1+ vehicle in
2014*

14,943,584

Lower Bound of 99.9%
Confidence Interval
2,013,067

Upper Bound of 99.9%
Confidence Interval
2,032,565

14,925,718

14,961,450

(SE 9,749)

(SE 17,866)

13.54%
13.48%
Percentage of Florida residents aged
15 to 85 years old in household with
access to 1+ vehicle in 2014 consisting
of non-Hispanic Blacks*
SE = Standard Error
* Estimate Based on 2014 1-year ACS/PUMS data file for Florida.

13.56%

2014 Overrepresentation of non-Hispanic Blacks in Statewide Seatbelt Citations
Actual
Seatbelt
Citations
Issued*

Expected Value Based on
Representation of
non-Hispanic Blacks Among
Florida Resident Driver
Population at 13.5%

Difference
Between
Actual and
Expected Citation
Numbers

99.9%
Confidence
Interval**

Total Number Seatbelt 168,199
Citations Issued Across
Florida in 2014
22,707
20,296
20,099 to 20,494
Total Number Seatbelt 36,838
Citations Issued Across
Florida to non-Hispanic
Blacks in 2014
* Department of Highway Safety and Motor Vehicles, Seat Belt Violation Data Collection 316.614(9), F.S. Annual

Report 2014

**99.9% Confidence Interval = the interval within which the population estimate is likely to lie 999 times out of
1,000, assuming no non-sampling error

31

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

2014 Seatbelt Citation Rates
Non-Hispanic whites

State of
Florida

Palm Beach
County
Sheriff’s
Office

Numerator

Denominator

(number of
seatbelt
citations
issued in
2014)*

(number of
residents
aged 15 to
85 years old
in household
with access
to 1+ vehicle
in 2014)**

85,368
(total
issued by
all
reporting
agencies)
1,487

Orange
County
Sheriff’s
Office

2,692

Broward
County
Sheriff’s
Office

1,205

Non-Hispanic Blacks
Seatbelt
Citation Rate
per 100,000
Residents

Numerator

Denominator

(number of
seatbelt
citations
issued in
2014)*

(number of
residents
aged 15 to 85
years old in
household
with access
to 1+ vehicle
in 2014)**

8,797,997

970

36,838

2,022,816

1821

(SE 10,974)

(99.9% CI:
966 to 975)

(SE 9,749)

(99.9% CI:
1,792 to 1852)

643,016

231

(total
issued by
all
reporting
agencies)
1,118

163,337

685

(SE 3,122)

(99.9% CI:
228 to 235)

(SE 2,933)

(99.9% CI:
645 to 729)

422,060

638

167,551

1,797

(SE 2,511)

(99.9% CI:
625 to 651)

(SE 3,157)

(99.9% CI:
1,688 to 1,921)

588,017

205

348,903

389

(SE 2,891)

(99.9% CI:
202 to 208)

(SE 4,358)

(99.9% CI:
373 to 406)

3,011

1,357

Seatbelt
Citation Rate
per 100,000
Residents

SE = Standard Error
99.9% CI = 99.9% Confidence Interval (i.e., the interval within which the population estimate is likely to lie 999
times out of 1,000, assuming no non-sampling error)
* Department of Highway Safety and Motor Vehicles, Seat Belt Violation Data Collection 316.614(9), F.S. Annual

Report 2014

** Estimate Based on 2014 1-year ACS/PUMS data file for Florida.

32

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

2011 Seatbelt Citation Rates
Non-Hispanic Whites

State of
Florida

Numerator

Denominator

(number of
seatbelt
citations
issued in
2011)*

(number of
residents
aged 15 to 85
years old in
household
with access
to 1+ vehicle
in 2014)**

133,727
(total issued
by all
reporting
agencies)
199

Non-Hispanic Blacks
Citation Rate
per 100,000
Residents

Numerator

Denominator

Citation Rate
per 100,000
Residents

(number of
seatbelt
citations
issued in
2011)*

(number of
residents
aged 15 to 85
years old in
household
with access
to 1+ vehicle
in 2014)**

8,667,254

1543

59,720

1,882,495

3172

(SE 10,908)

(99.9% CI:
1536 to 1550)

(total issued
by all
reporting
agencies)

(SE 10,312)

(99.9% CI:
3,114 to 3,233)

148,893
229
42,409
134
540
Escambia
County
(99.9% CI:
(99.9% CI:
(SE 1,588)
(SE 975)
Sheriff’s
129 to 139)
501 to 586)
Office
SE = Standard Error
99.9% CI = 99.9% Confidence Interval (i.e., the interval within which the population estimate is likely to lie 999
times out of 1,000, assuming no non-sampling error)
* Department of Highway Safety and Motor Vehicles, Seat Belt Violation Data Collection 316.614(9), F.S. Annual

Report 2011

** Estimate Based on 1-year 2011 ACS/PUMS data file for Florida.

33

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union

2014 Racial Disparity Analysis
Seatbelt Citation Rate
per 100,000 residents

State of
Florida
Palm Beach
County
Sheriff’s Office
Orange County
Sheriff’s Office
Broward
County
Sheriff’s Office

Disparity Between nonHispanic white and nonHispanic Black Seatbelt
Citation Rates

99.9% Confidence Interval for
Racial Disparity Indicator based
on Standard Errors for
Denominator

NonHispanic
whites
970

NonHispanic
Blacks
1821

1.88

1.85 to 1.90

231

685

2.96

2.83 to 3.10

638

1797

2.82

2.70 to 2.95

205

389

1.90

1.85 to 1.95

2011 Racial Disparity Analysis
Seatbelt Citation Rate per
100,000 residents

State of Florida
Escambia County
Sheriff’s Office

Non-Hispanic
whites
1543
134

Non-Hispanic
Blacks
3172
540

Disparity Between nonHispanic white and nonHispanic Black Seatbelt
Citation Rates

99.9% Confidence Interval for
Racial Disparity Indicator based
on Standard Errors for
Denominator

2.06
4.04

2.03 to 2.09
3.88 to 4.22

34

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union
1

Fla. Stat. Ann. § 316.614 (9) (West 2015).

2

This report addresses disparities in the rate at which law enforcement agencies issued seatbelt citations to people identified
as “non-Hispanic white” and “non-Hispanic Black” by law enforcement officers. It does not analyze data regarding Latinos,
who are referred to as “Hispanic” in the seatbelt citation data reported by law enforcement agencies pursuant to the Florida
Safety Belt Law. See infra Section IV. Methodology, Data & Sources at 29-30.

3

Daniel Lieberman, Bullies With Badges? Store Owner Films Police Arresting Employee, FUSION, Nov. 25, 2013,
http://fusion.net/story/4180/bullies-with-badges-store-owner-films-police-arresting-employee/; Alice Brennan & Dan
Lieberman, Florida City’s ‘Stop and Frisk’ Nabs Thousands of Kids, Finds 5-Year-Olds ‘Suspicious’, FUSION,
http://fusion.net/story/5568/florida-citys-stop-frisk-nabs-thousands-of-kids-finds-5-year-olds-suspicious/ (last visited Nov.
28, 2015).
4

See Complaint at 22–24, Sampson v. City of Miami Gardens, No. 1:13-cv-24312 (S.D. Fla. Nov. 27, 2013); Morgan Whitaker,
Police Sued for Alleged Racial Profiling, Stop and Frisk, MSNBC.COM, Dec. 6, 2013,
http://www.msnbc.com/politicsnation/police-sued-alleged-racial-profiling.
5

AMERICAN CIVIL LIBERTIES UNION, THE WAR ON MARIJUANA IN BLACK AND WHITE 144 (2013),
https://www.aclu.org/sites/default/files/field_document/1114413-mj-report-rfs-rel1.pdf (discussing racial disparities in
marijuana possession arrest rates across Florida); id. at 66-68 (discussing data from the National Household Survey on Drug
Abuse and Health documenting that whites and Blacks use marijuana at similar rates).
6

Kyle Swenson, Biking While Black Is a Crime, MIAMI NEW TIMES, Oct. 31, 2013, http://www.miaminewtimes.com/2013-1031/news/biking-while-black-fort-lauderdale/full/.

7

In 2013, Fort Lauderdale’s population was 172,374, according to a 1-year estimate from the American Community Survey.
U.S. Census Bureau; American Community Survey, 2013 American Community Survey 1-Year Estimates, Table DP05;
generated by Nina Papachristou using American FactFinder; http://factfinder2.census.gov; (Dec. 2, 2015).
8

Alexandra Zayas & Kameel Stanley, How Riding Your Bike Can Land You In Trouble With the Cops—If You’re Black, TAMPA BAY
TIMES, Apr. 17, 2015, http://www.tampabay.com/news/publicsafety/how-riding-your-bike-can-land-you-in-trouble-with-thecops---if-youre-black/2225966; Kameel Stanley, Some in Tampa Want Assurance Federal Bike Ticket Review Will Be
Thorough, TAMPA BAY TIMES, May 11, 2015, http://www.tampabay.com/news/publicsafety/some-in-tampa-want-assurancefederal-bike-ticket-review-will-be-thorough/2229232; Noah Pransky, Black Arrest Numbers Off-The-Charts in Tampa,
WTSP.COM, May 6, 2015, http://www.wtsp.com/story/news/investigations/2015/05/05/majority-of-tampa-arrests-are-blackpeople/26940149/; Justice Probe Welcome, But Tampa Police Should Change Now, TAMPA BAY TIMES, Apr. 23, 2015,
http://www.tampabay.com/opinion/editorials/editorial-justice-probe-welcome-but-tampa-police-should-changenow/2226780.
9

Jim DeFede, Race Matters: Marijuana Cases Flood Court System, CBS MIAMI, June 29, 2015,
http://miami.cbslocal.com/2015/06/29/race-matters-marijuana-cases-flood-court-system/.
10

A 2014 survey of individuals over the age of 12 by the U.S. Department of Health & Human Services Substance Abuse and
Mental Health Services Administration estimated that 15.7% of Blacks and 13.7% of whites had used marijuana at some point
in the past year. Results from the 2014 National Survey on Drug Use and Health: Detailed Tables, Table 1.24B (Marijuana Use
in Lifetime, Past Year, and Past Month among Persons Aged 12 or Older, by Demographic Characteristics: Percentages, 2013
and 2014) (2014), http://www.samhsa.gov/data/sites/default/files/NSDUH-DetTabs2014/NSDUH-DetTabs2014.htm#tab1-24b.

11

See Fla. Stat. Ann. § 316.614(9) (West 1987).

12

Fla. Stat. Ann. § 316.614(9) (West 2015).

13

Id.

14

See Jessie-Lynne Kerr, New Florida Seat Belt Law Begins Tuesday in Florida, JACKSONVILLE.COM, June 29, 2009,
http://jacksonville.com/news/metro/2009-06-29/story/new_florida_seat_belt_law_begins_tuesday_in_florida (quoting
Representative Audrey Gibson: "I do have an issue with racial profiling not only of blacks but also of brown minorities, and I
understand that seat belts save lives," she said. "But I view this as a primary stop bill, not a primary seat belt law, so let's call
it what it really is."); Matthew L. Wald, U.S. Presses States for Strict Seat Belt Laws, N.Y. TIMES, Mar. 5, 2004,
http://www.nytimes.com/2004/03/05/us/us-presses-states-for-strict-seat-belt-laws.html (quoting concerns of Florida Senate
president James E. King, Jr. that primary enforcement of seatbelt requirements had led to racial profiling in Pennsylvania).
15

Florida Highway Safety and Motor Vehicles Website, http://www.flhsmv.gov/resource-center/cabinet-and-legislaturereports/ (listing “Seat Belt Violation Reports” from 2005 to 2014).

16

Department of Highway Safety and Motor Vehicles, Seat Belt Violation Data Collection 316.614(9), F.S. Annual Report 2014.,
http://www.flhsmv.gov/html/pdf/sbv2014.pdf [hereinafter “FDHSMV 2014 Report”).

35

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union
17

According to U.S. Census ACS/PUMS data, an estimated 14,943,584 people (Standard Error 17,866) in Florida were of driving
age (aged 15 to 85 years old) and living in a household with access to at least one vehicle in 2014. Of that figure, an estimated
2,022,816 people (Standard Error 9,749) were non-Hispanic Blacks. Accordingly, non-Hispanic Blacks made up an estimated
13.5% of the 2014 Florida population of driving age and living in a household with access to at least one vehicle, with a 99.9%
confidence interval of 13.5% to 13.6%. See Section V. Calculations.

18

Department of Highway Safety and Motor Vehicles, Seat Belt Violation Data Collection 316.614(9), F.S. Annual Report 2011,
http://www.flhsmv.gov/html/pdf/SBV2011.pdf [hereinafter “FDHSMV 2011 Report”).

19

The June 2014 Final Report sampled 165 randomly selected controlled roads and freeway exits across 15 counties.
Observers recorded front seat drivers and outboard passengers’ seat belt use, as well as their sex, age, race, and type of
vehicle. FLA. DEP’T OF TRANSP., JUNE 2014 SEAT BELT USE IN FLORIDA: FINAL REPORT 8 (2014),
http://www.floridaoprc.ce.ufl.edu/Document.asp?DocID=465 (last visited Jan. 16, 2016). FDOT observed that in April/May
2014, prior to the campaign, 89.6% of white people observed wore seat belts and 83.9% of Black people observed wore seat
belts. Id. at 16.
20

Id. at 16.

21

NAT’L HIGHWAY TRAFFIC SAFETY ADMIN., OCCUPANT RESTRAINT USE IN 2013: RESULTS FROM THE NATIONAL OCCUPANT PROTECTION USE
SURVEY CONTROLLED INTERSECTION STUDY 4 (2015), http://www-nrd.nhtsa.dot.gov/Pubs/812080.pdf. The NOPUS study was
conducted by observing vehicle occupants at controlled intersections, and recording seat belt use for occupants 8 years of age
and older in both front and rear seats. Id.
22

When asked how often respondents wore seatbelts, 84 percent of Blacks reported using their seat belts “all of the time,” as
compared to 87 percent of Whites. NAT’L HIGHWAY TRAFFIC SAFETY ADMIN., 2007 MOTOR VEHICLE OCCUPANT SAFETY SURVEY, VOLUME 2:
SEAT BELT REPORT 15 (2008),
http://www.nhtsa.gov/DOT/NHTSA/Traffic%20Injury%20Control/Articles/Associated%20Files/810975.pdf. When the
interviewers followed up this question by asking subjects when was the last time they did not wear a seat belt while driving, a
number of respondents who had just reported “all of the time” use responded that the last time they did not wear their seat
belt was yesterday or that very morning. Id. at 22. When respondents who conceded to not wearing their seat belt in the past
day or week while driving were subtracted from the “all of the time” category, the usage rate among both blacks (77%) and
whites (82%) declined, and the differential between the two groups increased to 5 percentage points. Id. at 23.
23

A 2014 observational survey of seat belt use in Ohio found that Whites were belted 85.7% of the time, and African Americans
were belted 80.9% of the time. Robert L. Seufert et al., Observational Survey of Seat Belt Use in Ohio 24 (2014),
http://www.highwaysafetyoffice.ohio.gov/Reports/2014ObservationSurvey.pdf. A 2014 observational study in Rhode Island
found that seat belt use “did not differ greatly by race.” Katherine A. Raboin & Neil K. Chaudhary, 2014 Seat Belt Use in Rhode
Island 11 (2014), http://www.dot.ri.gov/documents/community/safety/2014_Seat_Belt_Usage.pdf. Among drivers, Whites were
belted 86.3% of the time while African Americans were belted 85.2% of the time; among passengers, Whites were belted
88.4% of the time while African Americans were belted 85.5% of the time. Alexander Weiss, Illinois Traffic Stop Statistics Act
Report for the Year 2004, NORTHWESTERN UNIV. CTR. FOR PUB. SAFETY, IL. DEP’T OF TRANSP., 30 (2005),
http://www.idot.illinois.gov/Assets/uploads/files/Transportation-System/Reports/Safety/Traffic-Stop-Studies/2004/
2004%20Illinois%20Traffic%20Stop%20Summary.pdf. In both studies, the researchers qualified their findings by noting that
the observations for non-White passengers were limited to a small sample size.
24

In observational studies, researchers are required to determine the race and ethnicity of drivers that are in motion. Aside
from the general risk that perceptions of race are subjective, the speed at which observational observers must identify the
race of subjects in motion creates an added layer of potential error. Self-reported data through questionnaires and
interviews, may not reflect actual seat belt usage rates because drivers who do not use seat belts may be less likely to
respond honestly, or respond at all. See, e.g., WEISS, supra note 23 at 30.
25

Fla. Stat. Ann. § 316.614(9) (West 2015).

26

Compare Department of Highway Safety and Motor Vehicles, Seat Belt Violation Data Collection 316.614(9), F.S. Annual
Report 2007, with Department of Highway Safety and Motor Vehicles, Seat Belt Violation Data Collection 316.614(9), F.S.
Annual Report 2014.
27
Department of Highway Safety and Motor Vehicles, Seat Belt Violation Data Collection 316.614(9), F.S. Annual Reports 20052014.
28

Letter from Nusrat Choudhury, Staff Attorney, ACLU, to Police Chief Eric Ward, Tampa Police Department, (July 2, 2015) at
3 (on file with author).

29

Letter from Kirby C. Rainsberger, Senior Assistant City Attorney, City of Tampa, to Nusrat Choudhury, Staff Attorney, ACLU,
(July 20, 2015) at 2 (on file with author).

36

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union
30

U.S. Census Bureau; American Community Survey, American Community Survey 2014 1-year Estimate; Table DP05;
generated by Nina Papachristou; using American FactFinder; http://factfinder2.census.gov (Jan. 19, 2016).
31

The U.S. Census’s 2014 1-year ACS/PUMS data does not provide a reasonable estimate of the race-specific resident driver
population of the City of Tampa in 2014 as it did for Palm Beach, Orange, and Broward Counties due to a spatial mismatch
between the territorial limits of the City of Tampa, in which TPD exercises law enforcement jurisdiction, and the Public Use
Microdata Areas (PUMAs) from which ACS/PUMS data is drawn. Using the U.S. Census’s TIGERweb mapping utility, we
discovered that several of the PUMAs that include territory within the Tampa city limits also included sizable resident
populations of immediately adjacent municipalities. As a result, PUMAs encompassing the territory of the City of Tampa fail to
align closely with city limits and do not permit using ACS/PUMS data to estimate the race-specific population of resident
drivers in Tampa.
32

E-mail from Kimberlyn Duncan, Records Custodian/Supervisor, City of Miami Police Department to Nusrat Jahan
Choudhury, Staff Attorney, ACLU, (July 28, 2015, 9:46 EST) (on file with author).

33

The U.S. Census’s 2014 1-year ACS/PUMS data does not provide a reasonable estimate of the race-specific resident driver
population of the Duval County due to a spatial mismatch between the territorial limits of the county, in which the Jacksonville
Sheriff’s Office exercises law enforcement jurisdiction, and the relevant PUMAs from which 2011 1-year ACS/PUMS data is
derived. As with the City of Tampa, the PUMA geographies relevant to Duval County fail to align closely with the county limits
in a manner that would permit estimating the race-specific population of drivers in Duval County.
34

U.S. Census Bureau; American Community Survey, American Community Survey 2011 1-Year Estimate; Table DP05;
generated by Nina Papachristou; using American FactFinder; http://factfinder2.census.gov (Jan. 19, 2016).
35

Floyd v. City of New York, 959 F. Supp. 2d 540, 557 (S.D.N.Y. 2013), dismissed (2d Cir. Sept. 25, 2013).

36

Florida Mandatory Seat Belt Laws, Florida DMV Online Guide, http://www.dmvflorida.org/seat-belt-laws.shtml (last visited
July 29, 2015); see also Fla. Stat. Ann. § 316.614 (8) (“Any person who violates the provisions of this section commits a
nonmoving violation, punishable as provided in chapter 318.”); Fla. Stat. Ann. § 318.18 (2) (listing penalty for nonmoving traffic
violations).
37

See Fla. Stat. Ann. § 318.15 (1)(a) (“If a person fails to comply with the civil penalties provided in § 318.18 within the time
period specified . . . fails to enter into or comply with the terms of a penalty payment plan with the clerk of the court . . . fails
to attend driver improvement school, or fails to appear at a scheduled hearing, the clerk of the court shall notify the
Department of Highway Safety and Motor Vehicles of such failure within 10 days after such failure. Upon receipt of such notice,
the department shall immediately issue an order suspending the driver license and privilege to drive of such person effective
20 days after the date the order of suspension is mailed . . . .”); see also Radley Balko, The Criminalization of Poverty, WASH.
POST, May 23, 2014, https://www.washingtonpost.com/news/the-watch/wp/2014/05/23/the-criminalization-of-poverty/; Sam
Sanders, Study Finds the Poor Subject to Unfair Fines, Driver’s License Suspensions, NPR, Apr. 9, 2015,
http://www.npr.org/sections/thetwo-way/2015/04/09/398576196/study-find-the-poor-subject-to-unfair-fines-drivers-licensesuspensions.
38

Rakesh Kochhar & Richard Fry, Wealth Inequality Has Widened Along Racial, Ethnic Lines Since End of Great Recession
PEW RESEARCH CENTER, Dec. 12, 2014, http://www.pewresearch.org/fact-tank/2014/12/12/racial-wealth-gaps-great-recession/.
39

Keith Morelli, Tampa Police Reviewing Death of Diabetic Driver, TAMPA BAY TIMES, Apr. 17, 2014,
http://tbo.com/news/crime/tampa-police-reviewing-officers-in-death-of-diabetic-driver-20140417/.
40

Michael S. Schmidt and Matt Apuzzo, South Carolina Officer is Charged with Murder of Walter Scott, N.Y. TIMES, Apr. 7, 2015,
http://www.nytimes.com/2015/04/08/us/south-carolina-officer-is-charged-with-murder-in-black-mans-death.html (officer
indicted for shooting Walter Scott after stopping him for a broken taillight); Richard Pérez-Peña, University of Cincinnati
Officer Indicted in Fatal Shooting of Driver, N.Y. TIMES, July 29, 2015, http://www.nytimes.com/2015/07/30/us/university-ofcincinnati-officer-indicted-in-shooting-death-of-motorist.html?emc=edit_na_20150729&nlid=53860164&ref=cta&_r=0
(Samuel Dubose shot and killed by officer during traffic stop); see, e.g., Jack Hitt, Police Shootings Won’t Stop Unless We Also
Stop Shaking Down Black People, MOTHER JONES (Sept./Oct. 2015), http://www.motherjones.com/politics/2015/07/policeshootings-traffic-stops-excessive-fines (discussing police violence during traffic stops).
41

Jaeah Lee, The Texas County Where Sandra Bland Died Is Fraught With Racial Tensions, MOTHER JONES, Jul. 17, 2015,
http://www.motherjones.com/politics/2015/07/texas-waller-county-sandra-bland-racial-tensions. Bland later died in
custody. Id.
42

Wesley Lowery, A Disproportionate Number of Black Victims in Fatal Traffic Stops, WASH. POST., Dec. 24, 2015,
https://www.washingtonpost.com/national/a-disproportionate-number-of-black-victims-in-fatal-trafficstops/2015/12/24/c29717e2-a344-11e5-9c4e-be37f66848bb_story.html.

37

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union
43

Report of Jeffrey Fagan, PhD, Floyd v. City of New York, No. 08 Civ. 01034 (SAS) 39 (S.D.N.Y. Oct. 15, 2010),
http://ccrjustice.org/files/Expert_Report_JeffreyFagan.pdf (last visited Jan. 19, 2016).
44
Police seized non-weapon contraband in only 1.8% of stops of Blacks and 1.7% of stops of Hispanics, whereas they did so in
2.3% of the stops of whites. Police seized weapons in only 1.0% of stops of Blacks and 1.1% of stops of Hispanics, whereas
they did so in 1.4% of stops of whites. See id.; Second Supplemental Report of Jeffrey Fagan, Ph.D., Floyd v. City of New York,
No. 08 Civ. 01034 (SAS) at 35 (Nov. 29, 2012), http://www.ccrjustice.org/files/FaganSecondSupplementalReport.pdf.
45

Floyd v. City of New York, 959 F. Supp. 2d 540, 561–63, 589, 667 (Aug. 12, 2013).

46

Jeffrey Fagan, et. al, Final Report, An Analysis of Race and Ethnicity Patterns in Boston Police Department Field
Interrogation, Observation, Frisk, and/or Search Reports (June 15, 2015),
https://s3.amazonaws.com/s3.documentcloud.org/documents/2158964/full-boston-police-analysis-on-race-and-ethnicity.pdf
(“[T]he analyses revealed that the percentage of Black and Hispanic residents in Boston neighborhoods were also significant
predictors of increased FIO activity after controlling for crime and other social factors.”); AMERICAN CIVIL LIBERTIES UNION OF
MASSACHUSETTS & AMERICAN CIVIL LIBERTIES UNION, BLACK, BROWN AND TARGETED: A REPORT ON BOSTON POLICE DEPARTMENT STREET
ENCOUNTERS FROM 2007-2010 7–9 (October 2014), https://aclum.org/app/uploads/2015/06/reports-black-brown-andtargeted.pdf.
47

AMERICAN CIVIL LIBERTIES UNION OF MASSACHUSETTS, ENDING RACIST STOP AND FRISK, https://aclum.org/our-work/aclumissues/racial-justice/ending-racist-stop-and-frisk/ (last visited Jan. 19, 2016).

48

Alison Manning, Boston Police Data Shows Black Men Were Stopped Most Often, BOSTON.COM (Jan. 16, 2016),
http://www.boston.com/news/local/massachusetts/2016/01/19/boston-police-data-shows-black-men-were-stopped-mostoften/4eT2npQZnhVuASnQWWPOqN/story.html.

49

AMERICAN CIVIL LIBERTIES UNION OF ILLINOIS, STOP AND FRISK IN CHICAGO 6, 9 (March 2015), http://www.aclu-il.org/wpcontent/uploads/2015/03/ACLU_StopandFrisk_6.pdf.

50

American Civil Liberties Union of Illinois, Stop and Frisk Settlement, http://www.aclu-il.org/stop-and-frisk-settlement/ (last
visited Jan. 19, 2016).

51

At Temple University, Dr. Ralph B. Taylor is a professor in the Department of Criminal Justice, and a member of the Center
for Security and Crime Science (affiliations provided for identification purposes only).
52

See Lorie Fridell, By the Numbers: A Guide for Analyzing Race Data from Vehicle Stops, Police Executive Research Forum
10 (2005), http://www.justice.utah.gov/Documents/Research/Race/PERF%20by%20the%20Numbers.pdf; Lorie A. Fridell,
Understanding Race Data From Vehicle Stops: A Stakeholder’s Guide, Police Executive Research Forum 74 (2005),
http://www.policeforum.org/assets/docs/Free_Online_Documents/RaciallyBiased_Policing/understanding%20race%20data%20from%20vehicle%20stops%20%20a%20stakeholders%20guide%202005.pdf.
53

See Fridell, Understanding Race Data From Vehicle Stops: A Stakeholder’s Guide, supra note 52 at 72–74.

54

See Fridell, Understanding Race Data From Vehicle Stops: A Stakeholder’s Guide, supra note 52 at 7 (discussing the
deficiency of using race-specific resident population figures as a benchmark against which to understand police stop data).
55

See, e.g., WEISS, supra note 23 , at 24–33; (discussing considerations relevant in developing an appropriate benchmark for
analyzing racial disparities in traffic stops); see also Fridell, By the Numbers: A Guide for Analyzing Race Data from Vehicle
Stops, supra note 51, at 7–11; 75–88.
56

Florida Highway Safety and Motor Vehicles, Cabinet and Legislature Reports & Statistics, http://www.flhsmv.gov/resourcecenter/cabinet-and-legislature-reports/ (listing “Seat Belt Violation Reports” from 2005-2014) (last visited Dec. 7, 2015).
57

FDHSMV 2014 Report, supra note 26.

58

FDHSMV 2011 Report, supra note 18.

59

U.S. Census Bureau, American Community Survey (ACS), What Is the American Community Survey?,
https://www.census.gov/programs-surveys/acs/about.html (last visited Jan. 19, 2016).
60

U.S. Census Bureau, American Community Survey (ACS), Public Use Microdata Sample (PUMS) Documentation,
https://www.census.gov/programs-surveys/acs/technical-documentation/pums.html (last visited Jan. 19, 2016).

61

Id.

62

U.S. Census Bureau; American Community Survey, 2014 1-Year Public Use Microdata Sample Estimates; generated by
Ralph B. Taylor; using American FactFinder; http://factfinder2.census.gov (Jan. 2016); U.S. Census Bureau; American
Community Survey, 2011 1-Year Public Use Microdata Sample Estimates; generated by Ralph B. Taylor; using American
FactFinder; http://factfinder2.census.gov (Jan. 2016).

38

Racial Disparities in Florida Safety Belt Law Enforcement | American Civil Liberties Union
63

U.S. Census Bureau, Geography, Public Use Microdata Areas (PUMAs), https://www.census.gov/geo/reference/puma.html
(last visited Jan. 8, 2016) (“PUMAs were initially adopted by the ACS because they were the only wall-to-wall geographic
entities below the state level that met the minimum population threshold of 65,000 required to disseminate ACS 1-year period
estimates.”).

64

U.S. Census Bureau, TIGERweb, TIGERweb Application, http://tigerweb.geo.census.gov/tigerwebmain/TIGERweb_apps.html
(last visited Jan. 8, 2016).

65

The U.S. Census Bureau’s geographic definition for PUMAs changed between the 2000 and 2010 census. U.S. Census
Bureau, Geography, Summary of Changes to the PUMA Criteria and Guidelines from 2000 to 2010,
https://www.census.gov/geo/reference/puma_changes.html (last visited Jan. 8, 2016). The 2010 PUMA codes (“PUMA2K12”)
were used to identify the PUMAs corresponding to Palm Beach County, Orange County, and Broward County in order to permit
estimation of the number of race-specific residents of driving age and living in a household with access to a vehicle in each
county from the 2014 1-year ACS/PUMS data files. The 2000 PUMA codes (“PUMA2k”) were used to obtain 2011 1-year
ACS/PUMS estimates for the number of race-specific residents of driving age and living in a household with access to a
vehicle in Escambia County because the 2010 PUMA codes did not go into effect until 2012.
66

This report uses the term “accurate” to mean appropriately taking into account sampling errors; it does not consider nonsampling errors.

67

U.S. Census Bureau, PUMS Accuracy of the Data 12 (2014), http://www2.census.gov/programssurveys/acs/tech_docs/pums/accuracy/2014AccuracyPUMS.pdf (“Replicate weights can be used to approximate what we refer
to as direct standard errors. Standard errors for the published ACS tabulations are calculated using a replicate weights
method. Direct standard errors will often be more accurate than generalized standard errors, although they may be more
inconvenient for some users to calculate.”).
68

D.R. Judkins, Fay’s method for variance estimation, Journal of Official Statistics 6: 223–239 (1990); P.J. McCarthy,
Replication: An approach to the analysis of data from complex surveys., Vital and Health Statistics, series 2. Hyattsville, MD:
National Center for Health Statistics (1966).
69

Stata. Survey Data Reference Manual: Release 14. College Station, TX: Statacorp (2015).

39