Skip to main content

Full text of "DTIC AD1046118: The Ferguson Effect - Are Police Anxieties To Blame"

See other formats


NAVAL 

POSTGRADUATE 

SCHOOL 

MONTEREY, CALIFORNIA 


THESIS 


THE FERGUSON EFFECT—ARE POLICE ANXIETIES 

TO BLAME? 

by 


Stephen Edward Simonds Jr. 


March 2017 


Thesis Co-Ad visors: 

Erik Dahl 
Patrick Miller 


Approved for public release. Distribution is unlimited. 



THIS PAGE INTENTIONALLY LEFT BLANK 



REPORT DOCUMENTATION PAGE 


Form Approved OMB 
No. 0704-0188 


Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing 
instruction, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection 
of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including 
suggestions for reducing this burden, to Washington headquarters Services, Directorate for Information Operations and Reports, 1215 
Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork 
Reduction Project (0704-0188) Washington DC 20503. 


1. AGENCY USE ONLY 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED 

(Leave blank) March 2017 Master’s thesis 


4. TITLE AND SUBTITLE 5. FUNDING NUMBERS 

THE FERGUSON EFFECT—ARE POLICE ANXIETIES TO BLAME? 


6. AUTHOR(S) Stephen Edward Simonds Jr. 


7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 

Naval Postgraduate School 
Monterey, CA 93943-5000 


9. SPONSORING /MONITORING AGENCY NAME(S) AND 
ADDRESS(ES) 

N/A 


8. PERFORMING 
ORGANIZATION REPORT 
NUMBER 


10. SPONSORING/ 
MONITORING AGENCY 
REPORT NUMBER 


11. SUPPLEMENTARY NOTES The views expressed in this thesis are those of the author and do not reflect the 
official policy or position of the Department of Defense or the U.S. Government. IRB number_N/A_. 


12a. DISTRIBUTION / AVAILABILITY STATEMENT 

Approved for public release. Distribution is unlimited. 


13. ABSTRACT (maximum 200 words) 


12b. DISTRIBUTION CODE 

A 


The “Ferguson Effect” is a relatively recent and controversial theory suggesting law enforcement 
officers across the country have become less proactive in their policing efforts following the August 2014 
officer-involved shooting of Michael Brown in Ferguson, Missouri. This thesis attempts to settle the 
Ferguson Effect debate by determining whether open-source data about police productivity can be 
collected and analyzed either to support or contradict the Ferguson Effect. 

Publicly available data repositories through the Public Safety Open Data Portal, Public Data Initiative, 
and related governmental links are utilized for raw dataset acquisition. Three agencies are chosen for data 
collection and analysis: (1) the Burlington Police Department, Burlington, Vermont, (2) the Montgomery 
County Police Department, Montgomery County, Maryland, and (3) the Philadelphia Police Department, 
Philadelphia, Pennsylvania. In each case study, the data is insufficient to confirm or deny the existence of 
the Ferguson Effect, although the limited data available does suggest that in these three cities, no 
noticeable de-policing is detected following the killing of Michael Brown. 

The conclusion of the study yields several limitations. The most evident deficiency identified in this 
study involves the transparency initiative and Open Government program, specifically with regard to the 
Public Data Initiative (PDI) and gathering of police-related data. The PDI needs to establish stricter 
guidelines and compliance for participating agencies. Additionally, this thesis suggests that less emphasis 
should be placed on crime correlations and more value placed on de-policing and anxieties being 
experienced by officers to measure accurately the existence of a Ferguson Effect. 


15. NUMBER OF 
PAGES 

73 


16. PRICE CODE 


20. LIMITATION 
OF ABSTRACT 


14. SUBJECT TERMS 


-1 

Ferguson Effect; de-policing; free-floating anxiety; Michael Brown; Freddie Gray; Ferguson, 
Missouri; community partnership; organizational justice; self-legitimacy; police productivity 

17. SECURITY 

18. SECURITY 

19. SECURITY 

CLASSIFICATION OF 

CLASSIFICATION OF THIS 

CLASSIFICATION OF 

REPORT 

PAGE 

ABSTRACT 

Unclassified 

Unclassified 

Unclassified 


NSN 7540-01-280-5500 


1 


Standard Form 298 (Rev. 2-89) 
Prescribed by ANSI Std. 239-18 




























THIS PAGE INTENTIONALLY LEFT BLANK 


11 



Approved for public release. Distribution is unlimited. 


THE FERGUSON EFFECT—ARE POLICE ANXIETIES TO 

BLAME? 


Stephen Edward Simonds Jr. 

Lieutenant, St. Mary’s County Sheriff’s Office, Leonardtown, Maryland 
B.S., Towson University, 1997 


Submitted in partial fulfillment of the 
requirements for the degree of 


MASTER OF ARTS IN SECURITY STUDIES 
(HOMELAND SECURITY AND DEFENSE) 

from the 

NAVAL POSTGRADUATE SCHOOL 
March 2017 


Approved by: Erik Dahl 

Thesis Co-Advisor 


Patrick Miller 
Thesis Co-Advisor 


Erik Dahl 

Associate Chair for Instruction 
Department of National Security Affairs 



THIS PAGE INTENTIONALLY LEFT BLANK 


IV 



ABSTRACT 


The “Ferguson Effect” is a relatively recent and controversial theory suggesting 
law enforcement officers across the country have become less proactive in their policing 
efforts following the August 2014 officer-involved shooting of Michael Brown in 
Ferguson, Missouri. This thesis attempts to settle the Ferguson Effect debate by 
determining whether open-source data about police productivity can be collected and 
analyzed either to support or contradict the Ferguson Effect. 

Publicly available data repositories through the Public Safety Open Data Portal, 
Public Data Initiative, and related governmental links are utilized for raw dataset 
acquisition. Three agencies are chosen for data collection and analysis: (1) the Burlington 
Police Department, Burlington, Vermont, (2) the Montgomery County Police 
Department, Montgomery County, Maryland, and (3) the Philadelphia Police 
Department, Philadelphia, Pennsylvania. In each case study, the data is insufficient to 
confirm or deny the existence of the Ferguson Effect, although the limited data available 
does suggest that in these three cities, no noticeable de-policing is detected following the 
killing of Michael Brown. 

The conclusion of the study yields several limitations. The most evident 
deficiency identified in this study involves the transparency initiative and Open 
Government program, specifically with regard to the Public Data Initiative (PDI) and 
gathering of police-related data. The PDI needs to establish stricter guidelines and 
compliance for participating agencies. Additionally, this thesis suggests that less 
emphasis should be placed on crime correlations and more value placed on de-policing 
and anxieties being experienced by officers to measure accurately the existence of a 
Ferguson Effect. 


v 



THIS PAGE INTENTIONALLY LEFT BLANK 


vi 



TABLE OF CONTENTS 


I. INTRODUCTION.1 

A. RESEARCH QUESTION AND HYPOTHESIS.3 

B. PURPOSE OF THE STUDY.3 

C. METHODOLOGY.4 

II. LITERATURE REVIEW.7 

A. INTRODUCTION.7 

B. WHAT IS THE FERGUSON EFFECT?.8 

C. DEBATING THE FERGUSON EFFECT.10 

D. OPPONENTS OF THE FERGUSON EFFECT.16 

E. CONCLUSION.19 

III. METHODOLOGY.21 

A. PROCEDURES.22 

B. DATA SETS.23 

C. CASE STUDIES.25 

D. DATA ANALYSIS.26 

IV. RESULTS AND ANALYSIS.27 

A. CASE STUDY 1—BURLINGTON POLICE DEPARTMENT, 

BURLINGTON, VERMONT.28 

1. Variable 1—Traffic Stops.29 

2. Variable 2—Arrests.30 

3. Variable 3—Violent Crime.31 

B. CASE STUDY 2—MONTGOMERY COUNTY POLICE 

DEPARTMENT, MONTGOMERY COUNTY, MARYLAND.32 

1. Variable 1—Traffic Stops.32 

2. Variable 2—Arrests.33 

3. Variable 3—Violent Crime.34 

C. CASE STUDY 3—PHILADELPHIA POLICE DEPARTMENT, 

PHILADELPHIA, PENNSYLVANIA.35 

1. Variable 1—Traffic Stops.35 

2. Variable 2—Arrests.37 

3. Variable 3—Violent Crime.37 

D. CASE STUDY COMPARISON.38 


vii 
































V. DISCUSSION.39 

A. LIMITATIONS AND WEAKNESSES.40 

B. RECOMMENDATIONS.41 

C. CONCLUSION.42 

LIST OF REFERENCES.45 

INITIAL DISTRIBUTION LIST.51 


viii 









LIST OF FIGURES 


Figure 1. Scope of Data—Two Years Before and After Ferguson Incident.23 

Figure 2. Burlington Police Department, Variable 1—Traffic Stops (August 

2012-December 2015).29 

Figure 3. Burlington Police Department, Variable 2—Arrests (August 2012- 

August 2016).30 

Figure 4. Burlington Police Department, Variable 3—Violent Crime (August 

2012- August 2016).31 

Figure 5. Montgomery County Police Department, Variable 1—Traffic Stops 

(January 2013-August 2016).33 

Figure 6. Montgomery County Police Department, Variable 3—Violent Crime 

(July 2013-August 2016).34 

Figure 7. Philadelphia Police Department, Variable 1—Traffic Stops (July 

2013- August 2016).36 

Figure 8. Philadelphia Police Department, Variable 3—Violent Crime (August 

2012-August 2016).37 











THIS PAGE INTENTIONALLY LEFT BLANK 


x 



LIST OF TABLES 


Table 1. Desired Variables.23 

Table 2. Case Study Departments and Obtainable Variables.27 


xi 





THIS PAGE INTENTIONALLY LEFT BLANK 



LIST OF ACRONYMS AND ABBREVIATIONS 


DEA 

DOJ 

FBI 

FOIA 

NIBRS 

PDI 

SRS 

UCR 


Drug Enforcement Administration 

Department of Justice 

Federal Bureau of Investigation 

Freedom of Information Act 

National Incident Based Reporting System 

Police Data Initiative 

Summary Reporting System 

uniform crime reporting 



THIS PAGE INTENTIONALLY LEFT BLANK 


xiv 



EXECUTIVE SUMMARY 


The “Ferguson Effect” is a relatively recent and controversial theory suggesting 
law enforcement officers across the country have become less proactive in their policing 
efforts following the August 2014 officer-involved shooting of Michael Brown in 
Ferguson, Missouri. Many advocates of the Ferguson Effect believe violent crime rates 
have significantly increased in many major U.S. cities because of the void created by the 
withdrawal of police activity. 

If police are becoming more hesitant in taking proactive approaches, logic would 
suggest these new practices should be measureable through a visible reduction in police 
productivity and quantifiable outputs. This thesis attempts to settle the Ferguson Effect 
debate by asking if open source data concerning police productivity can be collected and 
analyzed either to support or contradict the Ferguson Effect. If so, what would that data 
show? 

The goal of the literature review is to acquire, review, and analyze available 
materials relating to the Ferguson Effect phenomenon. During the course of conducting 
the review, it has become evident that subsections of the public, government, and media 
differ in their acceptance or rejection of the theory. Additionally, the theory has become 
highly politicized and is an issue of debate amongst government officials. Despite 
varying degrees of accepted legitimacy, most literature conceded additional data was still 
needed to establish a scholarly conclusion. 

The research sets out to resolve the Ferguson Effect debate by conducting three 
case studies utilizing trend analysis on law enforcement open source quantitative datasets. 
Publicly available data repositories through the Public Safety Open Data Portal, Public 
Data Initiative (PDI) and related governmental links are utilized for raw dataset 
acquisition. The scope of data sought includes at least two years before and after the 
shooting of Michael Brown. 

Fimitations in available datasets ultimately determined which three agencies were 
chosen for collection and analysis: (1) the Burlington Police Department, Burlington, 


xv 



Vermont, (2) the Montgomery County Police Department, Montgomery County, 
Maryland, and (3) the Philadelphia Police Department, Philadelphia, Pennsylvania. 

Initial data values seek productivity including information relating to traffic stops, 
pedestrian stops, arrests, call response times, use of sick leave, new applicants, hires, and 
retirees. Obtaining this data proved challenging through the PDI. On the other hand, 
crime related data was slightly more accessible, although each agency varied in reporting 
content, organization, and classification of information. 

In each case study, data is insufficient to confirm or deny the existence of the 
Ferguson Effect, although the limited data available does suggest that in these three 
cities, no noticeable de-policing is detected following the killing of Michael Brown. 

The conclusion of the study yields several limitations. The most evident 
deficiency involves the transparency initiative and Open Government program, 
specifically with regard to the PDI and gathering of police related data. The PDI needs to 
establish stricter guidelines and compliance for participating agencies. For those law 
enforcement agencies wishing to participate, minimum guidelines should be established 
and upheld. Additionally, those agencies not meeting those standards should be 
eliminated from participating in the program. 

To date, conversations involving the Ferguson Effect are usually associated with 
crime statistics. This thesis suggests that less emphasis should be placed on crime 
correlations and more value placed on de-policing and anxieties being experienced by 
officers to measure accurately the existence of a Ferguson Effect. Despite the large 
amount of anecdotal evidence suggesting law enforcement is becoming less proactive, the 
limited data available does not seem to show a significant amount of de-policing 
following the events in Ferguson. More data is urgently needed to help determine 
whether de-policing is in fact a widespread trend. 

Theoretically, big data should offer possible correlations between police 
productivity and the increase or reduction of crime. Fikewise, data relating to 
productivity should offer independent insight into the existence of a Ferguson Effect 
regardless of reported crime. The reality is that access to this type of big data is not 



readily available to the public. If the open data project matures into its intended vision, 
such data may provide insight into the existence of a Ferguson Effect. Currently, 
however, the data is insignificant and lends no insight whatsoever. 

This study aspires to close previous gaps and offer a more complete picture of 
national sentiment by employing big data analytics for police productivity. Limitations in 
data prevented the research questions from being answered but highlighted a need for 
better data collection methods. 


XVII 



THIS PAGE INTENTIONALLY LEFT BLANK 



ACKNOWLEDGMENTS 


I would like to thank my wonderful wife, Susan, for her continued support and 
encouragement during this academic journey. I would also like to thank my children, 
Logan, Hunter, Grace, and Valerie, for their understanding of the time spent on studying 
over family activities. I would like to thank the Center for Homeland Defense and 
Security, Naval Postgraduate School, for affording me the opportunity to participate in 
what has truly been a life-changing experience. Lastly, I would like to extend my 
gratitude to Sheriff Timothy K. Cameron, St. Mary’s County Sheriffs Office, for his 
endorsement and continued support throughout the program. 



THIS PAGE INTENTIONALLY LEFT BLANK 


xx 



I. 


INTRODUCTION 


The “Ferguson Effect” is a relatively recent and controversial theory suggesting 
law enforcement officers across the country have become less proactive in their policing 
efforts following the August 2014 officer-involved shooting of Michael Brown in 
Ferguson, Missouri. Many advocates of the Ferguson Effect believe violent crime rates 
have significantly increased in many major U.S. cities as a result of the void created by 
the withdrawal of police activity. Until new crime data became available recently, the 
Ferguson Effect resulted in two levels of debate. Many critics of the concept questioned 
whether in fact a rise in violent crime had occurred, particularly with regard to homicide 
in some major cities post-Ferguson. This part of the debate was initially driven by a lack 
of available crime data on a national level for the year 2015. For those who 
acknowledged violent crime had spiked, the second level of debate concerned identifying 
possible causes for the increase and whether it could be linked to less aggressive policing. 

In September 2016, the Federal Bureau of Investigation (FBI) released new data 
from the Uniform Crime Reporting (UCR) Program, which contained data through the 
end of 2015. The findings indicate, “Crime in the United States, 2015 reveals a 3.9 
percent increase in the estimated number of violent crimes and a 2.6 percent decrease in 
the estimated number of property crimes last year when compared to 2014 data.” 1 This 
data appears to establish that a rise in violent crime on a national level has in fact 
occurred. However, a larger debate still remains, and is the subject of this thesis: Is the 
Ferguson Effect responsible for the spike in violent crime? 

The recent change in violent crime may or may not be a consequence of the 
Ferguson Effect. Alternate theories, such as the impact of illegal drugs within 
communities, have been suggested as competing reasons for the spike. Regardless, crime 
itself should not be a prerequisite in determining if police are experiencing increased 
anxieties while performing sworn duties. If police are becoming hesitant in their 


1 “Latest Crime Statistics Released, Increase in Violent Crime, Decrease in Property Crime,” 
September 26, 2016, https://www.fbi.gov/news/stories/latest-crime-statistics-released. 


1 



proactivity, logic would suggest these new practices should be measureable through a 
visible reduction in police productivity and quantifiable outputs. 

A number of documented instances are available in which individual officers are 
second guessing their actions for fear of adverse publicity. This development was 
recognized as early as August 2015 when a male police officer in Birmingham, Alabama, 
was pistol whipped unconscious with his own gun by a felon he had pulled over on a 
traffic stop. The officer told CNN that he hesitated to use force against the driver because 
he “didn’t want to be in the media like I am right now.” 2 A more recent example involves 
a veteran female Chicago police officer who was nearly beaten to death by a man on PCP 
on October 5, 2016. The officer stated she did not fire her weapon at the suspect because 
“she didn’t want her family or the department to have to go through the scrutiny the next 
day on national news.” 3 

Existing research, however, is largely inconclusive as to whether a Ferguson 
Effect is having a negative impact on police productivity. Some studies have examined 
the correlation between the number of arrests conducted and the level of violent crime 
within particular areas. Although helpful, this method does not address the question of 
whether police have become less proactive in general. Further opportunity remains to 
evaluate whether police are experiencing greater apprehension in performing their duties, 
and what, if any, long-term consequences this apprehension may pose. If police are 
becoming less proactive after Ferguson, then data relating to self-initiated productivity 
should demonstrate a relative decline. Research conducted thus far has not addressed 
police productivity independent of its presumed correlation with crime. As such, the 
possibility that the Ferguson Effect may be largely impacting areas that have not 
experienced a rise in violent crime has not been fully considered. 


2 Heather Mac Donald, “Officer Beaten by a Convicted Felon Hesitated for Fear of Being Called 
Racist: Welcome to Post-Ferguson Policing,” National Review , August 16, 2105, http://www.national 
review.com/article/422605/Birmingham-cop-beaten-unconscious-feared-racism-charge. 

3 Stephen Gossett, “Top Cop Says Beaten Officer Was Afraid to Shoot Due to Media Attention,” 
Chicagoist, October 6, 2016, http://chicagoist.com/2016/10/06/top_cop_says_beaten_officer_was_afr.php. 


2 



A. RESEARCH QUESTION AND HYPOTHESIS 

This thesis attempts to settle the Ferguson Effect debate using open source data 
analysis. The study examines the prospect that police productivity related data can be 
collected, analyzed, and correlated either to support or contradict the Ferguson Effect. 

The following research questions are addressed in this study. 

• Can open source data concerning police productivity be collected and 
analyzed either to support or contradict the Ferguson Effect? If so, what 
does that data show? 

• Has police productivity decreased overall since August 2014? 

• What, if any, impact does a reduction in police productivity have on 
violent crime? 

• What future challenges do law enforcement administrators face if the 
Ferguson Effect is deemed viable? 

The hypothesis of this thesis is that open source data supports a notable reduction 
in police productivity following the events in Ferguson. To a lesser degree, the researcher 
expects to find that these reductions in police activity are correlated with sporadic 
increases in violent crime within some major U.S. cities post-Ferguson. 

B. PURPOSE OF THE STUDY 

What is the homeland security value in settling the debate of whether the 
Ferguson Effect is to blame for the rise in violent crime? The topic has already 
established itself as a homeland security issue due to public perceptions, media coverage, 
and perpetuating mass protests, many of which have resulted in violence and police 
fatalities. The value of conducting this research lends to identifying the root causes for 
increased crime and determining if officers are truly becoming less proactive in their 
policing as a result of the events sparked in Ferguson. 

The researcher’s law enforcement experience suggests a substantial percentage of 
officers across the nation are informally adopting reactive policing practices in place of 
proactive methods. Take for instance the example an officer who drives by a street comer 
in a known drug area and sees what could possibly be a hand-to-hand drug transaction. 


3 



Pre-Ferguson, the officer may opt to investigate by stopping and conducting field 
interviews with the subjects in question. Post-Ferguson, the same officer may choose to 
continue driving by without further investigation. This action is not to suggest officers are 
not responding to calls for service; i.e., someone calls in a loitering complaint with 
possible drug dealing on the same corner and the officer responds to address the 
complaint. However, if it is true that officers are becoming less proactive because of 
higher levels of anxiety, the matter should be resolved before such reductions in policing 
lend to a rise in crime. 

C. METHODOLOGY 

This thesis attempts to gather and analyze empirical data to support or contradict 
the Ferguson Effect, and to examine the implications of this theory as it relates to 
homeland security and safety. 

The Taskforce on 21st Century Policing resulted in development of the Police 
Data Initiative (PDI), which “focuses on generating and implementing new data and 
technology innovations within key jurisdictions, civil society groups, and federal, state, 
and local agencies.” 4 As a result of the initiative, many law enforcement agencies now 
share statistical data to include calls for service, traffic stops, use of force incidents, and 
arrests, to name a few. Currently, 120 law enforcement agencies participate in the sharing 
of statistical data. Information submitted by participating agencies can be accessed 
through the Public Safety Data Portal. 5 Testing the Ferguson Effect Theory using the 
described methodology depends on the ability to access large amounts of productivity 
related information through the PDI. 

In this thesis, PDI performance-related datasets are analyzed and graphed in 
attempt to identify patterns and associations visually before and after the incident in 
Ferguson. Additionally, violent crime data for the same selected jurisdictions is compared 


4 “The Police Data Initiative (PDI),” accessed November 18, 2016, https://publicsafetydatapoital.org/ 
participating-agencies/. 

5 “Public Safety Open Data Portal,” accessed October 9, 2016, http://publicsafetydatapoital.org/. 

4 



and contrasted with acquired productivity data. The data visualization software, “Tableau 
for Students,” 6 is utilized for this project. 

Chapter II, Literature Review, discusses existing literature surrounding the 
Ferguson Effect including influence from multiple perspectives. Available studies are 
reviewed, compared, and contrasted. Chapter III, Methodology, explains procedures 
utilized within the study, to include desired productivity variables, case study selection 
factors, timeline parameters, data selection, and data analysis. Chapter IV, Results and 
Analysis, describes the agencies selected for the case study, results of the data, and visual 
trend analysis. Chapter V, Discussion, concludes with a discussion of case study 
limitations and recommendations for future research. 


6 “Tableau for Students,” accessed August 15, 2016, http://www.tableau.com/academic/students. 

5 



THIS PAGE INTENTIONALLY LEFT BLANK 


6 



II. LITERATURE REVIEW 


A. INTRODUCTION 

The “Ferguson Effect” is a recent and controversial concept suggesting law 
enforcement officers across the country, and especially in larger cities, have become less 
proactive in public safety efforts following the August 2014 officer-involved shooting in 
Ferguson, Missouri. Many supporters of the Ferguson Effect believe violent crime rates 
have been directly and negatively impacted as a result of the void created by the 
withdrawal or absence of police presence in many communities. 

Fiterature surrounding the Ferguson Effect continues to develop but is still limited 
due to the recentness of the event and because much of the terminology used is still 
evolving. Referenced materials are drawn from online sources to include newspaper 
articles, editorials, magazine articles, and blog commentaries. Additionally, academic 
studies and governmental papers are reviewed and analyzed. More value is placed on 
academic studies and news reports over editorials and commentaries. 

The primary search term, “Ferguson Effect,” is utilized to acquire digital 
information through a variety of channels to include open source online searches, Google 
Scholar, the Naval Postgraduate School Dudley Knox Fibrary, and ProQuest. Materials 
gathered for study include data from August 2014 through January 2017. 

The goal of the literature review is to acquire, review, and analyze available 
materials relating to the Ferguson Effect phenomenon. During the course of conducting 
this review, it became evident that subsections of the public, government, and media 
differ in their acceptance or rejection of the theory. Additionally, the theory has become 
highly politicized and is an issue of debate amongst government officials. Despite 
varying degrees of accepted legitimacy, most literature conceded additional data was still 
needed to establish a scholarly conclusion. Subsequently, the deficiency of scholarly 
study presents a unique opportunity for empirical research towards either substantiating 
or discrediting the notion of a Ferguson Effect. 


7 



B. WHAT IS THE FERGUSON EFFECT? 

According to a New York Times report in August 2015, “at least 35 of the nation’s 
cities [were] reporting increases in murders, violent crimes or both.” 7 One city that 
experienced an uptick in violent crime following the events in Ferguson, Missouri, was 
St. Louis, which experienced a 60% increase in homicides from 2014 to 2015. 8 St. Louis 
police Chief Sam Dotson recognized the spike early on and provided the following 
explanation during an interview with a reporter in November 2014, “It’s the Ferguson 
effect ... I see it not only on the law enforcement side, but the criminal element is feeling 
empowered by the environment.” 9 Specifically, the environmental change Chief Dotson 
was referring to within St. Louis was a result of diminished police presence due to 
officers being diverted from normal assignments to assist Ferguson along with receiving 
additional civil unrest training. Chief Dotson has since been credited with the coining of 
the phrase, “Ferguson Effect.” His description of the term primarily deals with a noticed 
observation and suspected cause. At least in St. Louis, he believed the spike in violent 
crime occurred because fewer officers were on the street to deter crime and make arrests. 
It should be noted in the case of St. Louis that the proactive behavior of officers was 
circumstantial and not necessarily intentional. 

Milwaukee Police Chief Edward Flynn presented a similar concept, “free-floating 
anxiety,” to articulate similar concerns about officers becoming less proactive. Flynn 
acknowledged a widespread angst being experienced within his own department and 
extended the belief to other officers across the nation. Flynn differs from Dotson in that 
he addresses an anxiety or unwillingness to be proactive, not because resources have been 
exhausted, but because officers are making a conscious decision to be less proactive. He 
advised, “There’s no denying that antipolice protests and viral videos that emerged after 
the fatal shooting of an unarmed black teenager in Ferguson, Mo., have unnerved the 

7 Monica Davey and Mitch Smith, “Murder Rates Rising Sharply in Many U.S. Cities,” New York 
Times , August 31, 2015, http://nyti.ms/lUiHLnX. 

8 Ibid. 

9 Christin Byers, “Crime Up after Ferguson and More Police Needed, Top St. Louis Area Chiefs Say,” 
St. Louis Post Dispatch, November 15, 2014, http://www.stltoday.com/news/local/crime-and-courts/crime- 
up-after-ferguson-and-more-police-needed-top-st/article_04d9f99f-9a9a-5 Ibe-a231-1707a57b50d6.html. 


8 



rank and file.” 10 His conviction stems from observing a decline in traffic stops and field 
interviews being performed by his officers in Milwaukee, despite police still “responding 
to crimes as they always have.” 11 While Flynn acknowledges increased officer anxieties 
and their impact on proactive policing, he did not believe they were the primary cause of 
increased crime in Milwaukee. Instead, Flynn faulted, “state budget cuts that reduced 
mental health services and other social programs, and a 2011 law that dramatically 
weakened the power of police to arrest people with guns,” to be the actual culprits. 12 
Traffic stops and field interviews are just two examples of self-initiated activities 
conducted by police, but this kind of policy activity often appears to have been reduced 
following the events in Ferguson. Even the mayor of Ferguson, James Knowles III, said, 
“We barely pull anybody over anymore,” proposing “a sharp decline in traffic stops 
might actually be a sign that the police are afraid to do their job.” 13 

Some have used a completely different definition of the Ferguson Effect, 
believing the term “more accurately describes a political movement where oppressed 
black Americans are increasingly standing up for their rights.” 14 For the purposes of this 
thesis, however, the more widely accepted description of the Ferguson Effect is used, 
meaning that police have become less proactive. Moreover, this thesis more specifically 
approaches the phenomenon from Flynn’s perspective, arguing that increased anxiety 
among law enforcement officers has lessened their proactiveness. The term often used to 
describe this practice is “de-policing.” 


10 Darryl Fears, “In Milwaukee, Weak Evidence for ‘Ferguson Effect’,” The Washington Post, 
December 6, 2015, http://search.proquest.com/docview/17393457767accountidM2702. 

11 Davey and Smith, “Murder Rates Rising Sharply in Many U.S. Cities.” 

12 Fears, “In Milwaukee, Weak Evidence for ‘Ferguson Effect’.” 

13 M. International, N. York, and T. Paris, “A Year Later, Ferguson Remains Unsettled: Residents 
Dispute Extent of Changes Following Racially Charged Death,” 1M, 2015, http://search.proquest.com/ 
do c vie w/1701637793? account id= 12702. 

14 Patrik Johnson, “Just What is the ‘Ferguson Effect’? It Depends on How You View Police,” The 
Christian Science Monitor, June 12, 2015, http://www.csmonitor.com/USA/Justice/2015/0612/Just-what- 
is-the-Ferguson-effect-It-depends-on-how-you-view-police. 


9 



C. DEBATING THE FERGUSON EFFECT 

To date, only a handful of scholarly articles have been written on the Ferguson 
Effect, and most of this research has focused on the impact of the events in Ferguson on 
the crime rate. One early study entitled, “Was There a ‘Ferguson Effect’ on Crime in St. 
Fouis?,” was written by Richard Rosenfeld, PhD, University of Missouri—St. Fouis, and 
released through the Sentencing Project in June 2015. 15 Rosenfeld analyzed crime 
statistics for the city of St. Fouis surrounding the timeframe Michael Brown was killed. 
His findings are based on tallying and comparing crime statistics in St. Fouis before and 
after the Ferguson incident. With regard to homicide, Rosenfeld’s study concluded “with 
reasonable certainty that the events in Ferguson were not responsible for the steep rise in 
homicide in St. Fouis.” 16 That conclusion was based on the finding, “The increase in 
homicide in 2014 predated Michael Brown’s killing on August 9.” 17 Rosenfeld noted, “If 
there was a Ferguson effect on crime in St. Fouis, it was most pronounced in the growth 
of property crimes,” warning, “temporal consistency is not a sufficient condition to 
establish substantive proof.” 18 Absent in the study were comparison statistics of 
proactive policing data for St. Fouis before and after Ferguson, to include traffic stops 
and other self-initiated calls for service. 

On October 23, 2015, FBI Director James Comey delivered a speech at the 
University of Chicago Faw School addressing his concerns for the continued rise in 
violent crime and possible explanations. Although he does not mention the Ferguson 
Effect by name, this portion of Comey’s speech articulates the crux of the theory, and is 
worth quoting at length. 

Most of America’s 50 largest cities have seen an increase in homicides 

and shootings this year, and many of them have seen a huge increase. 

These are cities with little in common except being American cities— 


16 Richard Rosenfeld, Was There a “Ferguson Effect” on Crime in St. Louis? (Washington, DC: The 
Sentencing Project, 2015), http://sentencingproject.org/wp-content/uploads/2015/09/Ferguson-Effect.pdf. 

16 Ibid. 

17 Ibid. 

18 Ibid. 


10 



places like Chicago, Tampa, Minneapolis, Sacramento, Orlando, 
Cleveland, and Dallas. 

In Washington, D.C., we’ve seen an increase in homicides of more than 20 
percent in neighborhoods across the city. Baltimore, a city of 600,000 
souls, is averaging more than one homicide a day—a rate higher than that 
of New York City, which has 13 times the people. Milwaukee’s murder 
rate has nearly doubled over the past year. 

And who’s dying? 

Police chiefs say the increase is almost entirely among young men of 
color, at crime scenes in bad neighborhoods where multiple guns are being 
recovered. 

That’s yet another problem that white America can drive around, but if we 
really believe that all lives matter, as we must, all of us have to understand 
what is happening. 

Communities of color need to demand answers. 

Police and civilian leaders need to demand answers. 

Academic researchers need to hit this hard. 

What could be driving an increase in murder in some cities across all 
regions of the country, all at the same time? What explains this map and 
this calendar? Why is it happening in all of different places, all over and 
all of a sudden? 

I’ve been part of a lot of thoughtful conversations with law enforcement, 
elected officials, academics, and community members in recent weeks. 
I’ve heard a lot of theories—reasonable theories. 

Maybe it’s the return of violent offenders after serving jail terms. Maybe 
it’s cheap heroin or synthetic drugs. Maybe after we busted up the large 
gangs, smaller groups are now fighting for turf. Maybe it’s a change in the 
justice system’s approach to bail or charging or sentencing. Maybe 
something has changed with respect to the availability of guns. 

These are all useful suggestions, but to my mind none of them explain 
both the map and the calendar in disparate cities over the last 10 months. 

But I’ve also heard another explanation, in conversations all over the 
country. Nobody says it on the record, nobody says it in public, but police 
and elected officials are quietly saying it to themselves. And they’re 
saying it to me, and I’m going to say it to you. And it is the one 


11 



explanation that does explain the calendar and the map and that makes the 
most sense to me. 

Maybe something in policing has changed. 

In today’s YouTube world, are officers reluctant to get out of their cars 
and do the work that controls violent crime? Are officers answering 911 
calls but avoiding the informal contact that keeps bad guys from standing 
around, especially with guns? 

I spoke to officers privately in one big city precinct who described being 
surrounded by young people with mobile phone cameras held high, 
taunting them the moment they get out of their cars. They told me, “We 
feel like we’re under siege and we don’t feel much like getting out of our 
cars.” 

I’ve been told about a senior police leader who urged his force to 
remember that their political leadership has no tolerance for a viral video. 

So the suggestion, the question that has been asked of me, is whether these 
kinds of things are changing police behavior all over the country. 

And the answer is, I don’t know. I don’t know whether this explains it 
entirely, but I do have a strong sense that some part of the explanation is a 
chill wind blowing through American law enforcement over the last year. 

And that wind is surely changing behavior. 19 

Director Comey indirectly refers to the Ferguson Effect as being a “chill wind 
blowing through American law enforcement over the last year,” 20 which is most closely 
related to Flynn’s concept of “free floating anxiety.” These types of occurrences and the 
labels given to them are not new and did not begin with Ferguson. In a recent study 
addressing “de-policing,” Stephen Rushin and Griffin Edwards write that “Within the 
academic literature and public discourse, this phenomenon has taken on many different 
names including ‘passive law enforcement,’ ‘selective disengagement,’ ‘tactical 
detachment,’ or officer ‘retreat.” 21 Their study used a “panel of American law 


19 James B. Comey, “Law Enforcement and the Communities We Serve: Bending the Lines toward 
Safety and Justice” (speech. University of Chicago Law School, Chicago, Illinois, October 23, 2015), 
https://www.fbi.gov/news/speeches/law-enforcement-and-the-communities-we-serve-bending-the-lines- 
toward-safety-and-justice. 

20 Ibid. 

21 Stephen Rushin and Griffin Edwards, “De-Policing,” DRAFT, Cornell Law Review 102 
(forthcoming, 2017): 13. 


12 



enforcement agencies and difference-in-difference regression analyses,” to examine 
“whether the introduction of public scrutiny and external regulation is associated with 
changes in crime rates.” 22 The study indicated the “de-policing phenomenon is not 
entirely implausible,” and “represent[s] an important recognition of the possible negative 
side effects associated with external regulation of American law enforcement.” 23 They 
make it a point to add their findings “should not necessarily deter policymakers from 
enacting regulations” upon local law enforcement. 24 

The notion of the Ferguson Effect being connected with increased crime partially 
stems from a May 29, 2015, opinion piece in The Wall Street Journal written by 
Manhattan Institute Fellow Heather Mac Donald. Mac Donald noted a rise in violent 
crime for the months following Ferguson in such cities as Baltimore, Milwaukee, St. 
Fouis, Atlanta, Chicago, Fos Angeles, and New York. Mac Donald also reported the rise 
in violent crime began during the second half of 2014, following the shooting of Michael 
Brown. 25 Mac Donald suspected correlations between the two, suggesting, “the most 
plausible explanation of the current surge in lawlessness is the intense agitation against 
American police departments over the past nine months,” adding the personal view, 
“acquittals of police officers for the use of deadly force against black suspects are now 
automatically presented as a miscarriage of justice.” 26 

A year after her initial observations, Mac Donald wrote a follow-up op-ed, “The 
Nationwide Crime Wave Is Building,” with the subtitle, “As the homicide rate keeps 
rising in many cities, even some who dismissed the ‘Ferguson effect’ admit the 
phenomenon is real.” 27 The article recaps differences of political opinion surrounding the 
theory and provides further references towards a rise in violent crime. Mac Donald 

22 Rushin and Edwards, “De-Policing,” 6. 

23 Ibid., 54. 

24 Ibid. 

25 Heather Mac Donald, “The New Nationwide Crime Wave,” The Wall Street Journal , May 29, 2015, 
http://www.wsj.com/articles/the-new-nationwide-crime-wave-1432938425. 

26 Ibid. 

27 Heather Mac Donald, “The Nationwide Crime Wave Is Building,” The Wall Street Journal, May 23, 
2016, http://www.wsj.com/articles/the-nationwide-crime-wave-is-building-1464045462. 


13 



articulates, “Despite this mounting evidence, the Ferguson effect continues to be distorted 
by its critics and even by its recent converts. The standard line is that it represents a 
peevish reaction from officers to ‘public scrutiny’ and expectations of increased 
accountability.” 28 She describes the problem as being “the activist-stoked hostility 
toward the police on the streets and ungrounded criticism of law enforcement that has 
flowed from the Obama administration and has been amplified by the media.” 29 Mac 
Donald believes “denial of the Ferguson effect is driven by a refusal to acknowledge the 
connection between proactive policing and public safety.” 30 

Officer anxiety is an integral part of Mac Donald’s understanding of the Ferguson 
Effect, as she believes such anxiety leads to a reduction in proactive policing and directly 
correlates to public safety. Scholars who have studied officers’ anxiety and perceptions 
include Scott Wolfe and Justin Nix, who published a study in October 2015 that sought to 
use survey data to measure police sentiment. 31 The study focused on a single geographic 
location within the United States, which attempted to establish empirical evidence 
through administering an online survey to an undisclosed mid-sized law enforcement 
agency. The survey involved a few hundred officers from the same department and 
focused on gathering police perspective on negative publicity and motivations. 32 Wolfe 
and Nix analyzed the received responses, which indicated “there appear to be a 
relationship between reduced motivation as a result of negative publicity and less 
willingness to work directly with community members to solve problems.” 33 

Others have sought to find other explanations for the rise in crime being 
experienced within many cities. Comey indicates some of these other possibilities in his 
speech, “Maybe it’s the return of violent offenders after serving jail terms. Maybe it’s 

28 Mac Donald, “The Nationwide Crime Wave Is Building.” 

29 Ibid. 

30 Ibid. 

31 Scott E. Wolfe and Justin Nix, “The Alleged ‘Ferguson Effect’ and Police Willingness to Engage in 
Community Partnership,” Law and Human Behavior 40, no. 1 (2016): 1-10, http://dx.doi.org/10.1037/lhb 
0000164. 

32 Sean Allocca, “The Facts Behind the ‘Ferguson Effect’,” Forensic Magazine, October 29, 2015, 
http://www.forensicmag.com/article/2015/10/facts-behind-ferguson-effect. 

33 Ibid. 


14 



cheap heroin or synthetic drugs. Maybe after we busted up the large gangs, smaller 
groups are now fighting for turf. Maybe it’s a change in the justice system’s approach to 
bail or charging or sentencing.” 34 Others, like Milwaukee Police Chief Edward Flynn 
believe “state budget cuts and lax gun laws are more likely behind spike,” in his 
jurisdiction. 35 “For police in Milwaukee, the real Ferguson effect has been a sudden and 
harsh cold shoulder from the public.” 36 

Criminology professors Tracy Sohoni and Charis Kubrin suggest the rise in 
violent crime hinges on the concept of a “legitimacy theory,” which proposes, “When 
individuals lose trust in the police they are also more likely to take matters into their own 
hands when conflicts arise.” 37 They write, “the highly publicized events in Ferguson, 
New York City, and Baltimore may pose a ‘legitimacy challenge’ to the criminal justice 
system, creating a legitimacy deficit that increases legal cynicism and encourages 
individuals to take the law into their own hands.” 38 

Rhys Blakely, U.S. editor at The Times, offers yet another explanation for the rise 
in violent crime, at least in the instances surrounding Baltimore. He claims “the higher 
rate of murders in Baltimore can be attributed to the riots interfering with the city’s 
illegal drugs market.” 39 Christian Science Monitor staff writer Henry Gass holds a 
different viewpoint, suggesting the Ferguson Effect seems to be in play at least for the 
Baltimore incident, noting that Freddie Gray was killed in April, and 43 homicides 
occurred the following month, making May the city’s most “deadliest month in 40 
years.” 40 

34 Comey, “Law Enforcement and the Communities We Serve: Bending the Lines Toward Safety and 
Justice.” 

35 Fears, “In Milwaukee, Weak Evidence for ‘Ferguson Effect’.” 

36 Ibid. 

37 Tracy Sohoni and Charis Kubrin, “Is the Ferguson Effect a Myth?,” The Baltimore Sun, September 
10, 2015, http://www.baltimoresun.com/news/opinion/oped/bs-ed-ferguson-effect-20150910-story.html. 

38 Ibid. 

39 Rhys Blakely, “U.S. Violent Crime Surges to 1970s Levels after Race Riots,” The Times, June 3, 
2015, http://www.thetimes.co.uk/tto/news/world/americas/article4458700.ece. 

40 Henry Gass, “FBI Chief Pins Crime Surge on ‘Ferguson Effect.’ Are Cops Afraid to Do Their 
Jobs?,” The Christian Science Monitor, October 26, 2015, http://www.csmonitor.com/USA/Justice/2015/ 
1026/FBI-chief-pins-crime-surge-on-Ferguson-effect.-Are-cops-afraid-to-do-their-jobs. 

15 



A study published by Stephen Morgan and Joel Pally of John Hopkins University 
in March 2016 analyzed the number of arrests in Baltimore City between the timeframe 
of March 2010 through December 2015. Researchers attempted to determine if the 
Ferguson Effect contributed to the civil unrest surrounding the death of Freddie Gray. 
With regard to the number of arrests analyzed, the study resulted in three primary 
conclusions: (1) “during the post-Ferguson, pre-Gray period from August 11, 2014 
through April 19, 2015, changes in arrest rates are consistent with a Ferguson effect on 
police conduct.” With regard to less serious crime, arrests decreased significantly 
although, “non-discretionary arrests for violent crimes, such as murder and robbery, held 
steady during this period,” (2) “Arrests declined further by an additional 30% from April 
20, 2015 through July 12, 2015, after removing the complex pattern of arrests that 
emerged during the week of unrest,” and (3) “Arrests increased during the remainder of 
2015, after the appointment of a new police commissioner.” 41 

D. OPPONENTS OF THE FERGUSON EFFECT 

Some critics of the Ferguson Effect continue to focus on the question of whether 
or not an increase in crime rates has occurred. David C. Pyrooz, for example, led a study 
titled, “Was There a Ferguson Effect on Crime Rates in Farge U.S. Cities?” 42 His 
research analyzed data for “81 U.S. cities with populations exceeding 200,000 persons,” 
using a “discontinuous growth model to determine if there was a redirection in 
seasonality-adjusted crime trends in the months following the Ferguson shooting.” 43 
Pyrooz found no evidence to “support a systematic post-Ferguson change in overall, 
violent, and property crime trends,” although he observed robbery rates “increased in the 
months after Ferguson.” 44 His study acknowledged that some “select cities did 
experience increases in homicide,” adding, “overall, any Ferguson Effect is constrained 

41 Stephen L. Morgan and Joel A. Pally, Ferguson, Gray, and Davis, An Analysis of Recorded Crime 
Incidents and Arrests in Baltimore City, March 2010 through December 2015 (Baltimore, MD: Johns 
Hopkins University, 2016), 4, http://socweb.soc.jhu.edu/faculty/morgan/papers/MorganPally2016.pdf. 

42 David C. Pyrooz et al., “Was There a Ferguson Effect on Crime Rates in Large U.S. Cities?” 

Journal of Criminal Justice 46 (2016): 1-8. 

43 Ibid. 

44 Ibid. 


16 



largely to cities with historically high levels of violence, a large composition of black 
residents, and socioeconomic disadvantages.” 45 Pyrooz’s study focused on crime, ‘‘What 
our analysis cannot speak to is the extent to which de-policing or a crisis in police 
legitimacy have occurred post-Ferguson, and if so, the impact it may have had on crime 
rates.” 46 

Much of the controversy over the Ferguson Effect, however, has focused on the 
question of whether or not any increase in crime can be attributed to de-policing on the 
part of law enforcement officers. Several top government officials have publicly 
supported this view, including as noted previously the Director of the FBI. 47 
Additionally, the acting chief of the Drug Enforcement Administration (DEA), Chuck 
Rosenberg, supports the likelihood of a Ferguson Effect and stated it may be responsible 
for the hesitancy of officers to do their jobs at the same levels of policing pre-Ferguson. 48 

However, others have opposed this view of the Ferguson Effect, seeing in it a 
suggestion that police officers have become negligent in their duties. The White House 
and Department of Justice (DOJ) under President Obama saw the issue this way, in stark 
contrast with the views of the FBI and DEA. Obama White House Press Secretary Josh 
Earnest summarized the White House position in the following statement, “The fact is the 
evidence does not support the claim that somehow our law enforcement officers across 
the country are shirking their duties and failing to fulfill their responsibility to serve and 
protect the communities to which they are assigned.” 49 The wording of Earnest’s 
response contends officers are not committing a dereliction of duty, which is very 
different than entertaining the possibility that de-policing is occurring. Although Comey 


45 Pyrooz et at., “Was There a Ferguson Effect on Crime Rates in Large U.S. Cities?” 1-8. 

46 Ibid. 

47 For more on Comey’s views, see Gass, “FBI Chief Pins Crime Surge on ‘Ferguson Effect.’ Are 
Cops Afraid to Do Their Jobs?.” 

48 Eric LichtBlau, “Officials Debate Whether ‘Ferguson Effect’ Is Real,” The New York Times, 
November 4, 2015, http://www.nytimes.eom/2015/l 1/05/us/politics/officials-debate-effect-of-scrutiny-on- 
police.html. 

49 Zeke Gillman, “FBI Director James Comey on the ‘Ferguson Effect’ at the Law School,” The 
Chicago Maroon, November 10, 2015, https://chicagomaroon.com/2015/ll/10/fbi-director-james-comey- 
on-the-ferguson-effect-at-the-law-school. 


17 



said the Ferguson Effect is a possible explanation for the rise in crime, he also 
acknowledged more data is needed to get a better handle on what is happening. 

Former Attorney General Loretta Lynch was also skeptical of the Ferguson 
Effect. Although she has not totally dismissed the idea. Lynch said she “has not seen 
anything resembling that in her experience,” adding, “we had a recent DOJ study that 
looked at that issue, and the conclusion was, we need more information.” 50 Lynch was 
likely referring to a study released the previous month under contract with support from 
the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice. 
The DOJ study was led by Richard Rosenfeld, who found a rise in homicide occurring in 
2015 across multiple cities. Unlike Rosenfeld’s previous Sentencing Project study, which 
only analyzed homicide statistics from the single city of St. Louis, the DOJ study was 
much broader and took into account year-end homicide data for 56 cities across the 
nation. 

The DOJ study determined “the homicide rise in 2015 in the nation’s large cities 
was real and, while not unprecedented, comparatively large.” 51 The significance of the 
DOJ study is that it acknowledged homicide rates in America’s largest cities did increase 
the year following Ferguson. However, while the study also confirmed that homicide 
rates had increased, no conclusive evidence was given as to the cause. Rosenfeld 
presented three possible explanations for the rise in homicide: (1) “an expansion of urban 
drug markets fueled by the heroin epidemic, (2) reductions in incarceration resulting in a 
growing number of released prisoners in the nation’s cities, and (3) a ‘Ferguson effect’ 
resulting from widely publicized incidents of police use of deadly force against minority 
citizens.” 52 Rosenfeld did not endorse the Ferguson Effect as the sole reason for the rise 
in crime; he merely suggested it could be a factor. 


50 Pete Kasperowicz, “Lynch: I Have Not Seen Ferguson Effect,” Washington Examiner, July 1, 2016, 
http://www.washingtonexaminer.com/lynch-i-have-not-seen-ferguson-effect/article/2595445. 

51 Richard Rosenfeld, Documenting and Explaining the 2015 Homicide Rise: Research Directions 
( Washington, DC: U.S. Department of Justice, Office of Justice Programs, National Institute of Justice, 
2016), https://www.ncjrs.gov/pdffilesl/nij/249895.pdf. 

52 Ibid. 


18 



Rosenfeld assesses the Ferguson Effect by using two opposing definitions of the 
term. On the one hand, he addresses the definition from the more widely accepted police 
perspective of community disengagement. On the other, he looks at the issue through the 
lens of longstanding grievances by the African American community with police. He 
refers to the latter perspective as a “legitimacy crisis.” Rosenfeld suggests the matter of 
de-policing can be gauged through information pertaining to self-initiated activity and 
arrest data. 53 He states in his hypothesis, “if de-policing was the operative mechanism, 
we should observe larger drops in arrests and other self-initiated police activities in cities 
that experienced the greatest homicide increases.” 54 Rosenfeld’s hypothesis sets the stage 
for future research but does not evaluate such speculation within this particular study. 

This thesis further explores Rosenfeld’s assumptions by taking into account arrest 
and other self-initiated police activity. Where this thesis diverges is allowing productivity 
data to stand independent of any changes in homicide rates. Rosenfeld’s earlier study 
surrounding St. Louis found the Ferguson Effect was not a factor in increased homicide; 
the DOJ study, however—that may be more accurate because it analyzed a larger amount 
of data—offers the Ferguson Effect as a possible explanation for the rise in homicide 
within 56 cities. 

E. CONCLUSION 

Most studies of the Ferguson Effect assume the reduction of police productivity 
and increased crime are related, and that one must exist to support the other. Moreover, 
although a rise in violent crime may be caused by a reduction in proactive policing, this 
thesis argues that the conflation of the two factors complicates this discussion. A rise in 
violent crime is not vital for de-policing to exist, and the withdrawal of a police presence 
may or may not be evidenced by a spike in homicide rates or violent crime. This scenario 
is especially true in smaller jurisdictions that regularly experience low homicide rates in 
the first place. For this reason, this study seeks to evaluate police productivity data on its 
own merit to gauge sentiment and de-policing practices regardless of crime associations 

53 Rosenfeld, Documenting and Explaining the 2015 Homicide Rise: Research Directions. 

54 Ibid. 


19 



although crime statistics are reintroduced for consideration as an associated variable for 
comparative purposes. 

Chapter III, Methodology, discusses the research approach used in obtaining data, 
agency selections, and data analysis. 


20 



III. METHODOLOGY 


The study seeks to resolve a theoretical debate through social science research; 
especially, by conducting case studies of several police departments, and for each case, 
conducting quantitative dataset collection and trend analysis. Specifically, the study seeks 
to use police productivity data that can be aggregated through open sources, analyzed, 
and correlated, to either support or contradict the Ferguson Effect. 

The following research questions are addressed in this study. 

• Can open source data concerning police productivity be collected and 
analyzed either to support or contradict the Ferguson Effect? If so, what 
does that data show? 

• Has police productivity decreased overall since August 2014? 

• What, if any, impact does a reduction in police productivity have on 
violent crime? 

• What future challenges do law enforcement administrators face if the 
Ferguson Effect is deemed viable? 

The research method follows a multiple case study approach concentrating on 
quantitative open source dataset collection and trend analysis. The datasets are examined 
for common correlations in an attempt to determine if police productivity trends alone 
may be considered an accurate indicator or verifying factor in corroborating the existence 
of a Ferguson Effect. Evaluating the system as a whole, the most obvious conclusion may 
indicate reductions in police productivity lead to simultaneous increases in crime. The 
premise is based on the assumption that crime will flourish in areas where police have 
withdrawn, thus creating a void for criminal activity. As previously stated, the researcher 
does not believe a rise in crime is needed to establish merit for the Ferguson Effect. The 
hypothesis is grounded on the belief police productivity will tell its own story as to 
whether officers are becoming apprehensive in performing the functions of their job. 
Crime statistics are included as an independent variable that may or may not have 
correlations to police productivity or the Ferguson Effect in all instances. 


21 



Dispatched calls for service are obligatory and officers are bound to respond to 
such incidents. For this reason, dispatched calls for service are not a reliable indication of 
police productivity within the context of the argument. On the other hand, self-initiated 
calls are mostly reliant upon an individual officer’s desire to be proactive and are largely 
dependent on that officer’s discretion and temperament. Self-initiated calls may consist of 
traffic stops, pedestrian stops, premise checks, and certain types of arrests like warrant 
services. Secondary indicators are also sought in hopes of drawing additional correlations 
about newly formed police anxieties. These factors consist of police retirements, new 
hires, use of sick leave, and call response times. 

A. PROCEDURES 

The study collects law enforcement open data sets to develop three explanatory 
case studies highlighting quantitative data trends plotted on a graph for the purpose of 
drawing possible police productivity correlations consistent with the Ferguson Effect. 
Data sets were constrained to utilizing publically available government sponsored open 
data repositories. 

The research set out to identify open data sources relating to police productivity. 
Data sets of interest include self-initiated calls for service, i.e., traffic stops. Also of 
interest but to a lesser degree is the number of arrests conducted by a department. Less 
emphasis is placed on arrest data because many arrests are outside the scope of 
discretionary measure; for example, arrests arising out of domestic violence situations. In 
these cases, officers are often mandated by state laws to conduct arrests if certain 
conditions are met. Lastly, data sets containing current and historical crime incident 
information are sought out through the same open sources channels. 

The scope of data sought after includes at least two years both prior to and after 
the shooting of Michael Brown, as noted in Ligure 1. Historical data leading up to the 
Lerguson incident is important in establishing baseline productivity trends. Data points 
occurring after Lerguson are compared and contrasted against any identified historical 
trends. 


22 



Ferguson Incident 


Aug 2012 Aug 2014 


Figure 1. Scope of Data—Two Years Before and After Ferguson Incident 

A matrix depicting ideal variables, shown in Table 1, is established outlining 
desired data set components relating to the categories of police productivity, morale, 
administrative function, and violent crime. The listed variables become the benchmark 
for choosing related datasets, which in turn, dictate which agencies may be best suited for 
case study. 


Aug 2016 


Table 1. Desired Variables. 


Productivity 

Morale 

Administrative 

Crime 

Traffic Stops 

Use of Sick 

Leave 

Applicants 

Violent 

Crime 

Pedestrian Stops 


Retirees 


Arrests 




Other Self-Initiated Calls for 
Service 




Response Time to Calls 





The FBI’s UCR Program defines violent crime as being “composed of four 
offenses: murder and nonnegligent manslaughter, forcible rape, robbery, and aggravated 
assault ... [and includes] offenses which involve force or threat of force.” 55 

B. DATA SETS 

On January 20, 2009, President Obama signed the Memorandum on Transparency 
and Open Government. The memorandum affirmed the Obama Administration’s pledge 

55 “UCR, 2015 Crime in the United States, Violent Crime,” accessed October 12, 2016, https://ucr. 
fbi.gov/crime-in-the-u.s/2015/crime-in-the-u.s.-2015/offenses-known-to-law-enforcement/violent-crime. 


23 














to ensure trust with the public through openness in government transparency, 
participation, and collaboration. The memorandum provided recommendations for an 
Open Government Directive within the executive departments. 56 In December of the 
same year, the White House issued an Open Government Directive “requiring federal 
agencies to take immediate, specific steps to achieve key milestones in transparency, 
participation, and collaboration.” 57 The memorandum tasked executive departments and 
related agencies with the “goal of creating a more open government” through the 
following steps: “Publish Government Information Online, Improve the Quality of 
Government Information, Create and Institutionalize a Culture of Open Government, and 
Create an Enabling Policy Framework for Open Government.” 58 

The final report of the President’s Task Force on 21st Century Policing was 
released in May 2015 and contributed to the development of the PDI, which took into 
account many of the technological recommendations defined in the document. The PDI 
was launched in May 2015 and immediately “mobili z ed 21 leading jurisdictions across 
the country to take fast action on concrete deliverables responding to these Task Force 
recommendations in the area of data and technology.” 59 

U.S. Chief Data Scientist Dr. DJ Patil explains, “PDI is centered on two key 
components: (1) using open data to build transparency and increase community trust, and 
(2) using data to enhance internal accountability through effective analysis.” 60 The Police 
Foundation’s Public Safety Open Data Portal (https://publicsafetydataportal.org) was 
created to “serve as a central clearinghouse for accessing, visualizing and analyzing local 
and national law enforcement and public safety open datasets.” 61 As a result of the 

56 “Transparency and Open Government,” accessed October 11, 2016, https://www.whitehouse.gov/ 
the_press_office/TransparencyandOpenGovernment. 

57 “Open Government Initiative,” accessed October 11, 2016, https://www.whitehouse.gov/open/ 
about. 

58 “Open Government Directive,” accessed October 11, 2016, https://www.whitehouse.gov/open/ 
documents/open-government-directive. 

59 Megan Smith and Roy L. Austin, Jr. May 18, 2015 (6:00 a.m.), “Launching the Police Data 
Initiative,” The White House Blog , accessed October 11, 2016, https://www.whitehouse.gov/blog/2015/ 

05/18/launching-police-data-initiative. 

60 “The Police Data Initiative (PDI).” 

61 Ibid. 


24 



project, many law enforcement agencies are now sharing statistical datasets to include 
calls for service, traffic stops, use of force incidents, and arrests to name a few. The 
number of participating agencies has grown from the original 21 to 120 agencies 
currently. 62 The Public Safety Open Data Portal itself is merely a pointer system to other 
local and state open data portals and does not store any of the listed datasets on their 
servers. 


The referenced portal indexes a variety of law enforcement datasets organized by 
both category and agency name. Additionally, a search feature can locate datasets by 
inputting specific keyword(s). Portal data is categorized into the following dataset 
classifications: assaults on officers, calls for service, incidents, officer involved shooting, 
stops/citations/arrests, use of force and miscellaneous. 63 

The Public Safety Open Data Portal was chosen as the primary repository for 
obtaining law enforcement datasets due to its appeal of transparency and sharing of 
datasets by 120 participating law enforcement agencies. 

C. CASE STUDIES 

Three police departments are chosen as case studies, selected based on dataset 
availability and data comprehensiveness. The following criteria are established and 
become the rationale behind final agency selections. 

• Agencies utilized in the case studies are nominated from the list of 
participating agencies registered through the Public Safety Open Data 
Portal. 

• Consideration is given to those agencies that had overlying variables for 
the desired timeframe of 24 months prior to the events in Ferguson and 24 
months post incident. The two most common variables for the dates in 
question boil down to traffic stops and arrests. Datasets related to violent 
crime are chosen as the third common variable. 

• Additional consideration is given to choosing three different departments 
that are varied in size (small, medium, and large) along with geographic 
separation (East, Midwest, and West). 


62 “The Police Data Initiative (PDI).” 

63 Ibid. 


25 



D. DATA ANALYSIS 


A visual analysis of the related case studies is plotted on a timeline to determine 
correlations and patterns with regard to productivity on one hand, and impacted crime on 
the other. Data visualization software is employed for this project, specifically, a program 
named, “Tableau for Students.” 64 

The analytical technique of simple time trend analysis is used on each case study. 
“The essential logic underlying a time series design is the match between the observed 
(empirical) trend and either of the following: (a) a theoretically significant trend specified 
before the onset of the investigation or (b) some rival trend, also specified earlier.” 65 The 
final analytical step pertained to identifying potential “slope” patterns between the three 
case studies. 

Prior to graphing or analyzing data, the anticipated hypothesis suggests a decline 
in police productivity in those months following the events in Ferguson, Missouri may be 
seen. Moreover, the researcher expects to observe direct correlations between reductions 
in police productivity and a rise in violent crime in more urbanized areas. 

Chapter IV reveals the three police departments used as case studies and explains 
the rationale behind the selections. Each case study analyzes three key variables. The first 
two variables, traffic stops and arrests, are associated with police productivity. The third 
variable concentrates on the variance of violent crime. Datasets relating to each variable 
are graphed on a timeline in relationship to the shooting of Michael Brown. 


64 “Tableau for Students.” 

65 Robert K. Yin, Case Study Research: Design and Methods (Los Angeles: Sage Publications, 2009), 
144-145. 


26 



IV. RESULTS AND ANALYSIS 


The Public Safety Open Data Portal is used as a starting point for locating datasets 
that meet the ideal model variables relating to police productivity. However, the majority 
of datasets are incomplete, irrelevant, or outdated. As a result, model variables are 
reduced to just two feasible subsets, Variable 1—Traffic Stops, and Variable 2—Arrests. 
Correspondingly, dataset limitations constrained choice of candidate agencies. 
Ultimately, case study targets are picked strictly on data availability as opposed to 
randomness, agency size, or geographical consideration. The following police 
departments are designated on the lone premise of common dataset obtainability. 

• Burlington Police Department, Burlington, Vermont 

• Montgomery County Police Department, Montgomery County, Maryland 

• Philadelphia Police Department, Philadelphia, Pennsylvania 

Table 2 depicts the available variable datasets for each agency along with the 
timeframe span for each dataset. Historical data in all variables is restricted to two years 
prior to the shooting of Michael Brown, August 2012, for consistency purposes. Those 
datasets containing historical data prior to August 2012 are excluded from analysis. On 
the opposite side of the equation, the endpoint for current data is capped at August 2016 
for the same justifications. Datasets with time and parameter shortcomings are ruled by 
available data within that particular set. 


Table 2. Case Study Departments and Obtainable Variables. 


CASE STUDY 

Variable 1 
Traffic Stops 

Variable 2 
Arrests 

Variable 3 
Violent Crime 

Burlington Police 
Department 

Aug 2012-Dec 2015 

Aug 2012-Aug 
2016 

Aug 2012- 
Aug 2016 

Montgomery County 
Police Department 

Jan 2013-Aug 2016 

Sep 12, 2016- 
Oct 12, 2016* 

June 2013- 
Aug 2016 

Philadelphia Police 
Department 

Jan 2014-Aug 2016 

Unavailable 

Aug 2012- 
Aug 2016 


*Online data limited to prior 30 days. 


27 










All three departments are geographically located on the east coast of the United 
States and range in size from 99 sworn officers to over 6,400 sworn officers. To provide a 
geographic frame of reference, Burlington, Vermont, is located approximately 1,100 
miles northeast of Ferguson, Missouri. 66 Montgomery County, Maryland, is 
approximately 800 miles east of Ferguson, 67 while Philadelphia, Pennsylvania, is 
approximately 900 miles east of Ferguson. 68 

Two critical dataset limitations have been discovered with regard to dataset 
availability for the chosen departments. The first concerned arrest data for the 
Montgomery County Police Department, which is limited to the last 30 days. The second 
deals with a total absence of arrest data for the Philadelphia Police Department. These 
issues are further addressed within the respective case study analysis sections. 

A. CASE STUDY 1—BURLINGTON POLICE DEPARTMENT, 

BURLINGTON, VERMONT 

Case Study 1 focused on the Burlington Police Department, located in Burlington, 
Vermont. The Burlington Police Department has 99 sworn officers and serves a 
population of 42,160. 69 Census data from 2010 indicates the following percentage 
breakdown of race and ethnicity: 88.9% White alone, 3.9% Black or African American 
alone, 0.3% American Indian or Alaska Native alone, 3.6% Asian alone, 2.6% Two or 
More Races, and 2.7% Hispanic or Latino alone. 70 Whites make up the predominate race 
in Burlington, Vermont. 


66 “Directions from Ferguson, Missouri to Burlington, Vermont,” Google Maps, accessed October 11, 
2016, https://goo.gl/maps/vL4zb6LN8KL2. 

67 “Directions from Ferguson, Missouri to Montgomery County, Maryland,” Google Maps, accessed 
October 11, 2016, https://goo.gl/maps/8dlFyPZ6DBp. 

68 “Directions from Ferguson, Missouri to Philadelphia, Pennsylvania,” Google Maps, accessed 
October 11, 2016, https://goo.gl/maps/xRUbJTYQ5cJ2. 

69 “2015 Crime in the United States, Table 78, Vermont, Full-Time Law Enforcement Employees by 
City, 2015,” accessed October 11, 2016, https://ucr.fbi.gOv/crime-in-the-u.s/2015/crime-in-the-u.s.-2015/ 
tables/table-78/table-78-state-pieces/table_78_full_time_law_enforcement_employees_vermont_by_cities_ 
2015.xls. 

70 “QuickFacts, Burlington City, Vermont,” accessed October 13, 2016, http://www.census.gov/quick 
facts/table/PST045215/5010675. 


28 



1 . 


Variable 1—Traffic Stops 


Traffic stop information includes data points from August 2012 through 
December 2015, as demonstrated in Figure 2. No statistics are readily available for the 
months of January 2016 through August 2016. The red line is a marker overlaid on 
August 2014 that indicates on the timeline the beginning events in Ferguson, Missouri. 
The red line is subsequently displayed for the same time period on each ensuing graph. 


Date Issued 



Figure 2. Burlington Police Department, Variable 1—Traffic Stops 
(August 2012-December 2015). 71 


The total number of traffic stops decreased from 567 in August 2014 to 473 the 
following month of September 2014. The number of traffic stops rebound the month of 
October and November, slightly decrease in December, and jump to 642 traffic stops 
during the month of January 2015. Historical data indicates fewer monthly traffic stops 
for the five months leading up to August 2014. Overall traffic stop trends for the time 


7 ' Source: “Burlington Police Department Traffic Stops,” accessed October 8, 2016, https://www.bur 
lingtonvt.gov/police/data/traffic. 


29 










period in question indicates a series of bell curves generally beginning in June or July and 
ending the following year around the same time. Approximately three of these bell curves 
are distinguishable in the timeline. August 2014 was in the same ballpark on the curve for 
August 2012, August 2013, and August 2014. The pattern of traffic stops in Burlington, 
Vermont appears largely unaffected by the events in Ferguson. 

2. Variable 2—Arrests 

In Figure 3, arrest information includes data points from August 2012 through 
August 2016. All departmental arrest information is included in the graph and no 
distinction is made between felony, misdemeanor, or traffic related arrests. 


Arrest Date/Time 



Figure 3. Burlington Police Department, Variable 2—Arrests 
(August 2012-August 2016). 72 


The number of total arrests actually increased the two months following August 
2014. Arrests begin to decline proportionately in November 2014 through January of 
2015 when compared to other yearly cycles. Roughly four bell curve cycles are evident 
on the timeline. August 2014 is slightly lower than peak for the prior and following six- 

72 Source: “Burlington Police Department Arrests,” accessed October 9, 2016, https://www.burlington 
vt.gov/police/data/arrests. 


30 






month period. It is not possible to determine if August 2014’s total arrests were slightly 
lower on the curve because of the Ferguson incident or another cause. 

3. Variable 3—Violent Crime 

Violent crime is not necessarily an indication of police productivity but it is 
included in each case study for added correlation consideration. Violent crime consists of 
murder, rape, aggravated assault, and robbery. The violent crime dataset spans the desired 
timeline from August 2012 through August 2106, as seen in Figure 4. Violent crime data 
was filtered from reported calls for service reported by the Burlington Police Department. 
It is important to note that “calls for service” are not necessarily an exact match to final 
offense count numbers later submitted to the FBI’s UCR Program. 


Cal Date/Time 



Figure 4. Burlington Police Department, Variable 3—Violent Crime 

(August 2012-August 2016). 73 


Violent crime rates for the five consecutive months following Ferguson, 
September 2014 through February 2015, were lower each of these months than they were 
in August 2014. No patterns or apparent correlations exist implicating the fact violent 


73 Source: “Burlington Police Department Calls for Service,” accessed October 9, 2016, https://www. 
burlingtonvt.gov/police/data/dashboard. 


31 




crime was affected by the Ferguson Effect in Burlington, Vermont, during the timeframe 
in question. 

B. CASE STUDY 2—MONTGOMERY COUNTY POLICE DEPARTMENT, 

MONTGOMERY COUNTY, MARYLAND 

Case Study 2 concentrates on the Montgomery County Police Department, 
located in Montgomery County, Maryland. The Montgomery County Police Department 
has 1,378 sworn officers 74 and serves a population of 1,040,116. July 1, 2015 census data 
indicates the following percentage breakdown of race and ethnicity: 57.5% White alone, 
19.1% Black or African American, 0.7% American Indian or Alaska Native, 15.4% 
Asian alone, 0.1% Native Hawaiian and Other Pacific Islander alone, 3.3% Two or More 
Races, and 19% Hispanic or Latino. 75 Montgomery County is the most culturally diverse 
region among the three case studies. 

1. Variable 1—Traffic Stops 

Traffic stop information for the Montgomery County Police Department includes 
data points from January 2013 through August 2016, as displayed in Figure 5. No 
statistics are readily available for the months of August 212 through December 2012. The 
amount of traffic stops were roughly 30 times greater than the amount of traffic stops 
conducted by the Burlington Police Department for the same month of August 2014. 


74 “2015 Crime in the United States, Table 80, Maryland, Full-Time Law Enforcement Employees by 
Metropolitan and Nonmetropolitan Counties, 2015,” accessed October 11, 2016, https://ucr.fbi.gov/crime- 
in-the-u.s/2015/crime-in-the-u.s.-2015/tables/table-80/table-80-state-pieces/table_80_full_time_law_enfor 
cement_employees_maryland_by_metropolitan_and_nonmetropolitan_counties_2015.xls. 

75 “QuickFacts, Montgomery County, Maryland,” accessed October 13, 2016, http://www.census.gov/ 
quickfacts/table/PST045215/24031. 


32 



Date Of Stop 



Figure 5. Montgomery County Police Department, Variable 1—Traffic Stops 

(January 2013-August 2016). 76 


In the month of August 2014, 17,132 traffic stops were conducted, which was 
1,733 less than the previous month of July 2014. Typically, traffic stops increased during 
the months of August from July for every other year on the timeline: 2013, 2015, and 
2016. Not every September in the timeline had increases in traffic stops over August 
months although September and October 2014 brought introduced increased traffic stops 
for both months. No indication can be seen that the Ferguson Effect influenced the total 
number of traffic stops being conducted during the months following events in Ferguson. 

2. Variable 2—Arrests 

Unlike arrest information supplied by the Burlington Police Department in Case 
Study 1, the Montgomery County Police Department does not strip Personal Identifiable 
Information (PII) from their arrest datasets, which is likely the reason they only post 
access to the last 30 days of arrest data. Arrest information is only available from 
September 12 through October 12, 2016. 

7(1 Source: “Montgomery County Police Department Vehicle Stops,” accessed October 8, 2016, https:// 
data.montgomerycountymd.gov/Public-Safety/Traffic-Violations/4mse-ku6q. 


33 








The Montgomery County Police Department arrest dataset does not contain 
enough data points to compare any historical data or subsequent months. For this reason, 
no inferences can be drawn from this limited dataset. 

3. Variable 3—Violent Crime 

As provided in Figure 6, violent crime information for the Montgomery County 
Police Department includes data points from July 2013 through August 2016. Similar to 
Case Study 1, it is important to remember the chart is adapted from a “calls for service” 
dataset that is not always equal to the final offense count numbers submitted by the 
agency to the FBI’s UCR Program. 


Start Date / Time 



Figure 6. Montgomery County Police Department, Variable 3—Violent Crime 

(July 2013-August 2016). 77 


The violent crime rate in Montgomery County for August 2014 was 255. Violent 
crime dropped to 202 in September 2014 and rose to 247 in October 2014. The last half 

77 Source: “Montgomery County Police Department Violent Crime,” accessed October 9, 2016, 
https://data.montgomerycountymd.gov/api/views/4mse-ku6q/rows.csv?accessType=DOWNLOAD. 


34 







of 2013 appears to be the lowest period of violent crime on the timeline. The highest 
amount of violent crime occurred in November 2015 when 287 incidents occurred. 
Again, no obvious pattern appears to impact a significant rise or decline in violent crime 
during the timeframe in question. 

C. CASE STUDY 3—PHILADELPHIA POLICE DEPARTMENT, 

PHILADELPHIA, PENNSYLVANIA 

Case Study 3 evaluates the Philadelphia Police Department, located in 
Philadelphia, Pennsylvania. The Philadelphia Police Department has 6,413 officers and 
serves a population of 1,567,8 1 0. 78 Census data from 2010 indicates the following 
percentage breakdown of race and ethnicity: 41% White alone, 43.4% Black or African 
American alone, 0.5% American Indian or Alaska Native alone, 6.3% Asian alone, 2.8% 
Two or More Races, and 12.3% Hispanic or Latino alone. 79 The black population is 2.4% 
greater than white residents although the majority of the city’s demographics are 
basically split between these two groups. 

1. Variable 1—Traffic Stops 

Traffic stop information for the Philadelphia Police Department in Figure 7 
includes data points from January 2014 through August 2016. No statistics are readily 
available for the months of August 2012 through December 2013. Also included in the 
dataset is information relating to pedestrian stops. Although the number of pedestrian 
stops is a relevant indicator of productivity, the data is not analyzed or graphed because 
comparable information is not available for the case studies involving Burlington and 
Montgomery County. 


78 “2015 Crime in the United States, Table 78, Pennsylvania, Full-Time Law Enforcement Employees 
by City, 2015,” accessed October 11, 2016, https://ucr.fbi.gOv/crime-in-the-u.s/2015/crime-in-the-u.s.- 

2015/tab les/table-78/table-78-state-pieces/table-78-pennsylvania-new. 

79 “QuickFacts, Philadelphia City, Pennsylvania,” accessed October 13, 2016, http://www.census.gov/ 
quickfacts/table/PST045215/4260000. 


35 



30K 


Datefilme 



Figure 7. Philadelphia Police Department, Variable 1—Traffic Stops 

(July 2013-August 2016). 80 


The average amount of traffic stops conducted during this time period is 21,976. 
Regrettably, little historical data is available. Furthermore, the number of stops conducted 
in January 2014 is suspiciously low. It is unknown if this number is actually the number 
of stops that were actually conducted, if a clerical error occurred, or the information is 
simply omitted from the dataset. The number of traffic stops conducted during the month 
of August 2014 is 18,977. Traffic stops continued to be higher for each consecutive 
month all the way through July 2016 when they dipped to 16,745. Case Study 3 is no 
different from the other two case studies regarding no noticeable trends pertaining to 
traffic stops. 


8<) Source: “Philadelphia Police Department Vehicle Stops,” accessed October 18, 2016, https://www. 
opendataphilly.org/dataset/vehicle-pedestrian-investigations/resource/e02050cl-51a4-48ff-80d2-88b2900 


47837. 


36 





2 . 


Variable 2—Arrests 


The Philadelphia Police Department offered no available datasets pertaining to 
arrests through its open data portal, which proved to be a significant drawback for a study 
comparison. 


3. Variable 3—Violent Crime 

In Figure 8, the violent crime dataset for the city of Philadelphia spans the desired 
timeline from August 2012 through August 2106. Case Study 3, like the other two case 
studies, is adapted from a “calls for service” dataset that is not always equal to the final 
offense count numbers submitted by the agency to the FBI’s UCR Program. 


Dispatch Date 



Figure 8. Philadelphia Police Department, Variable 3—Violent Crime 

(August 2012-August 2016). 81 


No evident positive or negative patterns of violent crime in response to the 
alleged Ferguson Effect appear during the timeframe in question. 


Source: “Philadelphia Police Department Violent Crime,” accessed October 9, 2016, https://data. 
phila.gov/api/views/sspu-uyfa/rows.csv?accessType=DOWNLO AD. 


37 






D. CASE STUDY COMPARISON 


Although the individual case studies offer no valid insights into settling the 
Ferguson Effect debate, the cases are compared and contrasted in attempt to identify 
common themes or patterns. 

A common theme surrounding the analysis of violent crime included recognition 
of patterns of activity that appeared to run consistently along bell curves on a yearly basis 
from one year to another. Despite this pattern, it has no apparent correlation with the 
phenomenon referred to as the Ferguson Effect. 

No commonalities are observed on the productivity side of the comparison, which 
especially holds true for Variable 1, traffic stops. Not only does each individual case 
study fall short of patterns or slopes, the case studies observed as a whole offer no 
additional awareness. 

Variable 2, arrests, is largely hindered by the amount of available data sets. Case 
Study 1 involving the Burlington Police Department is the only agency that contained a 
full dataset to analyze. Even so, being a smaller department, Burlington’s data is 
inconsequential due to the miniscule amount of data points it actually contains. Arrest 
data for Montgomery County may have given better understanding if the data contained 
the needed historical data past the 30 days offered. Likewise, a significant gap is left 
concerning the lack of arrest data for the Philadelphia Police Department. No intelligible 
inferences can be established from the arrest data available for the case studies. 

The final discussion chapter primarily addresses the limitations and weaknesses 
encountered during the study. Additionally, several recommendations are set forth for 
consideration in future studies. 


38 



V. DISCUSSION 


The shooting of an unarmed black teenager, Michael Brown, by a police officer, 
Darren Wilson, in Ferguson, Missouri, on August 9, 2014, led to criticisms against the 
Ferguson Police Department and national policing practices in general. The incident 
prompted public allegations of excessive use of force, police misconduct, civil rights 
violations, and racial divide between police and community members. Concerns of 
injustice for Michael Brown rallied local and national protests in Ferguson and other 
cities throughout the country. Ferguson gained national media attention early on and 
remained in media headlines for months to follow. In November 2014, a St. Louis Grand 
Jury announced Darren Wilson would not be indicted on charges of murder or 
manslaughter, which resulted in additional civil unrest and riots occurring in Ferguson. 82 
The DOJ conducted its own investigation and ultimately determined, “...it is not 
appropriate to present this matter to a federal grand jury for indictment, and it should 
therefore be closed without prosecution.” 83 

The research set out to resolve the Ferguson Effect debate by conducting three 
case studies utilizing trend analysis on law enforcement open source quantitative datasets. 
Publicly available data repositories through the Public Safety Open Data Portal, Public 
Data Initiative and related governmental links were utilized for raw dataset acquisition. 

Limitations in available datasets determined which three agencies were chosen for 
collection and analysis: (1) the Burlington Police Department, Burlington, Vermont, (2) 
the Montgomery County Police Department, Montgomery County, Maryland, and (3) the 
Philadelphia Police Department, Philadelphia, Pennsylvania. Data sought for productivity 
value included information relating to traffic stops, pedestrian stops, arrests, call response 
times, use of sick leave, new applicants, and retirees. Productivity related datasets proved 

82 Erin McClam, “Ferguson Cop Darren Wilson Not Indicted in Shooting of Michael Brown,” NBC 
News, November 25, 2014, http://www.nbcnews.com/storyline/michael-brown-shooting/ferguson-cop- 
darren-wilson-not-indicted-shooting-michael-brown-n255391. 

83 Department of Justice, Department of Justice Report Regarding the Criminal Investigation into the 
Shooting Death of Michael Brown by Ferguson Missouri Police Officer Darren Wilson (Memorandum) 
(Washington, DC: Department of Justice, 2014), 5, https://www.justice.gov/sites/default/files/opa/press- 
re leases/attachments/2015/03/04/doj_report_on_shooting_of_mic hael_brown_ 1. pdf. 


39 



challenging to obtain through the PDI. On the other hand, crime related data was slightly 
more accessible although each agency varied in reporting content, organization, and 
classification. 

In each case, the data was insufficient to confirm or deny the existence of the 
Ferguson Effect, although the limited data available did suggest that in these three cities, 
no noticeable de-policing occurred following the killing of Michael Brown. Several 
limitations became evident upon the conclusion of the study, with the most striking 
revelation being that major deficiencies appeared in the federal government’s 
transparency initiative and Open Government program. 

A. LIMITATIONS AND WEAKNESSES 

The three case studies were dependent on attaining complete datasets for the 
desired variables and timeframe to begin with a sound analytical framework. The ability 
to locate needed historical and recent datasets for each variable and agency would likely 
have established a clearer picture for correlation and comparison purposes. 

An alternate method for obtaining the needed information would have entailed 
completing a Freedom of Information Act (FOIA) request for each respective agency. A 
potential disadvantage in using FOIA requests as a data collection method is the often 
lengthy turnaround time. The Open Data Initiative was originally set up in part to relieve 
traditional FOIA requests to streamline the process of government openness and 
transparency. 

In hindsight, too much preliminary value was placed on the Open Data Initiative 
program along with the amount of expected available information. Aside from obtaining 
substandard data, numerous deficiencies in the program were exposed during the research 
process. 

• Incomplete and partial datasets 

• Minimum participation criteria and accountability 

• Effective program management and maintenance 


40 



The Public Data Initiative boasts the participation of 120 law enforcement 
agencies, and the limitations of the data can be seen through the example of the Camden, 
New Jersey, Police Department, which was one of the original 21 agencies to participate 
in the program. In May 2015, President Obama visited Camden to discuss the program’s 
merit and direction. The White House wrote about Camden’s participation in a 
subsequent blog entry on its website: 

By upgrading its technology practices, the Camden County PD will have 
more efficient data supply chains, and will be better positioned to use that 
data to improve its internal operations and to identify and solve policing 
problems in a timelier manner. The lessons learned in Camden can help 
law enforcement around the country both by example and also directly 
since some of the development work can be shared though open source 
best practice. 84 

Despite the fanfare, the Public Safety Open Data Portal’s webpage listing of 
“Participating Agencies” depicts the hyperlink to Camden’s data portal as “Not Currently 
Identifiable.” 85 Moreover, a basic Google search for the term, “Camden New Jersey 
Open Data Portal,” reveals a hyperlink to Camden’s open data portal 
(http://camdencountynj.ccdpw.opendata.arcgis.com/). The landing page highlights eight 
recently added datasets, none of which are law enforcement related. An “Explore County 
Data” link reveals the same eight datasets when accessed. A visible search bar is 
available at the top of the page and when the keyword “police” is typed, the results return 
the message, “No Datasets Found.” 

B. RECOMMENDATIONS 

The most evident deficiency identified in this study involves the transparency 
initiative and Open Government program, specifically with regard to the PDI and 
gathering of police related data. The PDI needs to establish stricter guidelines and 
compliance for participating agencies. For those law enforcement agencies wishing to 
participate, minimum guidelines should be established and met. Additionally, those 


84 Smith and Austin, “Launching the Police Data Initiative.” 

85 “The Police Data Initiative (PDI).” 


41 




agencies not meeting those standards should be eliminated from participating in the 
program. 

To date, conversations involving the Ferguson Effect are usually associated with 
crime statistics. This thesis suggests that less emphasis should be placed on crime 
correlations and more value placed on de-policing and anxieties being experienced by 
officers. Despite the large amount of anecdotal evidence suggesting law enforcement is 
becoming less proactive, the limited data available does not seem to show a significant 
amount of de-policing following the events in Ferguson. More data is urgently needed to 
help determine whether de-policing is in fact a widespread trend. 

Another method for further study is to conduct surveys of law enforcement 
officers to gauge police self-perception and how those perceptions have influenced 
performance of duties post-Ferguson. As this thesis was being concluded, one such 
survey was completed by the Pew Research Center. The study found, “majorities of 
police officers say that recent high-profile fatal encounters between black citizens and 
police officers have made their jobs riskier, aggravated tensions between police and 
blacks, and left many officers reluctant to fully carry out some of their duties.” 86 The 
national survey is “one of the largest ever conducted” and “draws on the attitudes and 
experiences of nearly 8,000 policemen and women from departments with at least 100 
officers.” 87 While important, survey data can only afford limited insight. Ultimately, 
evidence concerning the Ferguson Effect must be centered on hard data surrounding 
police activity and productivity, and it is perhaps the primary finding of this thesis that 
reveals such data is grossly inadequate. 

C. CONCLUSION 

Theoretically, big data should offer possible correlations between police 
productivity and the increase or decrease in crime. Fikewise, data relating to productivity 
should offer independent insight into the existence of a Ferguson Effect regardless of 

86 “Behind the Badge,” January 11, 2017, http://www.pewsocialtrends.org/2017/01/ll/behind-the- 
badge/. 

87 Ibid. 


42 



reported crime. The reality is that access to this type of big data is not readily available to 
the public. If the open data project matures into its intended vision, such data may 
provide insight into the existence of a Ferguson Effect. Currently, however, the data is 
insignificant and lends no insight whatsoever. 

The FBI acknowledges a need for better data collection and “recently announced 
that the National Incident Based Reporting System, or NIBRS, would become the 
Uniform Crime Reporting Program (UCR) standard by January 1, 2021.” 88 NIBRS was 
implemented in 1989 to accommodate the increased volume of information being 
collected about crime. The system “provides for 52 offense classifications,” and provides 
better insight into national crime. Due to the voluntary nature of NIBRS, participation 
from law enforcement has grown at a slow pace, which in turn, has “caused the FBI to 
maintain both the SRS and NIBRS.” 89 Moreover, “the FBI will conduct a NIBRS 
modernization study to “assess the current law enforcement agencies practices and 
evaluate a possible updating of NIBRS.” 90 

This study aspired to close those gaps and offer a more complete picture of 
national sentiment by employing big data analytics for police productivity. Fimitations in 
data prevented the research questions from being answered but highlighted a need for 
better data collection methods. 


88 James Kent, “A Long Time Coming: The FBI’s Goal of Pushing Uniform Crime Reporting into the 
21st Century,” Homeland Security Digital Library, December 15, 2016, https://www.hsdl.org/c/a-long- 
time-coming-the-fbis-goal-of-pushing-uniform-crime-reporting-into-the-21st-century/. 

89 Ibid. 

90 Ibid. 


43 



THIS PAGE INTENTIONALLY LEFT BLANK 


44 



LIST OF REFERENCES 


Allocca, Sean. “The Facts behind the ‘Ferguson Effect’.” Forensic Magazine, October 
29, 2015. http://www.forensicmag.com/article/2015/10/facts-behind-ferguson- 
effect. 

Blakely, Rhys. “U.S. Violent Crime Surges to 1970s Levels after Race Riots.” The Times , 
June 3, 2015. http://www.thetimes.co.uk/tto/news/world/americas/article4458700. 
ece. 

Burlington Vermont. “Burlington Police Department Arrests.” Accessed October 9, 2016. 
https://www.burlingtonvt.gov/police/data/arrests. 

-. “Burlington Police Department Calls for Service.” Accessed October 9, 2016. 

https://www.burlingtonvt.gov/police/data/dashboard. 

-. “Burlington Police Department Traffic Stops.” Accessed October 8, 2016. 

https://www.burlingtonvt.gov/police/data/traffic. 

Byers, Christin. “Crime Up after Ferguson and More Police Needed, Top St. Louis Area 
Chiefs Say.” St. Louis Post Dispatch, November 15, 2014. http://www.stltoday. 
com/news/local/crime-and-courts/crime-up-after-ferguson-and-more-police-need 
ed-top-st/article_04d9f99f-9a9a-5 Ibe-a231-1707a57b50d6.html. 

City of Philadelphia. “Philadelphia Police Department Violent Crime.” Accessed October 
9. 2016. https://data.phila.gov/api/views/sspu-uyfa/rows.csv?accessType=DOWN 
LOAD. 

Comey, James B. “Law Enforcement and the Communities We Serve: Bending the L ines 
toward Safety and Justice.” Speech, University of Chicago Law School, Chicago, 
Illinois, October 23, 2015. https://www.fbi.gov/news/speeches/law-enforcement- 
and-the-communities-we-serve-bending-the-lines-toward-safety-and-justice. 

Davey, Monica, and Mitch Smith. “Murder Rates Rising Sharply in Many U.S. Cities.” 
New York Times, August 31, 2015. http://nyti.ms/lUiHLnX. 

Department of Justice. Department of Justice Report Regarding the Criminal 

Investigation into the Shooting Death of Michael Brown by Ferguson Missouri 
Police Officer Darren Wilson (Memorandum). Washington, DC: Department of 
Justice, 2014. https://www.justice.gov/sites/default/files/opa/press-releases/ 

attachments/2015/03/04/doj_report_on_shooting_of_michael_brown_l.pdf. 

Lears, Darryl. “In Milwaukee, Weak Evidence for ‘Lerguson Effect’.” The Washington 
Post, December 6, 2015. http://search.proquest.com/docview/17393457767acc 
ountid= 12702. 


45 



Federal Bureau of Investigation. “2015 Crime in the United States, Table 78, 

Pennsylvania, Full-Time Law Enforcement Employees by City, 2015.” Accessed 
October 11, 2016. https://ucr.fbi.gOv/crime-in-the-u.s/2015/crime-in-the-u.s.- 
2015/tables/table-78/table-78-state-pieces/table-78-pennsylvania-new. 

-. “2015 Crime in the United States, Table 78, Vermont, Full-Time Law 

Enforcement Employees by City, 2015.” Accessed October 11, 2016. https:// 
ucr.fbi.gov/crime-in-the-u. s/2015/crime-in-the-u. s. -2015/tables/table-7 8/table-78- 
state-pieces/table_78_full_time_law_enforcement_employees_vermont_by_ 
cities_2015.xls. 

-. “2015 Crime in the United States, Table 80, Maryland, Full-Time Law 

Enforcement Employees by Metropolitan and Nonmetropolitan Counties, 2015.” 
Accessed October 11, 2016. https://ucr.fbi.gov/crime-in-the-u.s/2015/crime-in- 
the-u.s.-2015/tables/table-80/table-80-state-pieces/table_80_full_time_law_en 
forcement_employees_maryland_by_metropolitan_and_nonmetropolitan_countie 
s_2015.xls. 

-. “Latest Crime Statistics Released, Increase in Violent Crime, Decrease in 

Property Crime.” September 26, 2016. https://www.fbi.gov/news/stories/latest- 
crime-statistics-released. 

-. “UCR, 2015 Crime in the United States, Violent Crime.” Accessed October 12, 

2016. https://ucr.fbi.gov/crime-in-the-u.s/2015/crime-in-the-u.s.-2015/offenses- 
known-to-law-enforcement/violent-crime. 

Gass, Henry. “FBI Chief Pins Crime Surge on ‘Ferguson Effect.’ Are Cops Afraid to Do 
Their Jobs?.” The Christian Science Monitor , October 26, 2015. http://www.es 
monitor .com/US A/Justice/2015/1026/FBI-chief-pins-crime-surge-on-Fergu son- 
effect. -Are-cops -afraid-to -do-their-j ob s. 

Gillman, Zeke. “FBI Director James Comey on the ‘Ferguson Effect’ at the Law School.” 
The Chicago Maroon, November 10, 2015. https://chicagomaroon.com/2015/ll/ 
10/fbi-director-james-comey-on-the-ferguson-effect-at-the-law-school. 

Google Maps. “Directions from Ferguson, Missouri to Burlington, Vermont.” Google 
Maps. Accessed October 11, 2016. https://goo.gl/maps/vL4zb6LN8KL2. 

-. “Directions from Ferguson, Missouri to Montgomery County, Maryland.” 

Google Maps. Accessed October 11, 2016. https://goo.gl/maps/8dlFyPZ6DBp. 

-. “Directions from Ferguson, Missouri to Philadelphia, Pennsylvania.” Accessed 

October 11, 2016. https://goo.gl/maps/xRUbJTYQ5cJ2. 

Gossett, Stephen. “Top Cop Says Beaten Officer Was Afraid to Shoot Due to Media 

Attention.” Chicagoist, October 6, 2016. http://chicagoist.com/2016/10/06/top_ 
cop_says_beaten_officer_was_afr.php. 

46 



International, M., N. York, and T. Paris. “A Year Later, Ferguson Remains Unsettled: 

Residents Dispute Extent of Changes Following Racially Charged Death.” 2015. 
http ://search.proque st .com/doc vie w/1701637793?accountid= 12702. 

Johnson, Patrik. “Just What is the ‘Ferguson Effect’? It Depends on How You View 
Police.” The Christian Science Monitor, June 12, 2015. http://www.csmonitor. 
com/US A/Justice/2015/0612/Just-what-is-the-Ferguson-effect-It-depends-on- 
how-you-view-police. 

Kasperowicz, Pete. “Fynch: I Have Not Seen Ferguson Effect.” Washington Examiner, 
July 1, 2016. http://www.washingtonexaminer.com/lynch-i-have-not-seen-fer 
guson-effect/article/2595445. 

Kent, James. “A Fong Time Coming: The FBI’s Goal of Pushing Uniform Crime 

Reporting into the 21st Century.” Homeland Security Digital Fibrary, December 
15, 2016. https://www.hsdl.org/c/a-long-time-coming-the-fbis-goal-of-pushing- 
uniform-crime-reporting-into-the-21 st-century/. 

FichtBlau, Eric. “Officials Debate Whether ‘Ferguson Effect’ Is Real.” The New York 
Times, November 4, 2015. http://www.nytimes.com/2015/ll/05/us/politics/ 
officials-debate-effect-of-scrutiny-on-police.html. 

Mac Donald, Heather. “Officer Beaten by a Convicted Felon Hesitated for Fear of Being 
Called Racist: Welcome to Post-Ferguson Policing.” National Review, August 16, 
2105. http://www.nationalreview.com/article/422605/Birmingham-cop-beaten- 
unconscious-feared-racism-charge. 

-. “The Nationwide Crime Wave Is Building.” The Wall Street Journal, May 23, 

2016. http://www.wsi.com/articles/the-nationwide-crime-wave-is-building-146 
4045462. 

-. “The New Nationwide Crime Wave.” The Wall Street Journal, May 29, 2015. 

http://www.wsj.com/articles/the-new-nationwide-crime-wave-1432938425. 

McClam, Erin. “Ferguson Cop Darren Wilson Not Indicted in Shooting of Michael 

Brown.” NBC News, November 25, 2014. http://www.nbcnews.com/storyline/ 
michael-brown-shooting/ferguson-cop-darren-wilson-not-indicted-shooting- 
michael-bro wn-n25 5 391. 

Montgomery County Maryland. “Montgomery County Police Department Vehicle 

Stops.” Accessed October 8, 2016. https://data.montgomerycountymd.gov/Public- 
Safety/Traffic-Violations/4mse-ku6q. 

-. “Montgomery County Police Department Violent Crime.” Accessed October 9, 

2016. https://data.montgomerycountymd.gov/api/views/4mse-ku6q/rows.csv? 
accessType=DOWNFOAD. 


47 



Morgan, Stephen L., and Joel A. Pally. Ferguson, Gray, and Davis, An Analysis of 

Recorded Crime Incidents and Arrests in Baltimore City, March 2010 through 
December 2015. Baltimore, MD: Johns Hopkins University, 2016. http://socweb. 
soc.jhu.edu/faculty/morgan/papers/MorganPally2016.pdf. 

OpenDataPhilly. “Philadelphia Police Department Vehicle Stops.” Accessed October 18, 
2016. https://www.opendataphilly.org/dataset/vehicle-pedestrian-investigations/ 
resource/e02050cl-51a4-48ff-80d2-88b290047837. 

Pew Research Center. “Behind the Badge.” January 11, 2017. http://www.pewsocialtre 
nds. org/2017/01/11 /behind-the -badge/. 

Police Foundation. The. “Public Safety Open Data Portal.” Accessed October 9, 2016. 
http ://public safetydataportal .org/. 

Public Safety Open Data Portal. “The Police Data Initiative (PDI).” Accessed November 
18, 2016. https://publicsafetydataportal.org/participating-agencies/. 

Pyrooz, David C., Scott H. Decker, Scott E. Wolfe, and John A. Shjarback. “Was There a 
Ferguson Effect on Crime Rates in Large U.S. Cities?” Journal of Criminal 
Justice 46 (2016): 1-8. 

Rosenfeld, Richard. Documenting and Explaining the 2015 Homicide Rise: Research 
Directions. Washington, DC: U.S. Department of Justice, Office of Justice 
Programs, National Institute of Justice, 2016. https://www.ncjrs.gov/pdffilesl/nij/ 
249895.pdf. 

-. Was There a “Ferguson Effect” on Crime in St. Louis? Washington, DC: The 

Sentencing Project, 2015. http://sentencingproject.org/wp-content/uploads/2015/ 
09/Ferguson-Effect.pdf. 

Rushin, Stephen, and Griffin Edwards. “De-Policing.” DRAFT, Cornell Law Review 102 
(Forthcoming, 2017): 1-66. 

Smith, Megan, and Roy L. Austin, Jr. May 18, 2015 (6:00 a.m.). “Launching the Police 
Data Initiative.” The White House Blog. Accessed October 11, 2016. https://www. 
whitehouse.gov/blog/2015/05/18/launching-police-data-initiative. 

Sohoni, Tracy, and Charis Kubrin. “Is the Ferguson Effect a Myth?.” The Baltimore Sun , 
September 10, 2015. http://www.baltimoresun.com/news/opinion/oped/bs-ed- 
ferguson-effect-20150910-story.html. 

Tableau. “Tableau for Students.” Accessed August 15, 2016. http://www.tableau.com/ 
ac ademic/student s. 

United States Census Bureau. “QuickFacts, Burlington City, Vermont.” Accessed 

October 13, 2016. http://www.census.gov/quickfacts/table/PST045215/5010675. 


48 



-. “QuickFacts, Montgomery County, Maryland.” Accessed October 13, 2016. 

http ://www .census .gov/quickfacts/table/PST045215/24031. 

-. “QuickFacts, Philadelphia City, Pennsylvania.” Accessed October 13, 2016. 

http://www.census.gov/quickfacts/table/PST045215/4260000. 

White House, The. “Open Government Directive.” Accessed October 11, 2016. https:// 
www.whitehouse.gov/open/documents/open-government-directive. 

-. “Open Government Initiative.” Accessed October 11, 2016. https://www.whhe 

house.gov/open/about. 

-. “Transparency and Open Government.” Accessed October 11, 2016. https:// 

www.whitehouse.gov/the_press_office/TransparencyandOpenGovernment. 

Wolfe, Scott E., and Justin Nix. “The Alleged ‘Ferguson Effect’ and Police Willingness 
to Engage in Community Partnership.” Law and Human Behavior 40, no. 1 
(2016): 1-10. http://dx.doi.org/10.1037/lhb0000164. 

Yin, Robert K. Case Study Research: Design and Methods. Los Angeles: Sage 
Publications, 2009. 


49 



THIS PAGE INTENTIONALLY LEFT BLANK 


50 



INITIAL DISTRIBUTION LIST 


1. Defense Technical Information Center 
Ft. Bel voir, Virginia 

2. Dudley Knox Library 
Naval Postgraduate School 
Monterey, California 


51