U.S. Department of Justice 
Office of Community Oriented Policing Services 

DOJ Seal

COPS Logo

Theft From Cars in Center City Parking Facilities – A Case Study


by 
Ronald V. Clarke and Herman Goldstein 

www.cops.usdoj.gov 

THEFTS FROM CARS IN CENTER-CITY PARKING FACILITIES: 
A CASE STUDY IN IMPLEMENTING PROBLEM-ORIENTED POLICING 

by 
Ronald V. Clarke and Herman Goldstein 


Table of Contents 

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 

Project Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 

Defining the Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 

The Police View of the Problem . . . . . . . . . . . . . . . . . . . . . . .11 

From Mapping Hot Spots to Pinpointing Risks . . . . . . . . . . . . . . . . 13 

The Security of the Uptown Decks and Lots . . . . . . . . . . . . . . . . . 17 

The Search for Solutions. . . . . . . . . . . . . . . . . . . . . . . . . . 21 

Selling the Solutions. . . . . . . . . . . . . . . . . . . . . . . . . . . .25 

An Unexpected Decline in Thefts . . . . . . . . . . . . . . . . . . . . . . 29 

Lessons of the Case Study . . . . . . . . . . . . . . . . . . . . . . . . . 33 

Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . 39 

Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41 

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45 



Abstract 

This paper describes a problem-oriented policing project, extending 
over a period of more than two years, which was designed to reduce 
thefts from cars parked in the center city of Charlotte, North 
Carolina. A progressive tightening of focus led to a detailed analysis of 
the risks of theft, and the associated security features, in the 39 decks 
and 167 surface lots in the center city. This analysis showed (1) that 
risks of theft were much greater in lots than in decks and (2) that 
higher risks of theft in lots were associated with inadequate fencing, 
poor lighting, and the absence of attendants. These data played an 
important part in obtaining the agreement of lot owners and 
operators to make security improvements. Before most of these 
improvements had been made, however, thefts in the lots began to 
decline, possibly as the result of more focused patrolling by police and 
security personnel. The paper concludes with a discussion of the 
difficulties encountered by police in undertaking problem-oriented 
projects, and of ways to help them meet these difficulties. 


Acknowledgments 

This project was completed under an Advancing Community Policing 
Grant (#97-0C-WX-0060) made to the Charlotte-Mecklenburg Police 
Department by the U.S Department of Justice Office of Community 
Oriented Policing Services. 

Because the integration, analysis, and mapping of data constituted 
such an important part of this exploration, the project leaned heavily 
on the crime analyst assigned to the David One district, Matthew 
White. He also prepared the maps included in this report. In the data 
collection stage, crime analysts Monica Nguyen and Michael 
Humphrey lent a hand and, later, Kristin Knight, assisted in some of 
the statistical analysis. 

It was the captain of David One, Jerry Sennett, who initially proposed 
the project as a case study in problem-oriented policing. He 
subsequently participated in the various meetings at which the data 
were analyzed and possible solutions discussed, made the connections 
to the various groups that would be important in carrying out new 
strategies, and also took the lead in the several presentations of the 
study's findings. Captain Sennett assigned a member of his staff to 
monitor the project and to acquire some needed information–first 
Sergeant Craig "Pete" Davis and, more recently, Sergeant Harold 
Medlock. A team of two police officers, Anthony Crawford and 
Veronica Foster, carried out the enormous task of acquiring, through 
on-site inspections, the detailed information on existing parking 
facilities, including the counting of the number of parking spaces in 
each facility. Subsequently Officers Crawford and Robert Vandergrift 
conducted CPTED analyses of selected parking facilities used in the 
PowerPoint presentation. 

Steve Ward, a senior District Attorney who is (in a unique 
arrangement in American policing) assigned full time to working as an 
adviser within the CMPD, participated in all of the meetings held over 
the life of the project, and in formulating the recommendations for 
earlier intervention. 

Dennis Nowicki, the former Chief of the CMPD, initially suggested 
taking on several case studies in problem-oriented policing as a way of 
illustrating what was involved in a comprehensive carrying out of the 
concept. From the time of his appointment, shortly after the project 
got underway, the project has received strong support from the 
current Chief, Darrel Stephens. The former Director of Research and 
Planning, Dr. Richard Lumb, was generous in his arrangements for the 



allocation and scheduling of his staff's time, as was current Acting 
Director, John Couchell. This report on the project was greatly 
facilitated by detailed notes taken on early project meetings by Officer 
Lisa Carriker. Finally, James LeBeau of Southern Illinois University 
provided us with valuable statistical advice. 



Introduction 

The concept of problem-oriented policing (Goldstein, 1979, 1990) has 
been widely endorsed by the American and British police. In the 
United States, federal grant programs, supported by the 1994 Federal 
Crime Act, have promoted the concept and many police departments 
have made a commitment to it in one form or another (Scott, 2000). 
The annual problem-oriented policing conference sponsored by the 
Police Executive Research Forum (PERF) attracts between 1,000 and 
1,500 delegates per year, while submissions for the Herman Goldstein 
Award for Excellence in Problem-Oriented Policing total about 100 
per year (Scott and Clarke, 2000). In England and Wales, all 43 police 
forces claim to be undertaking some form of problem-oriented 
policing (Read and Tilley, 2000). The British police also have an annual 
conference devoted to advancing problem-oriented policing projects 
and have the Nick Tilley award to recognize outstanding projects. 

Despite these endorsements, advocates of problem-oriented policing– 
we included–have continued to express disappointment with the 
projects reported in its name (Clarke, 1997, 1998; Goldstein, 1994a,b, 
1996,a,b; Read and Tilley, 2000; Scott, 2000; Scott and Clarke, 2001). 
Many are little more than well-intentioned efforts to improve 
community relations, barely recognizable as problem-oriented policing. 
Where they do include problem-solving elements, the problems may 
be small-scale (sometimes confined to a single address), analysis may 
be perfunctory, and evaluation often consists of testimonials from 
citizens or the local newspaper. In the few reported cases of larger-
scale projects fitting the definition of problem-oriented policing, 
analysis seldom goes deeper than looking at calls for service data or 
statistics of reported crimes, responses frequently depart little from 
traditional enforcement strategies, and evaluation rarely explores 
alternative explanations for any drops in crime. 

The commentators cited above have not been led to conclude that this 
experience negates the value of the concept or is indicative of theory 
failure–perhaps because a sufficient number of successful problem-
oriented projects have been published to sustain faith in the concept.1 
Rather, they have assumed that the disappointing experience of 
applying problem-oriented policing results from implementation 
failure, which they attribute to a variety of sources. We will not be 
departing from this position, but will anchor our discussion of 
implementation difficulties in a detailed description of one problem-
oriented project in which we have been involved as consultants–an 
effort to reduce theft from cars in the center city of Charlotte, North 
Carolina. 

1 Particularly when situational prevention 
projects are included in the count. (When 
practiced by police situational prevention is 
indistinguishable from problem-oriented 
policing, Clarke, 1997) 



It is rare that those who contribute to the development of theories 
and concepts have the opportunity to play as active a role as we did in 
an actual effort at implementation, and to observe the entire 
implementation process so closely. Such a relationship has some 
drawbacks and hazards. Our involvement made the project atypical. 
One cannot, as a result, generalize from the project, either in weighing 
the results or in planning a replication, without allowing for this 
involvement. And while we tried to remain objective, we are 
vulnerable to the charge that our involvement compromised our 
objectivity. But the relationship had its unique benefits. By joining with 
the project team in muddling through the many complex issues that 
were encountered, we had the opportunity to gain unique insights into 
the difficulties of implementation. Thus, while we offer this case study 
as one more commentary on the state of problem-oriented policing, 
we believe it offers a somewhat unique and different perspective. As 
will be seen, it claims mixed results–some successes and some failures 
–both of which have lessons for the future. 


Project Background 

Our involvement in this project began with a request by Chief Dennis 
Nowicki to Goldstein, who was serving as an in-house adviser to the 
Charlotte-Mecklenburg Police Department (CMPD) under an 
Advancing Community Policing grant from the U.S. Department of 
Justice's Office of Community Oriented Policing Services (the COPS 
Office). As part of his work with the CMPD, he was asked to review 
the department's efforts to implement problem-oriented policing. The 
department had invested substantially in training in problem-oriented 
policing and in urging line officers to identify and address problems. 
Goldstein's review led him to conclude that more progress would be 
made in implementing the concept if time and resources could be 
focused on just a few projects in which an intensive, careful effort 
would be made to address a specific substantive problem. These 
projects could then be used as illustrations of the type of problem-
oriented policing project to which others could aspire. 

Captain Jerry Sennett and his officers in the David One district, which 
encompasses Charlotte's center city,2 suggested that a suitable 
candidate for this kind of intensive effort would be a project focused 
on thefts from parked cars in their district. These larcenies from autos 
(or LFAs) constituted a large proportion of all crimes reported in 
David One and bringing them down would make a substantial dent in 
the district's crime statistics. This was an important objective not only 
for the district captain, but also for the CMPD, given the significance 
attached to controlling crime in the recently redeveloped and 
revitalized center city (known locally as Uptown). LFAs had been 
resistant to control through conventional police operations and, in 
fact, were increasing at a rate faster than economic growth. Between 
1998 and 1999, they jumped from 1,011 to 1,313. The District One 
officers knew that these statistics were likely to underestimate the 
problem because victimization surveys have consistently found that 
only about 30 to 50 percent of LFAs are reported to the police. 
Moreover, they also believed that LFAs fuelled drug and alcohol use 
by the offenders involved. 

2 The David One district (one of twelve CMPD 
districts) is split roughly into two parts - the 
northern part consisting of Charlotte's Uptown, 
and a larger inner city area to the west and 
south consisting of mixed residential and 
industrial development.

Goldstein agreed that the David One LFAs could provide a suitable 
focus for the kind of project he had in mind and, soon after, he 
invited Clarke to join him in helping with the analysis and in 
identifying possible preventive measures. In a series of short visits 
extending over more than two years, Clarke and Goldstein met 
regularly with Captain Sennett and several of his officers and the 
crime analysts assigned to the district–a group that came to be 
referred to as the project team.3 Also attending most of these 
  



meetings was Steve Ward, a senior Assistant District Attorney who 
was assigned to work as an adviser within the CMPD and who was 
supportive of efforts to increase the effectiveness of the police while 
making more discrete use of the criminal justice system. The role 
played by Goldstein and Clarke was essentially consultative–to explain 
the process of problem-oriented policing, to help talk through the 
difficulties encountered during the project, to discuss the experience 
gained elsewhere in dealing with theft from vehicles, to raise points for 
further inquiry or action, and to make suggestions about data analysis. 
In tandem with the project team, this work resulted in refining the 
focus of the project, obtaining a better understanding of the problem, 
selecting and gaining agreement to solutions, and assessing 
effectiveness. 

3 At the same time, they also served as 
consultants to problem-oriented projects in 
other units of the CMPD. 



Defining the Problem 

The first meetings of the project team were largely taken up with 
defining the problem. It was soon decided to focus on the Uptown 
where in 1998 just over 50 percent of David One's LFAs were 
reported, and where, despite heavy levels of policing,4 most of the 
increase in LFAs had occurred. Uptown is a clearly defined 
geographical area of about one square mile, encircled by a freeway 
system. The area covers about 170 city blocks, which mostly hold 
office buildings, hotels, and associated retail and parking facilities. In 
the northern corner is a well-established, affluent residential district, 
and in the eastern corner is a second residential district, consisting of 
newly-built condominiums, on land which had been cleared of low 
income housing that had deteriorated in its quality. The west corner 
holds the Ericsson Stadium, home of the Carolina Panthers. A trolley 
line, that had fallen into disuse, but is now being restored, runs across 
the area in a Northeast/Southwest direction (see Figure 1 in 
Appendix). 

Each business day, Uptown accommodates some 50,000 commuters 
who travel into the city by car. Most of these cars are parked in decks 
or surface lots scattered throughout the area. This pattern is reflected 
in LFAs, 83 percent of which occurred in decks or lots in 1998, and 
only 17 percent of which occurred in residential property or on the 
streets. Hot spot mapping (see Figure 2 in Appendix) by Monica 
Nguyen, the crime analyst originally assigned to the project, showed 
that LFAs were concentrated in the center of Uptown where 
residences and street parking are largely absent, but where, in support 
of the businesses and nightlife, there are many decks and lots. In light 
of these facts, it was decided to tighten further the project's focus to 
deal only with Uptown LFAs occurring in lots and decks.5 

4 The center city businesses have provided 
funds that enable the CMPD to assign ten 
officers to foot and bike patrol that are in 
addition to the number that would otherwise 
have been assigned based on the countywide 
standards for allocating police personnel. 

5 For the purposes of this study, decks are 
either freestanding multilevel parking structures 
or parking garages/parking floors belonging to 
multilevel office buildings. Lots are open, 
surface-level parking facilities. 


The Police View of the Problem 

In the course of dealing with the problem over the previous few years, 
the police officers had developed their own view of the causes and 
potential solutions. They tended to blame a combination of careless 
victims, lenient courts, and offenders who were supporting drug or 
alcohol habits. In more detail, their diagnosis comprised the following 
elements: 

1. LFAs are quick and easy to commit. Most LFAs in Uptown are 
committed by breaking a window (which is often shattered using a 
spark plug) and taking items left inside the car. Thefts may take 
less than 30 seconds to commit and may not be discovered until 
several hours later when commuters return to their cars. Without 
having any need to touch the car, fingerprints are rarely left at the 
scene. Because of the large area to be patrolled, police rarely catch 
an offender in the act. 

2. In a congested criminal justice system, LFAs are not considered 
serious offenses by the courts and tend to be treated leniently. The 
few arrests that are made, therefore, rarely result in offenders 
being taken off the streets, which means they are free to repeat 
the same offense. 

3. The victims are office workers in the day and customers of clubs 
and restaurants at night. Many victims must share the blame for 
LFAs because they leave items, such as cell phones, compact discs, 
and clothes, inside the car in plain view. 

4. Offenders can find a ready market for items they steal in the 
numerous pawnshops and known street drug markets located in 
areas close to the Uptown. (Officers spoke of having identified 14 
such locations.) 

5. Offenders fall into three main groups: (1) habitual street criminals 
with drug habits, (2) petty offenders with alcohol problems who 
spend their days hanging around Uptown, and (3) transients from 
the city's homeless shelters, many of whom are also alcoholics. 

6. The transients comprise the largest group of offenders and LFAs 
are committed as part of their daily routine. This begins with their 
trek from the winter homeless shelter, in the west of the Uptown 
area, along the trolley line (which, until recently, was not in use), 
to the soup kitchen, on the other side of Uptown, where they eat 
their lunch (see Figure 1 in Appendix). Numerous surface lots and 
decks border the trolley line and it is easy for transients to find 
something worth stealing in a car and then escape along the 
trolley route. They sell items they have stolen to drug dealers and 
pawnshops in the general location of the soup kitchen and use the 
proceeds for alcohol. After midday, the transients follow the 
reverse course and sell stolen items at locations near the homeless 
shelter. En route they might detour to the convenience store just 
south of the trolley line to purchase cheap alcohol. 




This view governed the strategies pursued by the police. They had 
worked with cooperating suspects to identify other suspects for arrest. 
They had attempted to use territorial restrictions as part of sentences 
for convicted parking lot thieves. They had tried to build cases to 
enable them to prosecute certain persistent offenders as career 
criminals, which could result in lengthy prison sentences. They had 
performed surveillance at high-risk locations. And they had 
encouraged the placement of "no trespassing" signs in parking 
facilities to allow the opportunity for officers to detain and question 
suspects. Apart from these enforcement efforts, they had prompted 
media stories about not leaving valuables in cars, they had sought to 
initiate a business watch program in the downtown and they had 
placed warning notices on cars with tempting items left in plain view. 

These efforts had met with little success. Few offenders were arrested 
and successfully prosecuted and LFAs in Uptown continued to rise 
(from 513 in 1998 to 814 in 1999). Even so, some of the David One 
officers argued that more vigorous pursuit of the strategies would lead 
to better results. In particular, they wanted to see a more intensive 
media campaign directed to careless victims, greater efforts to arrest 
offenders through more direct surveillance and through work with 
cooperative suspects, and a stronger commitment from the District 
Attorney's office to prosecute alleged offenders and seek harsher 
sentences for those convicted. 

Most of the David One officers initially hoped that the attention 
focused on the problem through the project would strengthen their 
hands in pressing for these strategies. They expressed some frustration 
when their proposed solutions were questioned and an effort was 
launched to deepen the analysis. In the subsequent discussions that 
opened that analysis, the ability of the police to substantially increase 
arrests for LFAs was challenged (see Hesseling, 1995), and the 
likelihood that the District Attorney's office could secure harsh 
sentences was questioned.6 Studies were described that had found few 
benefits from local lock-your-car campaigns (Barthe, 2000; Burrows 
and Heal, 1980; Riley, 1980). In terms of the classic crime triangle 
(Spelman and Eck, 1989), it was pointed out that the police view of 
the problem was focused mostly on offenders and victims, rather than 
on the locations, i.e. the parking facilities and, in particular, the 
security of those facilities. 

6 We were supported in this latter argument by 
Steve Ward.

As a result of these initial discussions, it was agreed that, while the 
officers would continue to pursue the strategies they had developed, 
they would, at the same time, assist with analysis of parking security. 


From Mapping Hot Spots to 
Pinpointing Risks 

The hot spot mapping that showed LFAs were concentrated around 
parking facilities in the center of Uptown was of limited value in 
further analysis because each hot spot covered several blocks, 
containing not only a large number of parking facilities, but a broad 
diversity of facilities as well. However, in discussing the maps, officers 
would occasionally identify particular facilities that they believed 
accounted for most of the LFAs. Many of these were simply the 
largest facilities, where, because of their size, one might expect to 
encounter more LFAs, but some facilities identified as troublesome 
were quite small. This suggested that there were features of the 
design, location, or management of the facilities that might make the 
vehicles parked in them especially vulnerable to LFAs. 

It therefore became important to learn more about the parking 
facilities, but many of these were not even shown in the CMPD maps 
of Uptown, which had become outdated as a result of the 
construction boom driven by the city's thriving banking industry. New 
lots had been created as old buildings had been demolished pending 
redevelopment, and new buildings had sprung up on the sites of 
former lots. Even the maps maintained by the city's planning 
department did not show every facility. Consequently, it was decided 
to undertake a comprehensive inventory of parking facilities and to 
count the spaces in each. 

This was a major undertaking, absorbing the resources available to the 
project for a considerable period of time. The crime analyst who had 
taken over responsibility for the project, Matthew White, 
supplemented the available information from the planning department 
with the detailed knowledge of the David One officers regularly 
assigned to the area and with information obtained from a new aerial 
survey of the uptown area. This resulted in the identification of 206 
separate parking facilities with more than 20 parking spaces–39 decks 
and 167 surface lots. 



Table 1: Deck and Lot Size, Charlotte's Uptown, 2000 

Spaces 
Decks 
Lots 

0-49 
0 
52 

50-99 
2 
39 

100-199 
5 
43 

200-299 
3 
17 

300-499 
10 
14 

500-999 
12 
1 

1000+ 
7 
1 

Total 
39 
167 

The David One officers assigned to the project, Anthony Crawford 
and Veronica Foster and the analyst, White, undertook to count all of 
the spaces in all of the parking facilities–a laborious enterprise. 
Eventually it was established that there was a total of 42,574 spaces in 
the 206 facilities, 22,373 of which were in decks and 20,201 in lots. 
Table 1 summarizes information about the size of the parking 
facilities. 

With these figures in hand, it should have been easy to calculate the 
rate of LFAs per facility, but unfortunately LFAs were not recorded 
for individual parking facilities, but only for the block on which these 
stood (though location codes distinguished LFAs occurring in parking 
facilities from those occurring from cars parked on the street or on 
private property). This was because victims making reports could 
usually identify the block where the car was parked, but not the 
particular parking facility. There was no difficulty in assigning the theft 
to that parking facility when it was the only one on the block. Nor was 
there any difficulty in assigning thefts to particular facilities where the 
block contained a lot and a deck because the location code permitted 
these to be distinguished. Rather, the difficulty arose when blocks 
contained more than one lot or more than one deck, which was the 
case for more than half of the blocks. In these cases, LFAs were 
sorted, using their location codes, into those occurring in lots and 
those occurring in decks before calculating separate rates of LFAs, per 
parking space, for the decks and for the lots. This meant that all the 
lots on the block shared the same rate of LFAs, which might be 
different from the rate for the decks (or deck) in that block. 

Once the rate of LFAs per parking space had been determined for 
each facility, comparisons could be made of their theft risks. Two facts 
stood out in the results of this exercise. First, it was apparent that the 


parking facilities bordering the trolley line and the disused rail tracks 
to the west of Uptown had generally higher rates of theft (see Figure 
3 in Appendix). This lent support to the police analysis, which had 
implicated these as conduits for the transients in their daily 
movements about the city. Second, it was found that the rate of LFAs 
per parking space was much higher for lots than for decks. In 1999, 93 
LFAs occurred in decks, which between them had 22,373 spaces (a 
rate of 4.1 LFAs per 1,000 parking space). But 510 LFAs occurred in 
lots, which had a total of 20,201 spaces (a rate of 25.3 LFAs per 1,000 
spaces). The risk of LFAs per parking space for lots was thus about 
six times greater than for decks. 


People are sometimes fearful in garages and decks, especially when 
they are out of view of attendants and nobody else is around. They 
often assume therefore that their persons and their cars are more 
vulnerable to crime. That fewer thefts were found to occur in decks 
than lots would probably surprise them. In fact, the same result was 
obtained in research undertaken in London where it was explained in 
terms of the lack of security in many lots (Webb et al., 1992). Few lots 
have an attendant, they often lack adequate lighting, and many lack 
natural surveillance from passers-by or nearby buildings. They also 
tend to be more open to offenders on foot than decks. Pedestrian 
movement in and out of decks is restricted to elevators and stairwells 
so that a thief carrying stolen items may come into contact with 
others coming and going. Thieves in lots can make a quicker getaway 
through a route of their own choosing with greater certainty that they, 
and the items they are carrying, will not be seen. 

The implications of the large difference in theft rates between lots and 
decks were far-reaching. It suggested that decks could be eliminated 
from the project because they accounted for relatively few LFAs. It 
also suggested, however, that inadequate security of parking facilities– 
in both lots and decks–in the Uptown area could indeed be 
contributing to the LFA problem. If improvements in security were to 
be sought, these inadequacies needed to be documented. Recognition 
of this fact led to the next stage of the project–a survey of the 
security of Uptown's parking facilities, with the expectation that 
lessons could be learned from examining the decks, with their low 
rates of theft, as well as the lots. 


The Security of the Uptown Decks and Lots 

The first step in designing the survey was to review past research on 
theft in parking facilities. This could have proved a major undertaking, 
but Clarke's familiarity with this research7 enabled him to contribute a 
quick summary of the findings most relevant to the Uptown situation: 

• Center city parking facilities tend to be at greater risk than those 
in other parts of a city. This may be due to the concentration of 
parking, making it easier for thieves to find attractive targets. 
• Commuter lots where cars are left for long periods of the day 
have particularly high rates of theft. 
• Parking facilities used around the clock tend to have higher rates 
of theft, if for no other reason than targets can always be found 
there. 
• The availability of cash in pay-boxes, meters, and pay-and-display 
ticket machines attracts thieves. 
• For both decks and lots, the presence of attendants greatly 
reduces risks of theft. 
• CCTV systems installed in parking facilities can be effective in 
reducing thefts in those facilities. 
• Improved lighting can reduce crime in decks and underground 
garages, and in lots with evening or night use. (The research on 
this topic is limited.) 
• Lots with pedestrian throughways experience higher rates of theft 
and thefts have been reduced when pedestrian access is reduced. 
• Improvements in perimeter security can reduce vehicle-related 
thefts. (Again, the research is limited.) 
• Lots located near stores and shops have lower rates of theft 
because of the natural surveillance provided by shoppers and 
shop staff. 
• No evaluations of electronic access systems to public parking 
facilities have been published, but these have been found effective 
in preventing theft from parking areas in housing complexes. 


The research is far from comprehensive, much of it is small-scale and 
exploratory, and most of it was conducted in the United Kingdom. 
Nevertheless, it consistently indicates that better-secured facilities (in 
terms of attendants, natural surveillance, and access controls) have 
lower rates of crime, and it provided helpful guidance on what to 
include in the survey of Uptown's parking facilities. 

7 For example: Clarke, In Press; Clarke and 
Harris, 1992; Clarke and Mayhew, 1998; Eck and 
Spelman, 1987; Frank, 2000; Geason and 
Wilson, 1990; Laycock and Austin, 1992; 
Mancini and Jain, 1987; Meredith and Paquette, 
1992; Poyner, 1991; Phillips, 1999; Sallybanks 
and Brown, 1999; Sandby–Thomas, 1992; Smith, 
1996; Tilley, 1993; and VCRAT, 1999. 



This survey was not intended, of course, to meet the rigorous
standards of an academic research study. This would have been
beyond the resources available to the project. Even a detailed
environmental survey of the kind undertaken for a CPTED project
was not practicable. The need was for a limited survey that (1) would
give a snapshot of the security in Uptown's facilities, (2) would
provide pointers to improving security, and (3) could be undertaken
quickly without occupying too much of the time of the officers and
the crime analyst.


To meet this limited need, the officers and the analyst made a rough
assessment of a small set of security variables that were under the
control of each facility's operators. Some of the variables included
were common to both decks and lots, others were specific to each
kind of facility. The full list of data collected for the 38 decks and 167
lots was as follows:


For both decks and lots:
.. Lighting (weak/moderate/strong)8
.. Day-time attendant (yes/no)
.. Night parking available (yes/no)
.. Night-time attendant (yes/no)
.. Passkey (yes/no)

For decks:
.. Security guard service (yes/no)

For lots:
.. Pay box (yes/no)
.. Fence (none/partial/full) 

8 Definitions were as follows: weak means 
more than half the facility is dark or shadowed; 
moderate means less than half the facility is 
dark or shadowed; strong means none of the 
facility is dark/shadowed. 


Data were collected during the night shift when lighting levels could
be assessed. Despite the survey's limited objectives, it represented 
amajor data-gathering exercise, requiring an unusual commitment from
officers Crawford and Foster.


Once collected, the data yielded two sets of results for lots and decks
(1) simple counts of the distribution of the variables (Table 2) and (2) 
statistical relationships between these variables and LFAs, determined 
through analyses of variance undertaken by crime analyst Kristin 
Knight (Table 3). 


Table 2 shows that, leaving aside night parking, the decks are generally 
more secure than the lots. The most important difference between 
them, however, probably accounting for most of the difference in 
risks of theft, is that decks generally have attendants (74 percent in the 
day and 60 percent at night), whereas lots generally do not (six percent 
in the day and four percent at night). 

Table 2: Security-Related Features in Decks (N=38) and lots (N=167), 
Charlotte Uptown 

Percent with: 			Decks 	Lots 
Moderate/strong lighting 	77% 	39% 
Day attendant 			74% 	 6% 
Night parking 			87% 	46% 
Night attendant 		60% 	 4% 
Pass keys 			23% 	15% 
Security guard service 		50% 	N/A 
No pay boxes 			N/A 	74% 
Fully fenced 			N/A 	12% 

Note: N/A signifies not available 

The relationships between LFAs and security features were not strong 
(Table 3), and some of the variables appeared to be inter-correlated9 
The results in Table 3 should therefore not be over-interpreted, but 
they suggest that: 

1. Security improvements are unlikely to reduce thefts in decks 
because there is no relationship between security features and the 
rate of LFAs. Most decks needing attendants may have them 
already and other security features appear to bring little added 
value. 
2. Reductions in thefts from parking lots 
would result from 
employing more attendants10 and probably also from (1) 
improving lighting and fencing and (2) making greater use of 
passkeys and less use of pay boxes. 


9 More refined statistical analyses of these 
relationships would not have been justified 
given the limitations of the data gathering 
methods. 

10 Three reasons permit this to be confidently 
asserted (1) the relationship in Table 2 between 
attendants in lots and LFAs was nearly 
significant, even though so few lots had 
attendants (2) the primary factor in the reduced 
risks in decks was the presence of attendants 
and (3) the previous research consistently 
indicates that the presence of attendants 
reduces theft. 



Table 3: Relationship Between LFAs and Security Features in Decks (N=38) and 
Lots (N=167), Charlotte Uptown 

Relationship between LFAs and: 	Decks 	Lots 
Weak lighting 			N.S. 	F=1.698, p<0.02 
No day attendant 		N.S. 	F=1.463, p<0.05 
Night parking 			N.S. 	N.S. 
No night attendant 		N.S. 	N.S. 
"Security" 			N.S. 	N/A 
Pay boxes present 		N/A 	F=2.602, p<0.01 
No passkeys 			N/A 	F=1.671, p<0.015 
Poor fencing 			N/A 	F=1.512, p<0.025 

Notes: (1) Relationships determined using analysis of variance
(2) N.S. signifies not statistically significant 
(3) N/A signifies not available 


Together, the analyses in Tables 2 and 3 confirmed the decision to 
eliminate the decks from the project (their rates of theft were already 
low and it seemed unlikely they could be reduced further) and to 
concentrate efforts on improving the security of lots. Their rates of 
theft were much higher than those of decks and there were many 
indications that if their security were to be improved, thefts could be 
reduced. Searching for the best ways to improve the security of the 
Uptown lots–the next stage of the project–occupied the project team 
for more than a year. 


The Search for Solutions 

The SARA model, which teaches police the value of thinking 
sequentially about scanning, analysis, response, and assessment, has 
been of great value in introducing police to problem-oriented policing. 
But it can also be misleading in suggesting the sequence of steps to be 
followed in any project. In fact, projects rarely follow a linear path 
from the initial scanning and analysis stages through the stages of 
response and assessment. Rather, the process is iterative so that an 
unfolding analysis can result in refocusing of the project (as happened 
more than once in the present case), and questions about possible 
responses can lead to the need for fresh analyses. The longer and 
more complicated the project, the more iterations of this kind are 
likely to occur. 

Understandably, at the beginning of the project, police continued to 
utilize the responses they had advocated in the past, and continued to 
pursue them while the analysis was proceeding. Furthermore, the pros 
and cons of some of the solutions that had earlier been proposed by 
the police, such as the closing down of liquor stores patronized by 
suspects and the relocating the homeless shelter, were periodically 
reconsidered. But as the project progressed, these discussions differed 
markedly from the earlier discussions. They reflected a greater unity 
and focus in seeking solutions–a result, it appeared, of the exchanges 
among the participants, who brought different perspectives and 
experience to the table, and the gradual blending together of the 
growing findings and the street knowledge of the officers. It was 
becoming clearer that the heightened enforcement being pursued by 
officers was having little effect on the overall problem, since the 
numbers of LFAs in Uptown had risen from 513 in 1998 to 814 in 
1999. In addition, having been directly involved in collecting the data 
on the parking facilities, the David One officers had acquired greater 
understanding of the part played by inadequate security in LFAs. 

In March of 2000, arrangements were made for two David One 
officers, Sergeant Craig "Pete" Davis and Officer Crawford, and crime 
analyst White to visit Portland, Oregon. Steve Ward, the Assistant 
District Attorney participating in the project, had heard about 
Portland's success in preventing thefts from autos in the city's Lloyd 
district, a commercial and office district immediately adjacent to its 
downtown core. In an area about as large as and similar to Charlotte's 
Uptown, the number of LFAs had been reduced from about 900 per 
year in the early 90s to 200 to 300 per year after the program. Their 
visit served to support the exploration of new strategies by providing 
some specific examples of preventive measures that had been used by 
another police agency for reducing LFAs in addition to the traditional 
dependence on law enforcement. 



The team sent to Portland reported back that the successful program 
in Portland had been implemented through a partnership of local 
businesses, private security companies, the police, and the DA's office. 
The program was comprised of 

1. A streamlined legal process that resulted in more convictions and 
more severe penalties, with repeat offenders being more often 
sentenced to prison 

2. The installation of electronic single-arm gates at parking lot 
entrances (to deter thieves cruising in cars) 

3. The closing down of camps near the parking facilities that had 
been illegally established by the homeless 

4. The reorganization of security services to provide a bike patrol 
covering all the lots 

5. The provision of a direct radio link between the bike patrol and 
district police 


All these measures were thought to have played a part in the 
reductions achieved, but the most effective was generally believed to 
be the bike patrol. 

By this time the project team had acquired a sound understanding of 
the LFA problem and a broad knowledge of responses that had 
worked elsewhere or might work in Charlotte, given the specific nature 
of the problem in Uptown. They were now in a position to set about 
developing an intervention plan that would have an immediate impact 
on the problem as well as a sustained longer-term effect. Given the 
complex nature of the problem, it was clear that the plan, as in 
Portland, would involve several elements requiring partnerships with 
other agencies, including Uptown business interests, city departments, 
the parking lot operators, and the D.A.'s office. If it were to have any 
chance of being implemented, it could not be too costly and should 
anticipate likely bureaucratic and legal difficulties. 

It was also accepted that some recommendations might be 
implemented quite quickly, but others would require a longer time 
scale, perhaps of two or three years. Accordingly, the plan should take 
account of anticipated changes in the city that might have an impact 
on LFAs. Several of these changes, related to the anticipated 
continuation in the expansion of the Uptown economy, seemed likely 
to make the problem worse. These included an expected growth in 
nighttime activity resulting from the construction of more office 
space, the building of more housing and hotels, and the opening of 
more clubs and restaurants. The proposal for construction of a new 
basketball arena and baseball stadium was defeated in a recent 
referendum, but these projects may be renewed. 



Other anticipated changes seemed likely to reduce LFAs. One of these 
was that more lots would gradually be converted to freestanding decks 
or to office buildings with garages, with the accompanying security 
those facilities generally experience. Shorter term, a new trolley service 
was scheduled to be introduced (in 2002) on the tracks currently used 
by the transients as a conduit through the city. This new service, with 
the activity and natural surveillance it would, as a byproduct, generate, 
would make it harder for the transients, were they inclined to engage 
in theft, to gain access to the parking lots from the trolley line. This 
relieved the project team of the need to pursue a response which 
would assuredly be controversial–the possibility of trying to re-site the 
shelter or soup kitchen to keep transients, who were thought by the 
police to be among those responsible for the LFAs, away from the 
Uptown lots. 

A second imminent change was that an Uptown CCTV system was to 
become fully operational in the first half of 2001. This would be 
funded from the CMPD's block grant from the federal government 
and by contributions from business members of the Center City 
Crime Prevention Council, who made locations for the cameras atop 
some of the tallest buildings in Uptown available at no cost in 
exchange for a linkage to their security desks. The system would 
comprise nine cameras that would be monitored by the police from 
one central location for up to 10 to 12 hours per day. The precise 
proportion of the surface lots that is subject to surveillance by the 
cameras was not established.11 The police involved in the project 
roughly estimate that a majority of the lots are under observation. But 
even if coverage were limited, the cameras might still provide a 
convincing deterrent to casual thieves. 

11 A plan to undertake such a study, together 
with a study of the amount of surveillance given 
lots from surrounding buildings, was abandoned 
because of lack of time and resources. 

In deciding upon the final group of measures to include in the 
intervention plan, the project team avoided blanket requirements for 
every lot to be illuminated to a particular standard, or to have full-time 
attendants. While such requirements could effectively reduce LFAs, 
they might not bring uniform crime prevention benefits for all lots 
and could also bankrupt the operators of the smaller ones. A more 
selective, cost-effective approach was sought. With these criteria in 
mind, the intervention plan that was developed included five distinct 
recommendations: 

1. The police and the D.A.'s office would continue to develop 
aggressive policies of arresting offenders, seeking convictions, and 
seeking severe sentences for repeat offenders. This had always 
been a central aim of the police response in David One and was 
also an element of the successful Portland program. 



2. Parking lot operators would be asked to post the addresses of 
their lots at the entrance(s) to each lot. This would assist victims 
in reporting thefts, help police in responding to calls for 
assistance, and assist future analysis of LFAs by allowing these to 
be assigned to the specific lot in which the LFA occurred. 

3. Changes would be sought in the city's zoning ordinance that 
currently, requires, for aesthetic purposes, that all new lots be 
surrounded by screening (which in practice is usually a fence) that 
is no less than four feet in height and can have no more than 25 
percent of its surface left open.12 These fences, most often solid, 
have reduced surveillance of lots by passing motorists, 
pedestrians, and police officers on patrol. Furthermore, lots 
established before the ordinance came into effect in 1993 (and its 
amendment in 1995), which constitute a majority of all lots, were 
not required to have screening. The proposed new ordinance 
would require see-through fences to be erected for all new parking 
lots and, within a period of two or three years, for all existing lots. 

12 Section 12.303 of the City Code.

4. With the cooperation and agreement of lot operators, the police 
would seek to implement a rating scheme that would result in 
every lot being graded for its security on a number of variables. 
Grades would be determined by either the police or the building 
inspector and would be posted at the lot entrances, in the same 
way health inspection results are posted for Charlotte's restaurants. 
This rating scheme would be modeled on the Secure Parking 
scheme as originally proposed in the United Kingdom. Experience 
there has shown that the implementation of the proposal, with 
adjustments over time, provides a strong incentive for parking 
facility operators to improve security (VCRAT, 1999). 

5. Funds would be sought for a security bike patrol for the uptown 
lots similar to the successful patrol introduced in Portland. Such a 
patrol had also been found effective in rail commuter lots in 
Vancouver (Barclay et al., 1996). The patrol would be trained in 
what to look for, how to focus patrols for greatest effect, how to 
deal with suspicious persons, and when and how to call the police 
(their radios would be compatible with police radios). The patrols 
would give the customers and employees of Uptown businesses 
the same type of security that private patrols give to customers 
and employees at large shopping malls. 



Selling the Solutions 

The recommendations for more aggressive legal pursuit of offenders 
and the posting of lot addresses were relatively uncontroversial. But it 
was thought that the others, particularly the changes in fencing, the 
grading system, and the bicycle patrol, were likely to encounter 
resistance from lot operators because of the potential costs and, in the 
case of the grading system, the commonly expressed concern about 
increased government regulation. It was decided that a presentation 
should be put together which would be used in selling the intervention 
plan to lot operators and others. Crime analyst White undertook 
responsibility for developing the visuals on which the presentation 
would be based. These were refined over the ensuing months and 
eventually consisted of 50 PowerPoint slides that took nearly one hour 
to present. They covered the background to the project, the reasons 
for focusing on LFAs in parking facilities, the visit to Portland, the 
decision to study lot security, anticipated developments in Uptown, 
rejected solutions, the thinking behind the proposed intervention plan, 
and the next steps toward its implementation. 

Considerable care was taken to report the findings of the analyses in a 
readily understandable form. The crime analyst spent many hours 
developing maps that clearly showed differential risks of LFAs 
throughout Uptown. In reporting the statistical relationships between 
LFAs and security features, he avoided correlation coefficients (which 
many people have difficulty interpreting) and, instead, made use of 
maps showing lots that were close to one another, but which differed 
in their levels of security and in their rates of theft. For example, he 
found adjacent lots that varied in lighting quality and which had quite 
different rates of LFA. Figure 4 (see Appendix) shows the maps he 
made to illustrate the relationships between LFAs and lighting quality, 
fencing and the presence/absence of attendants.13 

13 While not entirely defensible on scientific 
grounds, it was an effective way of showing 
these relationships. 

Also, in preparation for the meetings, two members of the David One 
team, Officers Crawford and Robert Vandergrift, undertook CPTED 
surveys of three pairs of adjacent lots with widely differing LFA rates 
to identify the differences in the security of the lots. The striking 
results of these surveys were included in the presentation, together 
with photographs of the lots showing the differences between the two 
lots in each of the three pairs. 

This presentation formed the basis of a report made on the project to 
Chief Darrel Stephens (who had recently succeeded Chief Nowicki) 
and senior officers of the CMPD. This meeting was helpful in refining 
the presentation, particularly concerning the likely benefits of the new 
CCTV system, but it also raised important issues regarding the 


 

proposed bike patrol and security grading system. Concerning the bike 
patrol, it was pointed out that the additional officers funded by 
Uptown businesses were already undertaking bike patrols and these 
patrols had gradually been extended into parking facilities, but these 
did not seem to be controlling the problem. To recommend that bike 
patrols be undertaken by additional security officers would likely 
provoke the question as to why these should be expected to succeed 
when police bike patrols had not. As for the grading system, the 
project team was instructed to undertake a careful study of the police 
capacity for undertaking these surveys, which would involve technical 
difficulties as well as requiring a considerable resource investment. 

These comments resulted in a more cautious presentation, in which 
recommendations were phrased more tentatively with fuller discussion 
of the difficulties attached to each. Over the succeeding months, this 
presentation was made by Captain Sennett, assisted by the project 
team, to Uptown lot operators, to Charlotte's Center City Crime 
Prevention Council, and at a meeting with the city's planning 
department. The sequence of these meetings was carefully arranged so 
that agreement in principle to the intervention plan was obtained first 
from the parking lot operators, whose cooperation was vital, and that 
agreement was conveyed to those with whom the team subsequently 
met. 

The project team considered these meetings to be highly successful, 
which we can confirm having been present at the one with the Center 
City Crime Prevention Council. At least 80 people attended this 
meeting, representing a wide range of business and city interests, 
including the three largest parking lot operators (controlling among 
them 85 percent of the parking spaces in Uptown). Those present 
were clearly impressed by the professional nature of the presentation 
and by the wealth of detailed information presented about the 
problem. No criticism was voiced of the inability of police to control 
LFAs and very little dissent was expressed concerning the 
recommendations. To the contrary, expressions of support and offers 
of help in implementing the plan were made from the floor. 

During the period that these meetings were being held, the project 
team was undertaking work needed to advance the recommendations. 
Little new needed to be done about aggressive legal pursuit of 
offenders since this was already agreed policy, though David One 
officers now regularly request that a territorial exclusion order be part 
of the sentence imposed on an offender convicted on an LFA charge 
who is returned to the community under probation. The 
recommendation concerning the posting of lot addresses was quickly 
accepted and lot operators are already beginning to comply. Progress 
was initially slow in changing the fence ordinance, despite the 



endorsement of lot operators, and a letter of support written by Tim 
Crowe, a nationally-known expert in CPTED.14 After the presentation 
at the planning department, however, the Director of Planning agreed 
to lend his support, considered vital, to the new ordinance. He also 
suggested that it should be extended to include requirements about 
adequate lighting and, at the time of writing, the revisions to the 
ordinance are being drafted and subjected to the approval of 
interested parties before being formally submitted to the City Council 
for its approval. 

A security grading system that the police could administer is also being 
developed, again with the assistance of Tim Crowe. Lot operators 
made surprisingly few objections to the scheme. They asked only that 
(1) they be given a preliminary grade (A, B, C, or D) for each lot 
which would only be made final after they have the opportunity to 
make necessary improvements and (2) that they be given the 
opportunity to be re-graded whenever they make subsequent 
improvements. They also expressed reservations about the 
requirement for posting of grades at lot entrances. At the time of 
writing, these matters are still under discussion. 

The lack of opposition to the proposed bike patrol might have been 
due to the absence of details about costs and who would bear these. 
Captain Sennett obtained proposals from two interested security 
companies for a patrol that would call for two persons to be on duty 
at any one time, augmented by another two at peak hours–requiring 
that, in all, eight persons be trained and available to fill this level of 
staffing. The costs of the proposals were similar and were comparable 
to the cost of hiring fully equipped CMPD officers on an off-duty 
basis. At the time of writing, Sennett is working with the Center City 
Crime Prevention Council to find ways of paying for the patrols. An 
alternative being considered is to combine the existing bike patrols 
provided by Bank of America for lots used by its employees15 with 
similar patrols to be provided by another major Uptown bank. 

Meanwhile, David One (with funds from the local Alcohol Beverage 
Control authority) has very recently implemented a new 
communications system that allows the Bank of America security 
officers and those of some other Uptown businesses to have direct 
radio contact with the on-duty David One field supervisor and the 
David One district office. This will enable greater advantage to be 
taken of the CCTV surveillance of lots undertaken by several of these 
security companies. It is also a further step in forging a crime 
prevention partnership between the police and the private security 
firms operating in Uptown. 

14 Tim Crowe is author of the principal text on 
CPTED (Crowe, 1991) and had been engaged by 
Chief Stephens to provide CPTED training for the 
department. 

15 Bank of America has maintained foot 
patrols since 1996 and bike patrols since 1998. 
They operate 6:00 a.m. to midnight, with two to 
six security officers on duty at any one time. 


An Unexpected Decline in Thefts 

The work of selling and implementing the intervention plan spanned 
the end of 2000 and the beginning of 2001. It was during this period 
that it became evident that LFAs in Uptown were declining. In fact, 
the decline in 2000 was substantial (38 percent), the number having 
dropped from 814 in 1999 to 506 in 2000. Most of the decline took 
place in the lots and decks, not on the streets or on private property 
(see Table 4).16 

16 There is a significant difference in LFA locations across the years 
(observed Chi-Square 17.49; degrees of freedom 4; Critical Value 13.27; Alpha=0.01). 


Table 4: Location of LFAs in Charlotte Uptown 1998-2000 

		1998 	1999 	2000 	Reduction in 2000* 
Decks 		58 	93 	52 	44% 
Lots 		292 	510 	269 	47% 
Elsewhere 	163 	211 	185 	12% 
Totals 		513 	814 	506 	38% 

*Compared with 1999 

Clearly, the decline was not due to the intervention plan, which had 
not yet been implemented. Nor was it due to the CCTV system, 
whose first cameras only became operational in November 2000. 
Finally, it does not appear to have been due to any greater success in 
arresting offenders in 2000. In that year, 11 LFA arrests were made in 
Uptown compared with 25 in 1999 and 16 in 1998. 

The most likely explanation for the fall is that the lots began to attract 
more attention from police and security patrols in 2000, partly as the 
result of the project team's activities. This was argued in a report on 
the project made by David One to the COPS Office in April 2001: 

"Although not every David One district officer was directly involved 
in the project, many officers were aware of the district's heightened 
interest in larceny from auto in the district. As a result, officers 
became more aware of suspicious activity in surface parking lots as 
they traveled throughout the district. Because of this heightened 
awareness, officers stopped to talk to suspicious people who were in 
the parking lots. The prevention aspect of this interaction with 
suspicious persons should be credited to the project and to the level 
of attention the District Captain conveyed to his officers." 

Another reason why the lots attracted more attention from police and 
security patrols was that on July 13, 1999, a woman employed by a law 
firm in Uptown was fatally stabbed while approaching her vehicle in 
one of the surface lots. This homicide resulted in a heightened 


sensitivity to the safety of the lots. It also resulted in an expansion of 
the coverage of the private bike patrols into the lots used by 
employees of the buildings maintaining those patrols. 

The unprogrammed increase in the surveillance given to the surface 
lots by patrols does not account for the decline of LFAs in the decks, 
though this could have been the result of offenders being more 
generally scared away from Uptown. If so, it would be another example 
of the diffusion of the benefits of crime prevention activity beyond 
the targets of intervention (Clarke and Weisburd, 1994).

The number of LFAs continued to drop precipitately in the first 
quarter of 2001, as shown in Table 5.17 At this stage, a contributory 
factor could have been the CCTV system that gradually came into  
operation in the first quarter of 2001, accompanied by news stories  
about the system carried by the local papers and TV stations. While 
few arrests could be attributed to the CCTV cameras,18 they helped 
alert officers to suspicious persons in the surface lots and they might 
have raised the fear of apprehension among potential thieves. 

Table 5: LFAs in Charlotte uptown 
January-March, 1998-2001 

1998 	87 
1999 	201 
2000 	144 
2001 	68 
% Reduction in 2001* 	53% 

*Compared with 2000 

Furthermore, the decline of LFAs in Uptown did not result in 
displacement of LFAs to the rest of the David One district, where 
LFAs also declined from a total of 499 in 1999 to a total of 441 in 
2000 (see Table 6).19 This decline of 11.6 percent was somewhat 
greater than the 8.5 percent decline in LFAs reported for the CMPD 
as a whole (CMPD, 2000). If anything, this pattern suggests, once 
again, that there may have been some diffusion of benefits to the rest 
of David One from the Uptown reductions in surface lot LFAs. 

17 There is significant drop in LFAs across the years as shown by a 
one-way Goodness of Fit test (observed Chi- Square 86.64; degrees 
of freedom 3; Critical Value 16.26; Alpha 0.001). 

18 In the four months until the end of April, only four LFA arrests 
were made in Uptown and it is not known in how many of these the 
cameras played a role.

19 There is a significant difference in LFAs in David One across the years (observed Chi-
Square 33.037; degrees of freedom 2; Critical Value 9.21; Alpha 0.01).



Table 6: LFAs in David One District 
1998-2000 

			1998 	1999 	2000 
Uptown 			513 	814 	506 
Rest of David One 	498 	499 	441 
Totals 			1011 	1313 	947 

This welcome decline has not removed the need for the measures in 
the intervention plan, which could help to turn what might otherwise 
be a short-term improvement into a permanent reduction in LFAs. 
Indeed, Captain Sennett has resolved to press on with the plan, 
including the more difficult elements such as the grading system and 
the bike patrol. Full implementation of the latter might wait however 
to see how the situation develops over the next year or two. 


Lessons of the Case Study 

As advocates of problem-oriented policing, we constantly ask 
ourselves why a concept that is so straightforward, and even 
commonsensical, is so difficult to put into practice. As explained in 
the Introduction, this question is also repeatedly raised in the 
literature. Direct involvement in this case study in the CMPD, where 
the conditions for advancing problem-oriented policing are particularly 
favorable (e.g., overall reputation as a modern police agency, 
commitment to the concept, highly developed crime analysis 
operation, superior data retrieval, and mapping capacities), produced a 
number of insights that inform the larger, broader efforts to 
implement problem-oriented policing elsewhere. 

The most frequent explanation for the absence of a fuller 
implementation of problem-oriented policing is framed in terms of 
the difficulty that police experience in making the switch from their 
usual way of doing business. This conventional method of policing 
involves a quick in and out response to single incidents, commonly 
referred to as a fire brigade response, leaving officers free and ready to 
respond to the next, potentially more serious incident demanding their 
attention. The problem-oriented approach requires police to restrain 
the impulse to use traditional responses of questionable value and, 
instead, to undertake a slow, methodical analysis of classes of similar 
incidents so as to identify and implement longer-term, preventive 
measures. This process might take weeks or months, rather than the 
minutes or hours usually required for their normal method of 
responding. It could be characterized as demanding patience at the 
beginning of a project and persistence at the end. It requires not just a 
fundamental change of attitudes by individual officers and their 
supervisors, but also a radical change in police organization and 
management.20 In most cases, even when they have embarked 
enthusiastically on a project, police find these changes difficult to 
sustain in the environment in which they operate. This is why analyses 
are so often superficial, responses are uncreative, and assessments are 
perfunctory or absent. 

20 These changes are spelled out in detail in 
Chapter 9 (pp. 148-175) in Goldstein (1990). 

While some of these difficulties were experienced even in the present 
project,21 it is still underway after more than two years of continuous 
work, and despite the unexpected decline in LFAs, it remains focused 
on achieving the longer-term changes designed to prevent a 
resurgence of the problem. This degree of persistence is well beyond 
that which is normally reported. The absence of persistence, 
sometimes labeled lack of commitment, is frequently cited as the 
primary reason in explaining the failure to implement problem-
oriented policing. But, in our opinion, the much stronger reason for 
the lack of progress, reinforced by our experience working on this 

 



project (and the other CMPD projects in which we were involved) is 
the sheer difficulty of undertaking problem-oriented policing. It is 
both administratively and technically difficult, and unless these 
difficulties are addressed, there is little prospect of the problem-
oriented approach becoming a standard policing method. 

21 At its beginning, some of the officers were 
anxious to get going and expressed some 
frustration when initial efforts were made to 
extend the analysis and the search for proposed 
solutions. Without the periodic deadlines 
imposed by the regularly scheduled meetings 
between ourselves and team members, it is 
likethat progress on the project would have 
been even more delayed by the press of other 
business. Likewise, our involvement as 
consultants helped to ensure that problem 
analysis and the search for alternative 
responses were pursued further than might 
otherwise have been the case. 

As problem-oriented policing was initially conceptualized, it was never 
contemplated that the primary burden for implementing the concept 
would rest on line officers. Around the country, police officers have 
been introduced to problem-oriented policing through a variety of 
short-term training programs. Line officers have repeatedly 
demonstrated that they are among the most committed, from among 
the ranks and staffs of police agencies, to grasp the concept, conduct 
studies, and implement new responses. But the most skilled and 
committed among them will acknowledge that, when it comes to an 
ambitious, in-depth study of the type undertaken in this case study, 
there is little in their police training–relating to the analysis of data 
and, more generally, in research skills–that equips them to carry out 
such a study on their own. The specialized training and skills needed 
are more likely found in a crime analysis unit, and the primary 
responsibility for analysis, which is at the heart of problem-oriented 
policing, must be placed there.22 Heavy dependence, however, must 
continue to be placed on officers–for their important role in 
contributing their knowledge of problems as they exist on the street, 
in aiding in the collection and interpretation of data, in helping to 
weigh the merits of alternative responses, and most importantly, in 
working on the implementation of new strategies. An appropriate 
blend of talents must be achieved. Just as it is unlikely that police 
officers could, by themselves, carry out a problem-oriented policing 
project, so a crime analysis unit cannot implement problem-oriented 
policing without the involvement of both line police officers and 
police leadership. 

22 This observation may conjure up an 
impossible burden for police administrators who 
are so often strapped for resources. But it is not 
contemplated that any one police agency would 
invest, at any one time, in an in-depth analysis 
of a large number of problems. Rather, precisely 
because resources and staff are so scarce, a 
single police agency such as the CMPD might 
not be able to examine more than two or three 
such problems in a year. Ideally, the 
development of a departmentwide commitment 
to creating an atmosphere in which all members 
of the department think in terms of identifying 
and addressing problems at all levels of the 
agency–a process that should not be 
abandoned–will not only increase police 
effectiveness regarding more discrete beat-level 
problems, it would also serve as the means for 
nominating problems that are potentially good 
candidates for more in-depth inquiry. Moreover, 
if police departments that can afford the 
minimum staff commitment conducted even one 
such study and shared the results with others, 
the police field in general would enormously 
benefit. 


These assumptions were confirmed in the David One project. The 
officers had been introduced to problem-oriented policing, but had no 
training or prior experience in researching in the required depth a 
problem of this magnitude and complexity. With guidance, they 
responded with growing enthusiasm to the need for collecting 
information, acquiring information from elsewhere, and conducting 
surveys. The captain of David One took the lead in presenting the 
results of the study and negotiating with potential partners in 
implementing the new strategies. The crime analysts, with no prior 
experience working on a problem-oriented policing project that was 
this ambitious and that probed a problem in such depth, demonstrated 
that, with guidance, they could use the tools and data readily available 
to them to take their usual work to a new and higher level. 



While a problem-oriented policing project will always compete with 
the many urgent matters that arise in a police agency, the time 
consumed in carrying out a project can be greatly reduced. Our 
limited involvement was not enough to pick up on tasks required 
doing or to coordinate matters on site in ways that might have sped up 
the project. The numerous tasks fell to an otherwise busy team of 
people. Among the lessons of this project was the realization that 
continued involvement in ambitious problem-oriented policing 
projects, to be accomplished in a timely manner, requires more on-site, 
in-house coordination of the various component efforts. The 
commitment and enthusiasm of the officers and crime analysts 
involved in this project could have been even more effectively 
channeled with the sustained attention of a full-time coordinator who 
could have stayed on top of things, and who could thereby have 
brought the project to a speedier conclusion.23 A coordinator could, 
for example, most likely have sped up the David One project by: 

1. Expediting the collection of data on parking facilities and parking 
spaces 
2. Assisting the crime analyst in identifying questions for study 
3. Searching for other relevant experience in dealing with LFAs 
4. Relieving the police of acquiring certain information (e.g., about 
the costs of fencing and private patrols) and making some 
contacts (e.g., such as initial explorations with lot operators, the 
planning department and other partners) 
5. Assuring follow-up on the many points and questions raised at 
periodic meetings 
6. Undertaking a host of other essential tasks that fell to individual 
team members to perform along with their regular work 


The need for such coordination, in any project meeting the definition 
of problem-oriented policing, is a fact that must be faced by 
departments seriously committed to the approach. 

Without substantial and continuous involvement in research, it is not 
easy for officers engaged in problem-oriented policing to conduct a 
literature review to identify relevant studies and relevant prior 
experience in dealing with similar problems. And even if they are 
experienced in conducting a search, they confront other problems. 
The nature of the literature is such that they may learn about titles, 
but have difficulty finding copies of actual documents. Specialized 
libraries that are most likely to have the fullest collection of such 
materials are few in number and not conveniently located to all 
agencies. And without familiarity with this body of literature, it is 
often difficult to judge the quality of the research reported so as to 
decide what is worth focusing on. 

23 This same conclusion has been reached by 
others in reviewing efforts to implement 
situational crime prevention (e.g. Gladstone, 
1980; Laycock and Tilley, 1995). 



The Internet has helped some law enforcement agencies to deal with 
these difficulties, but for this type of search, computers also have their 
limitations. While we were able to meet the need for literature review 
in the present project, ways of helping police everywhere to profit 
from the available literature must be found. The forthcoming 
publication of the Problem-Oriented Guides for Police, a project 
developed under a grant from the COPS Office, constitutes a 
substantial step in this direction. These guides, of which 20 have been 
prepared, seek to present in a synthesized, readable form the lessons 
that have emerged from the experience of law enforcement and others 
in dealing with specific problems, such as those addressed in the LFA 
project. Another approach would be to expand the responsibilities of 
crime analysts, and give them appropriate training, so that they could 
be expected on request to undertake and report the results of focused 
literature reviews. 

Lastly, the project illustrated the difficulties faced by members of a 
police agency–both officers and crime analysts–in obtaining the 
considerable amounts of information needed to guide each stage of a 
project. They must decide what information is needed, identify 
sources and persuade those holding the information to release it, and 
then analyze and interpret it. The following brief list of the 
information collected for the present (incomplete) project will 
illustrate the scope of the work that will often be required. 

1. At the scanning stage, data about vehicle-related thefts in the David 
One area were examined to determine the relative proportions of 
auto thefts and LFAs occurring in the parking facilities and 
elsewhere. The hot spot mapping undertaken by the crime analyst 
facilitated this examination. 
2. At the analysis stage, maps of the Uptown area showing individual 
lots and decks had to be updated from planning records, from 
aerial photographs, and from physical checks made of facilities. 
The number of spaces in each parking facility had to be recorded 
and in many cases counted. Security surveys had to be undertaken 
of the 206 separate parking facilities identified. These data had to 
be subjected to correlational analysis and significance testing. 
Rates of LFAs had to be calculated for each block in the city. 
Computer maps of Uptown showing the distribution of rates of 
LFAs had to be constructed. 
3. At the response stage (still incomplete at the time of writing), cost 
data were obtained for employing full-time attendants, installing 
various kinds of fencing, and the projected bike patrol. 
Information was obtained about the projected new trolley line. 
CPTED surveys were undertaken of three parking facilities. 
Studies were designed (but not carried out because of lack of 
resources) to measure the surveillance given to each lot by the 




CCTV system and by the windows of overlooking buildings. 
Computer maps were made showing LFAs in adjacent lots with 
and without attendants and with different levels of lighting and 
fencing. A detailed grading scheme for lot security was developed 
with the assistance of Tim Crowe. LFA data for Uptown and the 
remainder of David One had to be analyzed for 2000/2001 to 
document the unexpected decline in thefts and to see whether 
displacement had occurred. 

The need for these data draws attention to the vital roles of both 
crime analysts and line officers in problem-oriented policing. But given 
the expectation that has built up about officer involvement, the 
examples emphasize the importance of giving more attention to the 
role of the crime analysts. As strong and engaged as the analysts were 
in this case study, and we know of none better, they had, prior to this 
project, no occasion to get deeply immersed in problem-oriented 
policing or situational prevention. If this were true in the CMPD, with 
its unusual investment in crime analysis, it must hold with even greater 
force elsewhere. This means that, if problem-oriented policing is to be 
properly implemented, ways will have to be found to provide a greater 
pool of those who can furnish the necessary analytic support. The 
following questions will need to be addressed. 

1. How can enough people with the appropriate blend of interests, 
basic research skills, and the appropriate computer skills be 
recruited for these positions, when–especially with regard to the 
computer skills–they are in such great demand in the more highly-
paid, private sector? 
2. How can crime analysts be given a form of specialized training 
designed to expand their capacities that would draw heavily on 
what is known about problem-oriented policing, situational crime 
prevention, and the relatively new specialty of environmental 
criminology? 


The first question lies outside our competence and might require a 
national plan to be formulated. As for the second, narrower question, 
we should note that the National Institute of Justice and the COPS 
Office have both played a useful role in drawing police attention to 
the capacities of the new mapping software and in providing training 
to analysts and officers in crime mapping.24 However, neither agency 
has invested in training designed to provide crime analysts with the 
skills and knowledge needed if they are to provide support for 
problem-oriented policing. Attempting to do so would expose a 
shortage of expertise and a lack of training materials, which is a 
situation that needs to be urgently addressed. 

24 For example, the Crime Mapping Research 
Center at the NIJ 
(www.ojp.usdoj.government/crmc/) holds an 
annual mapping conference and has published 
an important text on crime mapping (Harries, 
1999), while COPS has sponsored training in 
mapping by the Regional Community Policing 
Institutes (RCPIs). The Carolina Institute for 
Community Policing, in which the CMPD is 
heavily involved, has geographic information 
systems (GIS) as its major focus area. The 
Institute has provided much training over the 
course of its existence. 


Summary and Conclusions 

The purpose of the project was threefold: 

1. To illustrate, within the context of the CMPD, what is involved in 
a full implementation of problem-oriented policing by taking on a 
comprehensive, in-depth effort to address a specific piece of 
police business. 
2. Having focused on the problem of theft from vehicles, to develop 
specific strategies designed to increase the effectiveness of the 
CMPD in dealing with that problem. 
3. More broadly, to gain new insights into the complexities of 
introducing problem-oriented policing into a police agency. 


It is difficult to measure the degree to which the first objective has 
been achieved. The project has touched many members of the CMPD. 
Descriptions of it have been incorporated in some of the agency's 
training. Presentations have been made to management. And perhaps 
most importantly, those in a position to encourage new ways of 
thinking about policing now have, by virtue of their familiarity with 
this and a companion project on theft from constructions sites, a 
better understanding of what problem-oriented policing entails. 

With regard to the problem of theft from vehicles, the project has 
produced several specific strategies grounded in exhaustive study that 
are targeted at reducing such thefts in the Uptown area of Charlotte. 
The most promising proposals have yet to be implemented, but work 
is proceeding on putting them in place. In the interim, an unexpected 
decline has occurred in LFAs. No hard evidence is available to explain 
the decline, but the police involved feel that the project may have 
indirectly contributed to the decline through the attention focused on 
the problem. Police regularly assigned to the area appear to have 
intensified surveillance of the surface parking lots in Uptown. It is 
anticipated that full implementation of the newly devised strategies 
will contribute to a long-term, permanent reduction in LFAs. 

A major benefit of the project (apart from an anticipated long term 
reduction in LFAs) has been the deeper understanding acquired of the 
administrative and technical difficulties encountered by police in 
implementing problem-oriented policing. The project abundantly 
illustrated just how complex it is to examine a large problem that, 
though commonly confronted by the police throughout this nation 
and abroad, has rarely been put under such an intensive microscope. 
And this is just one of the many problems routinely handled by police 
which have not been similarly examined. The project confirmed that, 
in its most ambitious form, problem-oriented policing is indeed, 

 

contrary to the frequent claim, a complex process that requires much 
patience at its beginning, and much persistence in blasting through to 
the end. It is an iterative process, not lockstep, in which the gradual 
acquisition of data and information informs the project, leading to 
more questions, to redefinition, and even to changes in focus as it 
moves along. And the cycle repeats itself several times as more 
knowledge is acquired and possible strategies are explored and ideally 
tested. It cannot simply be introduced alongside other activities 
without an allocation of sufficient staff time, without special training, 
and without other adjustments in the management and organization of 
a police agency. It requires that police have improved access to 
information about prior experience dealing with the problems being 
addressed. And if it is to be adopted more widely by police agencies, it 
requires a substantially expanded and better-trained cadre of crime 
analysts to support the initiatives and efforts of career police. Ways of 
meeting these needs must be found if problem-oriented policing is to 
achieve its prime objective, which is to enable police agencies to 
engage in-house in the kind of analysis that helps them to improve 
their effectiveness in dealing with the problems that the public expects 
them to handle, and to share the results of their efforts with police 
elsewhere. 


Appendix 


Map of Uptown Study Area
Figure 1 



Map of Uptown Study Area and Original Hotspotting
Figure 2 

Map of Uptown Study Area and Analysis
Figure 3 



Maps Highlighting Fencing, Lighting, Attendant and Night Parking
Figure 4 


References 

Barclay, P., Buckley, J., Brantingham, P.J., Brantingham, P.L. and 
Whinn-Yates, T. (1996). "Preventing Auto Theft in Suburban 
Vancouver Commuter Lots: Effects of a Bike Patrol." In R.V. 
Clarke (ed.). Preventing Mass Transit Crime. Crime Prevention 
Studies, Vol. 6. Monsey, NY: Criminal Justice Press. 

Barthe, E. (2000). Unpublished Ph.D. Prospectus, School of Criminal 
Justice, Rutgers, The State University of New Jersey. 

Burrows, J. and K. Heal (1980). "Police Car Security Campaigns". In 
R.V.G. Clarke and P. Mayhew (eds.). Designing Out Crime. 
London: H.M.S.O. 

Charlotte-Mecklenburg Police Department (2000). Annual Summary. 
Unpublished Report. 

Clarke, R.V. (1997). Problem-Oriented Policing and the Potential 
Contribution of Criminology. Unpublished Report to the 
National Institute of Justice (Grant # 95-IJ-CX-0021). 

Clarke, R.V. (1998). "Defining Police Strategies: Problem Solving, 
Problem-Oriented Policing and Community-Oriented Policing." In 
T. O'Connor Shelley and A.C. Grant (eds.), Problem-Oriented 
Policing: Crime-Specific Problems and Critical Issues and Making 
POP Work, Volume 1. Washington, D.C.: Police Executive 
Research Forum. 

Clarke, R.V. (In Press) "Theft of and from Cars in Parking Facilities". 
Problem-Oriented Guides For Police, No 10. Office of 
Community Oriented Policing Services. Washington, DC: US 
Department of Justice. 

Clarke, R.V. and P. Harris, (1992). "Auto Theft and Its Prevention." In 
M. Tonry (ed.) Crime and Justice: A Review of Research, Vol.16. 
Chicago, IL: University of Chicago Press. p.1–54. 

Clarke, R. V. and P. Mayhew (1998). "Preventing Crime in Parking 
Lots: What We Know and Need to Know." In M. Felson and R.B. 
Peiser (eds.) Reducing Crime Through Real Estate Development 
and Planning. Washington, DC: Urban Land Institute. 

Clarke, R.V., and D. Weisburd. (1994). "Diffusion of Crime Control 
Benefits: Observations on the Reverse of Displacement." In 
Crime Prevention Studies, vol. 2, edited by Ronald V. Clarke. 
Monsey, N.Y.: Criminal Justice Press. 



Crowe, T. (1991). Crime Prevention through Environmental Design: 
Applications of Architectural Design and Space Management 
Concepts. Boston: Butterworth-Heinemann. 

Eck, J. E. and W. Spelman (1987). Problem-Solving. Problem-Oriented 
Policing in Newport News. Police Executive Research 
Forum/National Institute of Justice. Washington, DC: Police 
Executive Research Forum. 

Frank, A. (2000). "Police Shut Holes at Newark Airport." The Star-
Ledger, March 22. 

Geason, S. and P. Wilson (1990). Preventing Car Theft and Crime in 
Car Parks. Canberra, AUS: Australian Institute of Criminology. 

Gladstone, F. (1980) Co-ordinating Crime Prevention Efforts. Home 
Office Research Study No.47. London: HM Stationery Office. 

Goldstein, H. (1979). "Improving Policing: A Problem-Oriented 
Approach". Crime and Delinquency (April): 234–58. 

Goldstein, H. (1990). Problem-Oriented Policing. New York: McGraw 
Hill. 

Goldstein, H. (1994a) "Examining the Current Status of Problem-
Oriented Policing and Thinking Through an Agenda for Research 
and Technical Support for the Concept." Unpublished 
memorandum addressed to Craig Uchida, Office of Community 
Oriented Policing Services in the United States Department of 
Justice, July 11, 1994. 

Goldstein, H. (1994b) "Suggestions for Inclusion in the National 
Institute of Justice Research Agenda Relating to the Police." 
Unpublished memorandum addressed to Jeremy Travis, Director 
of the National Institute of Justice, September 29, 1994. 

Goldstein, H. (1996a) Problem-Oriented Policing: The Rationale, the 
Concept, and Reflections on Its Implementation. London: Police 
Research Group, Home Office. 

Goldstein, H. (1996b) "Problem-Oriented Policing." An address to the 
summer conference of the Association of Chief Police Officers, 
Manchester, England, July 3, 1996. 

Harries, K. (1999) Mapping Crime: Principle and Practice. National 
Institute of Justice. Washington, DC: US Department of Justice. 



Hesseling, R. (1995) "Theft from cars: reduced or displaced." 
European Journal of Criminal Policy and Research, 3: 79–92. 

Laycock, G. and C. Austin. (1992). "Crime Prevention in Parking 
Facilities." Security Journal. 3(3): 154–160. 

Laycock G. and N. Tilley (1995). "Implementing crime prevention." In 
M. Tonry and D.P. Farrington (eds.) Building a Safer Society: 
Approaches to Crime Prevention. Crime and Justice: A Review of 
Research, Vol.19. Chicago, IL: University of Chicago Press. 
p.535–584. 

Mancini, A.N. and R. Jain. (1987). "Commuter parking lots: vandalism 
and deterrence." Transportation Quarterly, 41: 539–53. 

Meredith, C. and C. Paquette (1992). "Crime Prevention in High-Rise 
Rental Apartments: Findings of a Demonstration Project." 
Security Journal. 3(3): 161–167. 

Phillips, C. (1999). "A Review of CCTV Evaluations: Crime Reduction 
Effects and Attitudes Towards Its Use". In K Painter and N. 
Tilley (eds.). Surveillance of Public Space: CCTV, Street Lighting 
and Crime Prevention. Crime Prevention Studies, Vol. 10. Monsey, 
NY: Criminal Justice Press. 

Poyner, B. (1991). Situational Crime Prevention in Two Car Parks." 
Security Journal, 2: 96–101. 

Riley, D. (1980). "An Evaluation of a Campaign to Reduce Car 
Thefts". In R.V.G. Clarke and P. Mayhew (eds.). Designing Out 
Crime. London: H.M.S.O. 

Read, T. and Tilley, N. (2000). Not Rocket Science? Problem-Solving 
and Crime Reduction. Crime Reduction Research Series Paper 6. 
London, UK: Home Office, Police Research Group. 

Sallybanks, J. and R. Brown (1999). Vehicle Crime Reduction: Turning 
the Corner. Police Research Series Paper 119. Policing and 
Reducing Crime Unit. London, UK: Home Office. 

Sandby-Thomas, M. (1992). Preventive Strategies to Reduce Car theft 
in Northern Ireland. Report No 2. Belfast: Extern Organization. 

Scott, M.S. (2000). Problem-Oriented Policing. Reflections on the First 
20 Years. Washington, D.C.: U.S. Department of Justice, Office of 
Community Oriented Policing Services. 

 

Scott, M.S. and R.V. Clarke (2000). "A Review of Submissions for the 
Herman Goldstein Award for Excellence in Problem-oriented 
Policing." In: C. Sole Brito and E. E. Gatto (eds.), Problem-
Oriented Policing: Crime-Specific Problems, Critical Issues and 
Making POP Work, Volume 3. Washington, D.C.: Police Executive 
Research Forum. 

Smith, M.S. (1996). Crime Prevention through Environmental Design 
in Parking Facilities. Research In Brief. National Institute of 
Justice. Washington, DC: U.S. Department of Justice. 

Spelman, W. and J. Eck (1989). "Sitting Ducks, Ravenous Wolves, and 
Helping Hands: New Approaches to Urban Policing." Public 
Affairs Comment, Vol. XXXV, No.2. 

Tilley, N. (1993). Understanding Car Parks, Crime and CCTV: 
Evaluation lessons from Safer Cities. London, UK: Home Office, 
Police Research Group. 

VCRAT (Vehicle Crime Reduction Action Team) (1999). Tackling 
Vehicle Crime: A Five Year Strategy. Communication Directorate. 
London: Home Office. 

Webb, B., B. Brown and K. Bennett (1992). Preventing Car Crime in 
Car Parks. Crime Prevention Unit Series, Paper 34. London: 
Home Office. 


DOJ Seal

FOR MORE INFORMATION: 
U.S. Department of Justice 
Office of Community Oriented Policing Services 
1100 Vermont Avenue, NW 
Washington, D.C. 20530 

To obtain details on the COPS Office program, call the 
U.S. Department of Justice Response Center at 1.800.421.6770 
Visit the COPS internet web site by the address listed below. 


e12021823 Created Date: January 29, 2003 


www.cops.usdoj.gov