Convenience Store Security Report | NACS – Research
Maintenance Underway
Parts of are unavailable until after the following maintenance window:
  • Jan 15, 5:00 PM - Jan 16, 5:00 PM (EST)
Thank you for your patience while we improve our systems.
Sign In Help

Advancing Convenience & Fuel Retailing


Skip Navigation LinksNACS / Research / Convenience Store Security Report

Convenience Store Security Report

It is specifically crimes of violence in convenience stores that concern this industry and that is the subject of this report. Three independent studies, examining various aspects of the crime issue, are summarized here along with discussion and recommendations that it is hoped will help advance industry knowledge in this most critical area.

Convenience stores are located in all areas of the United States and in all types of neighborhoods. Over eighty-eight percent (88%) of all Americans shop in convenience stores. Thus, the convenience store industry represents, in many respects, a functioning microcosm of our society. As such, the industry is susceptible to all of the phenomena to which society in general is subjected.

Any analysis of crime in convenience stores must be performed within the context of crime in America. For example, according to the FBI's 1990 Uniform Crime Report, robbery in the convenience store industry accounted for 6.1% of all robberies while over 56% of such crimes took place on streets or highways. In 1990, all categories of robbery increased, but convenience store robberies increased at the lowest rate (4%) of all categories, as compared with banks who experienced an 18% increase.

That the United States is a violent nation cannot be disputed. According to the U.S. Bureau of Justice "in 1991, the chance of being a violent crime victim is now greater than that of being hurt in a traffic accident." Nearly two million Americans were victims last year.

In 1990, the U.S. led the world with its murder, rape and robbery rates. Internationally, the U.S. murder rate quadruples Europe's and American women are eight times more likely to be raped than European women. Robbery in this country occurs at a rate 150 times that of Japan.

Criminologists and sociologists point to issues like gun control, prision overcrowding, recidivism, early release programs and social disorganization as some of the contributing factors to the tragedy of violence in America. What follows in this report does not examine those macro-issues which, it must be emphasized, certainly impact the issue of crime in convenience stores.

In 1973, industry participants and the Western Behavioral Sciences Institute (WBSI), under the direction of the late Dr. W.J. Crow, accomplished research in partnership with the Department of Justice on robbery prevention for convenience stores. Over 17,000 robberies were analyzed in that study which producted the deterrence measures in widespread use in the industry today. These measures include:

  • cash control;
  • clear sight lines into stores;
  • promoinent position of cash register;
  • eliminating escape routes;
  • balanced exterior and interior lighting; and
  • employee training.

Robbery Research Yields fifty Percent Rate Reduction, 1976-1986
From 1976-1986 the number of convenience stores doubled while the number of convenience stores robberies remained relatively constant. Indeed, the robbery rate per store was cut in half over that ten year period due to the implementation of the deterrence measure described above.

Despite these impressive gains, the Board of Directors of the National Association of Convenience Stores (NACS) regularly addressed the issue of crime. In 1989, the Board felt that it was an appropriate time to update the information gained in the 1973 research. For over a year, the Board invited input as to what might be helpful in broadening industry knowledge of the subject.

On March 21, 1990, the Board approved funding for three independent studies which are summarized in the pages that follow. These studies are entitled:

  1. 1991 National survey of Convenience Store Crime and Security-- a study to establish the base crime rates and distribution patterns for homicide, sexual assaults, robbery and other crimes of violence in convenience stores.

    Principal Researcher:
    F. Barry Schreiber, Ph.D.
    Professor of Criminal Justice
    St. Cloud State University
    St. Cloud. Minnesota
  2. Convenience Store Homicide and Rape -- a study to look at the detailed circumstances of these crimes for patters or trends.

    Principal Researcher:
    Rosemary J. Erickson, M.A.
    Athena Research Corporation
    Washington, D.C.
  3. An Assessment of Robbery Deterrence Measures at Convenience Stores: Multiple Clerk staffing, Central Station Based Interactive Television and Bullet-Resistant Barriers-- a study to analyze the effects of multiple clerks, bullet resistant barriers, and interactive television on robbery rates in convenience stores.

    Principal Researcher:
    Robert M. Figlio, Ph.D.
    Professor of Sociology
    University of California at Riverside
    Riverside, California

In addition to authorizing funding for these studies, the Board articulated an offical research objective:

"NACS will undertake research with the objective of determining what steps can be taken effectively to create the safest practical working and shopping environment for convenience store employees and customers. Such research will be designed to permit the drawing of factually supported conclusions, derived scientifically, as to the best practical means by which to evaluate real and perceived threats of crime at convenience stores and, thereby identify deterrent techniques to minimize such crime."

Moreover, the Board also established a 20-month time frame for the reporting of results, directing NACS to report the findings at the 30th Annual Meeting of the Association in Orlando, Florida in November 1991.

The level of participation by the membership of the National Association of Convenience Stores (NACS) is higher than any other study ever undertaken by NACS. Over 1,000 companies participated in at least one of the three studies for a rate of 82 percent of the member companies and 60 percent of the member store counts.

Based on its size and an analysis of other representative characteristics, the respondent sample is believed to be representative of NACS membership and the U.S. convenience store industry as a whole.

The number of companies, stores, and crimes indexed in these three studies make them the largest and most comprehensive analysis of convenience store crime.

Further, it is clear that the record number of participants and the countless hours of time and resources employed to get the necessary data assembled, depict the importance placed on this issue by the industry.

Key Findings
While the exective summaries of all three research projects are included in this report, what follows is an overview of the key findings across all three studies.

Distribution of Robbery: Seventy-nine (79%) of all U.S. Convenience Stores Robbery-Free
A main finding across all three studies is the fact that convenience store robbery is not evenly distributed across the industry. Indeed, nearly eighty percent of all stores in the U.S. did not have any robberies in 1990. Similarlly, seventy-eight percent (78%) were robbery-free in 1989.

The distribution pattern further finds that thirteen percent (13%) of all U.S. convenience stores experienced one robbery in 1990, while only seven percent (7%) had two or more that year. Essentially, sixty-five percent (65%) of all convenience store robberies were confied to six and one-half percent (6.5%) of all U.S. convenience stores.

Circumstances of Homicide: Eighty percent (80%) Classified as Acquaintance-based or Random/Senseless
The Homcide study was intended to review, in detail, the circumstances of this most heinous offense. It was hoped that some common thread or pattern would emerge from the data that might lead to new prevention programs. What emerged, however, was much different. Eightly percent (80%) of all homicides studied were either random senseless in nature (65%), that is, there appeared to be no apparent reason for the murder or they occurred between people who were prior acquaintances (15%).

In addition, less than $50.00 was stolen in almost half of the robbery/homicides and one third of the time there was no indication of a robbery. There was more than one perpetrator in forty-six percent (46%) of the cases and handguns were the weapon of choice seventy-one percent (71%) of the time.

Taken together, the homicide information does not enable one to draw any conclusions about how to deter future events. The combination of senseless, unprecipitated killing with a large percentage of murders occuring between acquaintances is hard to guard against, but homicide must be rigorously evaluated on an ongoing basis for new clues.

One Clerk vs. Two Clerks: Overall, No Statistically Significant Effect on Robbery Rate Reduction
There was a five hundred and seventy-six (576) store sample studied for information about the effect of one vs. two clerks. These stores were studied over a six and one-half year period, in the middle of which, two hundred and thirty (230) stores implemented a two clerk practice on the third shift.

Consistent with robbery distribution information, there were a large percentage (almost half) of these stores that were never robbed in the entire six and one-half year period. Indeed, robbery-free stores were the predominant group. Further, there is not a statistically significant relationship overall between crime rate reduction and two clerks apparent from this study. In fact, for stores in the sample that had experienced no robberies prior to the 1988 implementation of two clerks, the post 1988 robbery rate was actually higher (0.43 per store per year) than that of similarly situated one-clerk stores (0.33 per store per year).

There is, however, a very small number of stores in the sample (less than 20 stores) that experienced repeat robberies. An analysis of the one clerk vs. two clerk issue for these stores provides data that two clerks may produce a robbery rate reduction at repeat robbery locations. While many other factors like fluctuation in neighborhood crime levels, store operating practices, and an over-time regression to the mean may influence this small sample, the finding invites additional investigation.

Time of Day/Day of Week/Month of Year: Convenience Store Crime Mirrors Crime in U.S.
With respect to information on homicide and rape relative to time and day of occurence, all of the data closely parallels the FBI national distribution data. For example, homicide, both in convenience stores and in the U.S. is most likely in July and least likely in February. The same is generally true for rape where rape in convenience stores occured most frequently in July compared to August in the FBI report, but both rape in convenience stores and rape in the U.S. occur least frequently in February.

For time of day, the 1990 National Crime Victimization Survey reports that seventy-one percent (71%) of rapes occur at night which is consistent with our finding of eight-nine percent (89%) occuring at night.

Overall, these similarities define the fact that every little difference exists between U.S. and industry data relating to when violent crime occurs.

Robbery Counts: Industry Count Differs From FBI Data
The industry-wide census performed as part of this research obtained responses from 1,024 companies or eight-two percent (82%) of the membership of the National Association of Convenience Stores. This is the largest sample in existence on which robbery data for this industry has been assembled. For 1990, however, the industry counted 22,935 robberies compared to the FBI's count of 38,995-- indicating a wide discrepancy. While many reasons for this may exist, likely differences lie in how locations are classified and how the term robbery is defined. For example, gasoline drive-offs, shoplifting, and internal thefts are not included in the industry's count. More needs to be known about the difference and reconciliation with the FBI is one way of approaching the question.

Closed Circuit Interactive Television Systems: First Year Reduction, Second Year Increase
Closed circuit interactive television systems do not appear to have a consistent or dramatic effect on crime reduction. The study looked at 189 stores where CCITV systems were in place and for which one year of pre-installation crime data and two years of post-installation crime data were known. In this sample, there was a statistically significant reduction in robbery rates during the first year after system installation. However, data for the second year showed an increase in robbery rates, although not quite to preinstallation levels.

CCITV systems depend on proper installation, maintenance, and adequate central station response for their success. This study was not capable of evaluating those factors, and therefore their effect in years one and two are unknown. Therefore, while there appears to be an overall direction of crime rate reduction, it is not statistically significant.

High-resolution Color Monitor Systems: Significant First Year Reduction
While not an original target of the study, eight-one stores (81) were observed for one year before and one year after the installation of state-of-the-art video equipment. The stores had color cameras with color monitors installed in such a manner that customers and would-be robbers could see that they were on camera. In addition, the unit was mounted with a large, easily visible microphone on a pedestal where its presence was obvious.

Other Findings
As with the homicide study, the rape study was intended to identify patterns among the cases for possible prevention strategies. What was found was theat the crime of rape is typically not committed in conjunction with a robbery. Two-thirds, in fact, were non-robbery related. Nearly half the time when robbery was present, less than $50.00 was stolen. It was also found that the perpetrators acted alone in ninety-two percent (92%) of the cases and handguns, as with homicide, were the weapon of choice forty-one percent (41%) of the time.

Bullet-resistant Barriers
Of the stores in this study, less than one percent (1%) had locations with bullet resistant barriers, rendering an attempt to analyze their effect impossible.

As with all research, some questions are answered while others are raised. In general it's clear that certain convenience stores experience crime while the vast majority are crime free. This distribution drives much of the studies' findings in that most countermeasures have relatively insignificant overall effects because most stores don't have a crime problem.

 Please note: The above resources represent our best efforts to gather useful information from available and reliable sources. However, we cannot accept liability due to errors or omissions and this is not a subsititue for legal or other professional advice.  Reliance upon this material is solely at your own risk.

The Police Department of Overland Park, Kansas Fact Sheet for Crime Prevention: Convenience Store Robbery Reducing Criminal Opportunity. (PDF)