Unanticipated Consequences

Abstract

Among scholars, there is a discussion regarding whether types of places, or facilities, function as crime generators or whether the association between some categories of facilities and higher rates of offending is the result of a small proportion of all facilities within a given category, or problem places. This study seeks to further inform this debate by exploring whether policy changes that alter the social functioning of a category of facilities, specifically bars and taverns, modifies the spatial association with crime. Using routine activities theory as a framework, this study builds on previous research by exploring the association between alcohol-serving establishments and violent crimes, specifically assaults, following the implementation of a smoke-free law. Using data from a pair of adjoining communities in Iowa, findings indicate the frequency of reported assaults on blocks with bars as well as on adjoining blocks declined following the implementation of a law prohibiting smoking tobacco products within bars and taverns. Implications for policies and future research are discussed.

Keywords

ecology and crime/spatial analysis crime/delinquency theory violent behavior other drugs and crime

Introduction

Beginning with early research identifying hot spots of crime, researchers have been examining whether specific types of places, also referred to as facilities (Brantingham & Brantingham, 1995) and institutions (Peterson, Krivo, & Harris, 2000), function as crime generators. One line of inquiry has focused on analyses identifying a relationship between alcohol-serving establishments and violent crimes, specifically assaults. This line of research often finds a spatial association between the presence of bars and the frequency of assaults (Gorman, Speer, Gruenwald, & Labouvie, 2001; Nielsen & Martinez, 2003; Roncek & Maier, 1991; Sherman, Gartin, & Buerger, 1989).

Some scholars have suggested that such inquiries into the relationships between types of facilities and criminal offenses may be misleading, as these associations may not result from institutions themselves, but rather that the facilities are often located in areas characterized by higher volumes of traffic, bringing together large numbers of potential victims and motivated offenders and increasing criminal opportunities (Eck, Clarke, & Guerette, 2007; Wilcox & Eck, 2011). In other words, the spatial association between categories of facilities and criminal offending may be spurious.

One option for informing the discussion of whether or not categories of facilities function as crime generators is to examine whether changes to policies directed toward social functioning of a category of facilities alter the level of offending in and around these institutions. Where facilities contribute little to the level of offending, policies altering facility functioning should have little impact on offending. In contrast, if facilities are functioning as crime generators, then changes in institution functioning should correspond to changes in levels and rates of offending around the impacted establishments.

The current study explores whether the implementation of a smoke-free law changed the frequency of violent crimes taking place around alcohol-serving establishments in adjoining communities in Iowa. Specifically, the current study uses nearly 3 years of calls for service data to examine whether the number of reported assaults on blocks with bars changed following the implementation of a law prohibiting the smoking of tobacco products within all public restaurants and bars in neighboring communities. We begin with a review of the research examining the association between bars and assaults, which is followed by a summary of the discussions regarding whether categories of facilities function as crime generators and how a policy change can inform that discussion, a review of clean indoor air laws, which is then followed by a theoretical framework through which the current study was conducted. After presenting the findings, we conclude with a discussion of policy implications and recommendations for future research.

Bars, Taverns, and Assaults

Over the past three decades, research has focused on identifying and explaining the relationship between different types of facilities and variation in offending. This line of research gained substantial attention in 1989 when Sherman, Gartin, and Buerger were among the first to empirically examine variation in the density of crime down to the level of addresses, or places, within a city. This contrasted from previous research that had focused largely on explaining variation across neighborhoods (Sampson, Raudenbush, & Earls, 1997; Shaw & McKay, 1942). Relying on calls for service data, Sherman, Gartin, and Buerger (1989) found that most calls, particularly calls reporting domestic disturbances and assaults, originated from a small proportion of “hot spots” relative to the rest of city, even within the most violent neighborhoods. Specifically, around 50% of calls originated from just 3% of addresses within the city (Sherman et al., 1989). In addition, Sherman and his colleagues found bars and taverns often functioned as places reporting high frequencies of assaults. Later studies sought to further clarify the relationship between recreational alcohol-serving establishments and assaults.

In an examination focusing on whether bars contribute to an increased likelihood of assaults around bars, Roncek and Maier (1991) analyzed whether the presence of a bar on a block is associated with more violent crime independent of socioeconomic and demographic characteristics of residential blocks (see also Roncek and Bell, 1981). Using data from Cleveland, Roncek and Meier (1991, p. 742) demonstrated that a bar on a block increased the probability of an assault occurring on a block by roughly 20%. Research making use of varying degrees of geographic aggregation including census block groups (Gorman et al., 2001; Gruenwald, Freisthler, Remer, LaScala, & Treno, 2006), census tracts (Nielsen & Martinez, 2003), and postal codes (Livingston, 2008) has also identified a relationship between bars and violent crime. The finding by Roncek and Maier (1991) as well as similar findings by others (Frisbie, Fishbine, Hintz, Joelsons, & Nutter, 1977; Nielsen & Martinez, 2003; Peterson et al., 2000) suggests that this type of facility may function as a crime generator contributing to the spatial concentration of assaults.

The Question of Categories of Facilities as Crime Generators

There is an ongoing discussion of whether or not categories of facilities function as crime generators, altering the spatial distribution of crime by concentrating offenses within and near institutions. On one hand, social functioning of specific types or categories of facilities brings together targets and offenders while providing inadequate guardianship, thus increasing criminal opportunities across places within the categories (Roncek and Maier, 1991). With evidence from a substantial body of research indicating associations between types of facilities and offending often cited in favor of the facilities acting as crime generators (Kubrin, Squires, Graves, & Ousey, 2011), policies directed toward altering the social functioning of categories of facilities should likewise alter the incidence of offending around the affected establishments.

On the other hand, inquiries into associations between categories of facilities and offending may create misleading impressions that these specific categories are functioning as crime generators. Wilcox and Eck (2011) question whether researchers are successfully identifying categories of crime-generating facilities or whether a general effect of most places on crime is being observed, noting that offending is not evenly distributed within categories of facilities, but rather it tends to concentrate among a few problem places or establishments. In their explanation of the Iron Law of Troublesome Places, Wilcox and Eck argue that facility functioning has little bearing on the crime-generating properties of facilities. Rather, higher rates of offending among particular problem places within each category of facility are likely driven by higher levels of traffic as well as issues surrounding place management, the number and quality of potential victims, and the presence of highly active offenders unique to each troublesome place (see Eck et al., 2007). As a result, policies directed toward altering social functioning of facilities are likely to have little impact on crimes taking place in and near the facilities.

A policy change directed at altering the social functioning of a category of facilities provides an opportunity to inform the discussion regarding whether that type of facility is a crime generator or whether this type of institution tends to locate in high crime areas without significantly contributing to offending in the area. As noted previously, if a category of facility is contributing little to offending, policies altering social functioning within and around this type of facility should have little impact. In contrast, if facilities are functioning as crime generators, then policy changes impacting facility functioning should correspond to changes in levels of offending. One such policy change could include the implementation of smoke-free or clean indoor air laws prohibiting the smoking of tobacco products within bars and taverns.

Clean Indoor Air Laws and Policies

In addition to serving as locations for patrons to imbibe alcoholic beverages, bars and taverns provide a setting in which smoking often takes place, including among individuals who do not routinely identify as smokers (Room, 2004). As a public health concern, mortality from smoking and exposure to secondhand smoke is much more common than deaths resulting from violent crimes. The U.S. Department of Justice reported 16,137 murders nationwide in the United States in 2004 (Federal Bureau of Investigation, 2005). In contrast, an estimated 443,000 people, roughly 27 times more than murder victims, died as the result of smoking or exposure to secondhand smoke during the same time period (Adhikari, Kahende, Malarcher, Pechacek, & Tong, 2008). Not surprisingly, over the last decade, many states, counties, and municipalities have tried addressing the health-related consequences by initiating prohibitions on smoking in public places including bars and restaurants using smoke-free or clean indoor air laws. In 2000, no state prohibited smoking in bars and taverns. By 2010, 26 states and the District of Columbia had clean indoor air laws that included bars and taverns (Tynan, Babb, MacNeil, & Griffin, 2011).

Debates surrounding the adoption of clean indoor air laws that prohibit smoking in public spaces, including bars and taverns, have been directed largely by public health concerns, often focusing on the impact of secondhand smoke on employees and nonsmoking patrons (Collins, Shi, Forster, Erickson, & Toomey, 2010). Opponents of such laws often focus on the economic implications for businesses that would be impacted, arguing that it would diminish their profitability (see Jones & Holl, 2006).¹ Proponents argue that expected outcomes such as improvements to public health would outweigh any potential short-term negative economic outcomes. However, policy changes may also produce unanticipated consequences (Merton, 1936). In the context of the adoption of a smoke-free law, changes to the laws regulating recreational alcohol-serving establishments may also alter the social functioning within the facilities. In turn, this may produce changes beyond anticipated improvements to public health. Legal changes altering behavior in and around recreational alcohol-serving establishments may also alter the social structure of criminal opportunity in spaces outside the impacted bars and taverns. Indeed, following the implementation of a prohibition on smoking within alcohol-serving establishments, patrons and employees may increasingly gather outside the impacted businesses in order to continue smoking tobacco products (Moore, Annechino, & Lee, 2009). Routine activities theory can serve as a framework for examining how changes to regulations impacting facility functioning may also alter in criminal opportunities in and around establishments.

Theoretical Framework

According to routine activities theory, criminal events result from the convergence of three conditions in time and space: exposure to motivated offenders, the presence of suitable targets, and the absence of capable guardianship (Cohen & Felson, 1979). The absence of any one of these conditions significantly reduces the likelihood of a criminal event. Conceptual definitions of these three elements generally vary by crime type and situation. For the purposes of this analysis, the concepts are elaborated upon as they relate to opportunities for assaults around recreational alcohol-serving establishments.

The first condition related to the likelihood of criminal events is exposure to motivated offenders. Alcohol use is frequently associated with violent behavior (Felson & Staff, 2010). Although most people who drink alcohol are unlikely to exhibit violent behavior, for a certain proportion of the population, alcohol can function as a catalyst for aggression (Block & Block, 1995; Felson & Staff, 2010; Heinz, Beck, Meyer-Lindenberg, Sterzer, & Heinz, 2011). Indeed, bars and taverns often function as a location where individuals with histories of deviant and risk-taking behavior often congregate (Felson & Staff, 2010; Hindelang, Gottfredson, & Garofalo, 1978). With higher densities of potentially motivated offenders gathering at this type of facility, it is not surprising that previous research finds a strong spatial relationship between bars and violent crime (Gorman et al., 2001; Nielsen & Martinez, 2003; Roncek & Maier, 1991).

In addition to functioning as potential motivated offenders, bar patrons may present as suitable targets. Individuals who visit alcohol-serving establishments and imbibe may increase their likelihood of victimization by engaging in more risky behavior (McClelland & Teplin, 2001) or, to some extent, becoming incapacitated, which reduces their ability to avoid confrontations (Homel, Tomsen, & Thommeny, 1992). Indeed, previous research finds an increased risk of assault of frequent and heavy drinkers when they are drinking (Felson & Burchfield, 2004). The increased risk of victimization among frequent and heavy drinkers may result from frequenting bars and taverns, increasing the likelihood of encountering motivated offenders.

The third condition associated with the probability of criminal events is the lack of capable guardianship. There are three types of guardianship that reduce the likelihood of criminal events: guardians, handlers, and managers (Felson, 1995). Guardians are responsible for monitoring targets, handlers provide guardianship by controlling the actions of potential offenders, and managers monitor places either as their primary duty or in the course of their other duties. Bars create situations whereby all forms of guardianship are reduced. Guardians and handlers may be intoxicated and unable to discourage crime effectively or may not be physically present. Establishments with high densities of patrons or with high turnover may also decrease the effectiveness of place managers to function as a source of guardianship (Roncek & Maier, 1991). Finally, the discouragement potential of place managers may largely depend on their level of responsibility. For instance, security staff members are often strictly assigned place manager duties, while other bar staff have only a diffuse responsibility to monitor patrons (Homel & Clark, 1995).

Impact of Smoke-Free Laws on Criminal Opportunities Around Bars

There are many ways in which the implementation of a clean air law may alter the social functioning of bars in taverns increasing criminal opportunities. One response to these laws is for tobacco users to leave the premises of affected businesses, yet remain nearby, while smoking, often in groups (Moore et al., 2009). One study directly examined the social conditions outside bars and taverns following the implementation of a statewide smoke-free law. Using direct observations and interviews outside bars and taverns across California following the implementation of a statewide law prohibiting smoking within establishments, Moore, Annechino, and Lee (2009) observed that smoking frequently occurs among groups around the front door of recreational alcohol-serving establishments. Likewise, they noted oftentimes the situation was tense or felt unsafe to observers as well as to female patrons and employees. Specifically, Moore, Annechino, and Lee (p. 141) stated, “regardless of neighborhood, the space immediately outside the bar could be a dangerous zone ….” One reason given for the tension among smokers was the expulsion of unruly patrons from bars and taverns (Moore et al., 2009). Similarly, smoking bans may increase the number of suitable targets and motivated offenders by increasing alcohol consumption among patrons. Researchers using revenue data from bars and taverns find that overall sales (Cowling & Bond, 2005) and specifically liquor sales revenue (Collins et al., 2010) increase following the implementation of smoke-free laws. Given these changes to social functioning within and around bars and taverns following the implementation, there is reason to expect an increase in the number of offenses taking place around the affected establishments.

While Moore and his colleagues (2009) note anxiety surrounding the potential for conflict outside establishments among patrons smoking tobacco products, it is not clear whether the implementation of the smoke-free law changed the threat of violent encounters. As previously noted, other research has established an association between recreational alcohol-serving facilities and assaults in communities without smoking prohibitions in bars and taverns. Indeed, there are several characteristics of the altered social functioning outside bars following the implementation of a smoke-free law that may reduce criminal opportunities. Patrons congregating outside of bars and taverns to smoke may reduce the risk of victimization for fellow-gathered smokers by functioning as guardians for one another supplementing guardianship that may be provided by one or more security staff members assigned to business entrances (Homel & Clark, 1995). Likewise, groups of smoking patrons and employees may also function as handlers, discouraging aggressive patrons from behaving violently toward other patrons and employees gathered outside the establishments to smoke as well as customers entering and leaving bars and taverns.

A third potential outcome of the implementation of a smoke-free law including bars and taverns may be no changes to the frequency of victimization. In this scenario, the level offending is unrelated to facility functioning of the category of institutions; rather is the result of this type of facility tending to locate in high-traffic areas with greater opportunities for offending (Wilcox & Eck, 2011).

The current study seeks to build on the discussion of whether categories of facilities contribute to nearby offending by examining whether the implementation of a clean indoor air law alters the relationship between a type of institution and reported victimization. In particular, the current study examines whether the implementation of a smoke-free law alters the frequency of assaults around bars and taverns using data from a pair of adjoining cities in Iowa. With the increasing adoption of laws regulating where tobacco smoking may take place, the current study contributes to the research by exploring whether changing the social functioning in and around bars through policy changes also alters the frequency of violent offenses around the affected establishments. Likewise, the current study also seeks to inform policy makers and other community stakeholders of a possible unanticipated consequence of adopting laws regulating where smoking of tobacco products may take place—a change in the frequency of assaults around bars and taverns.

Research Setting

While much of the research examining the relationship between alcohol-serving establishments has taken place in communities with larger populations such as Cleveland (Roncek & Bell, 1981) or Miami (Nielsen & Martinez, 2003), the current study is set within a pair of adjoining smaller cities in Iowa. Cedar Falls, a community of roughly 40,000 residents, is the location of one of the three public universities in Iowa and has a large product development campus for a large agricultural manufacturer (U.S. Census Bureau, 2011). With roughly 10,000 students, the university contributes to significant seasonal variation in population. Waterloo, a community of 70,000 residents, has an economy driven largely by manufacturing of agricultural equipment and processing of agricultural products (U.S. Census Bureau, 2011).

Differences in economic bases between the two communities are reflected in household measures. Census figures indicate households in Waterloo have lower annual median incomes on average than those in Cedar Falls with reported incomes of US$38,779 and US$45,951 in 2009 dollars, respectively (U.S. Census Bureau, 2011). Regarding the racial and ethnic characteristics of the population, findings from the 2010 Census indicate Waterloo is more heterogeneous than Cedar Falls with 16% of Waterloo residents identifying as African American and another 6% as Hispanic. Seventy-seven percent of residents in Waterloo self-categorize as White. In contrast, over 93% of residents of Cedar Falls identify themselves as White (U.S. Census Bureau, 2011).

While these communities border one another, they vary substantially in the frequency of crime from one another as well as from national averages for similar size cities. In 2010, the overall violent crime rate of 180 offenses per 100,000 in Cedar Falls was well below the national average of 326 for cities with populations between 25,000 and 50,000 (Federal Bureau of Investigation, 2011). Likewise, an aggravated assault rate of 143 was reported for the community compared to a national rate of 202 offenses in similar sized cities. In contrast, the reported assault rate Waterloo was well above the national averages for similarly sized cities with 490 aggravated assaults per 100,000 residents compared to a national rate of 242 for cities with populations between 50,000 and 100,000 (Federal Bureau of Investigation, 2011). Together, the neighboring communities provide an opportunity to expand the current research examining the association between recreational alcohol-serving establishments and assaults beyond large cities. While these cities are not necessarily representative of other communities with similarly sized populations, together they provide a good setting to examine the impact of a policy regulating facility functioning on assaults around bars.

On July 1, 2008, a state statute, the Smokefree Air Act (2008), went into effect prohibiting smoking in many public locations including all bars and taverns open to the public within Iowa. Following the implementation of the law, bar and restaurant patrons and employees interested in smoking were required to go to unenclosed spaces outside of the affected establishments. While the law specifically prohibited smoking in bars and restaurants, stairways, stairwells, and entrances to businesses, it did not include a buffer onto surrounding sidewalks, streets, or parking lots. At the time that the law took effect, neither Cedar Falls nor Waterloo had prohibited smoking tobacco products within bars and taverns. As a result, the implementation of the law, combined with enforcement powers and actions by state agencies, changed rules regarding facility functioning for all bars and taverns in both communities, and provides an opportunity to examine how policies directed at a category of facilities may or may not change criminal opportunities.

Data and Methodology

The unit of analysis in the current study is residential city blocks, specifically all 2,188 residential city blocks in Cedar Falls and Waterloo, Iowa. Findings by Sherman et al. (1989) demonstrate the value of using geographic units more spatially disaggregated than neighborhoods after finding that even in neighborhoods with the highest crime rates, crime is highly concentrated among a few problem locations also referred to as “hot spots.” Likewise, while some research assessing the relationship between bars and assaults has used larger geographic units of analysis such as census tracts and zip codes (Britt, Carlin, Toomey, & Wagenaar, 2005; Livingston, 2008; Nielsen & Martinez, 2003; Speer, Gorman, Labouvie, & Ontkush, 1998), other research has demonstrated city blocks are well suited for such analyses (Roncek & Bell, 1981; Roncek & Maier, 1991). While the data used in these analyses usually do not specify whether specific offenses were associated with particular establishments, when examinations rely on more refined geographic units of analysis such as blocks, it is more likely that offenses occurring within the smaller spatial area around each establishment are an outcome of the social functioning within and around the establishments.

Dependent Variables

All calls for service data reporting assaults, which had taken place in either community for a period spanning from January 1, 2007, through November 15, 2009, were used to identify the location of assaults. While previous research has found spatial associations between bars and various violent crimes, this study focuses on assaults due to the frequency of this type of offense and the strength of the relationship between this category of facilities and reported offenses. During the time frame of the current study, assaults comprised nearly three of the four (72%) reported violent crimes in the adjoining communities (Federal Bureau of Investigation, 2008, 2009, 2010). Likewise, Roncek and Maier (1991) found that among several violent crimes, the number of assaults had a particularly strong relationship with the presence of one or more bars or taverns on a block. A total of 3,583 calls reporting assaults occurred over the time frame. The time span allows for an examination of assaults around bars for nearly 18 months both before and after the law prohibiting smoking within bars and taverns took effect on July 1, 2008. Each call included the date of the reported offense that was used to identify whether each incident occurred prior to or following the implementation of the smoke-free law. Likewise, with each reported offense, operators recorded the current location of the victim as well as the location of each assault as reported by the caller. Based on the reported location of each incident, the current study used a geographic information system (ArcGIS 10) to geocode each event and subsequently link each reported offense with a spatially corresponding census block. Offenses occurring at intersections were randomly assigned to one of the adjoining blocks. A standard address geocoding procedure was implemented using the Waterloo-Cedar Falls street layer as a reference. After removing offenses that took place outside the adjoining communities, 2,850 offenses were geocoded onto 2,790 residential and nonresidential blocks. With social characteristics of blocks limited to residential blocks, analyses were limited to 2,557 offenses on 2,188 blocks. Spatial join was utilized to aggregate reported incidents to a census block so that each block would contain a count of offenses that could be combined with social characteristics.

Relying on calls for service to identify the location of offenses has several limitations. It is possible a single incident could be included in multiple reports. Conversely, findings from the National Crime Victimization survey indicate that assaults often are not reported to the police (Rand & Robinson, 2011; Sherman et al., 1989). Nationally, 40% of aggravated assaults and 53% of simple assaults were not reported to the police in 2010 (Truman, 2011). Indeed, changes in the frequency of reported offenses in the current study could be the result of changes in reporting behaviors by victims and witnesses. Additionally, dispatchers exercise discretion in classifying offenses that may result in undercounts of assaults if they are identified as disturbances or disorderly conduct. Finally, using calls for service is an overinclusive measure of crime, including incidents that upon further examination were found not to be criminal. In some cases, police “unfound” reported offenses where a report of an incident is made to the police that upon investigation is found not to be criminal (Federal Bureau of Investigation, 2004).

Despite the limitations of using calls for service, this source of data is often used when the focus of a study is more spatially refined than cities, neighborhoods, or census tracts (Sherman et al., 1989). Calls for service data provide a measure of crime counts down to individual addresses, which allows for analyses with smaller geographical units than can be accomplished with other sources of data such as victimization surveys (Roncek & Faggiani, 1985).² Likewise, using calls for service data eliminates the problem of selection bias introduced by any systematic variation in recording of reported offenses by police officers (Klinger, 1997; Varano, Schafer, Famega, & Swatt, 2009). Indeed, data from the National Crime Victimization Survey indicate when victims report assaults, officers respond to 90% of victims and complete reports in less than half of the cases (Rand & Robinson, 2011).³

Using the calls for service data, three dependent variables are used to explore the relationship between bars and assaults both before and after the implementation of the law prohibiting smoking tobacco products in bars and taverns. Two variables are counts of reported assaults per block with one measuring offenses taking place within 18 months prior to the implementation of the clean indoor air law and the second measuring counts of offenses taking place postimplementation.⁴ A third dependent variable measures changes in the number of reported assaults per block through calculating the difference between the number of reported assaults per block prior to the implementation of the smoke-free law and postimplementation.

Independent Variables

The key independent variable in the current study is the presence of bars and taverns on a block. Reflecting previous practices, bars and taverns were identified using listings from a pair of regional phone directories and included all establishments listed as bars and taverns or were listed as restaurants but included “bar” or “tavern” in the name of the establishment (Polk City Directories, 2007a, 2007b, 2009a, 2009b; Roncek & Maier, 1991).⁵ Locations of bars and taverns were identified before the implementation of the smoking ban using 2007 listings, while post-smoking ban businesses were identified with 2009 listings and a pair of variables identifying whether or not each block contained a bar or tavern were created and included in corresponding analyses either before or after the clean air law was implemented. By separately identifying whether each block had one more bar in 2007 and 2009, we were able to account for the opening and closing of businesses. The variables bar block pre and bar block post measures whether a given block contained one or more bars or taverns during corresponding time periods.

In addition the presence of bars or taverns on blocks, additional independent variables were selected based on their relevance to spatially centered theories of criminal behavior and findings from previous block-level analyses of the distribution of assaults (Murray & Roncek, 2008; Roncek & Maier, 1991). The data for these variables originate from the 2000 Census and were obtained from the U.S. Census Bureau.⁶ Racial and ethnic composition of each block was measured with two variables, percentage Hispanic and percentage African American, based on the proportion of residents who identified themselves either as Hispanic or as African American, respectively. Previous research has found these variables are often related to the frequency of reported assaults on blocks (Peterson et al., 2000; Roncek & Maier, 1991). Other measures are often included in spatial analyses of associations between facilities and offending for their relevance to social disorganization as well as routine activities theory include percentage single-headed household (percentage of female single-headed households with children), percentage alone (percentage male-headed householders living alone), percentage youth (percentage of the population under age eighteen), and percentage vacant (percentage of vacant residences; Murray & Roncek, 2008; Nielsen & Martinez, 2003; Peterson et al., 2000).

Analysis

In the analysis, multivariate models are used to explore how a smoke-free law changes the relationship between bars and reported assaults. First, separate models are run independently examining whether an association is present between whether a block has a bar or tavern and the total number of reported assaults on a given block both before and after the implementation of a smoke-free law. With counts of reported assaults per block as a dependent variable, analyses were first conducted using a Poisson model. However, results yielded violations of the overdispersion assumption. As a result, the current analysis presents results from separate negative binomial regression models examining the relationship between bars on blocks and the likelihood of assaults taking place on and around those blocks. To further explore the relationship, an ordinary least squares (OLS) model with the change in counts of reported crimes per block before and after the implementation of the law is presented.

Previous research indicates that crime is not evenly distributed across communities, but rather it is often clustered creating concerns of spatial autocorrelation (Anselin, Cohen, Cook, Gorr, & Tita, 2000; Townsley, 2009). Within the context of the current study, spatial autocorrelation occurs when blocks with higher counts of reported assaults are located near other blocks also with higher counts of reported assaults. This nonrandom distribution of events may result in correlating error terms in regression models, violating the assumption of independence and, as a result, biasing estimates. One of the methods to correct such a problem is to introduce a spatial lag model (Anselin & Bera, 1998). In this case, a regression model includes a spatially lagged independent variable that is a weighted average of a given variable computed for a predefined block (Anselin, 2004).

Standard autocorrelation diagnostics were run to determine whether models developed in this study were affected by autocorrelation. Tests showed that the pre-ban and post-ban models needed to be adjusted for autocorrelation (Moran’s I was .22 and .23, respectively, both significant). Spatial lag terms were computed based on the queen’s neighborhood principle using the GeoDa algorithm (Anselin, 2004; Smirnov & Anselin, 2001) available within the GeoDa 0.9.5-i geostatistical package. Model 3 did not exhibit an autocorrelation problem.

Results

Both before and after the clean air act was implemented, about 2% of blocks had one or more bars. Fifty individual businesses located on 41 blocks were located for the initial, preclean indoor air law wave, while 47 bars and taverns situated on 37 blocks were identified in the post-ban wave. Before the clean air act was implemented, 1,401 offenses were identified on 650 residential blocks, while 1,156 assaults were located on 564 blocks after the smoking ban. During the pre-ban interval, 95 offenses were reported on 27 bar blocks, while in the post-ban time frame 47 reported assaults were located on 17 bar blocks. Descriptive statistics and correlations for both the independent and dependent variables are presented in Table 1. The correlations reveal that the presence of one or more bars on a block or an adjacent block is positively related to the number of reported assaults both prior to (.11) and after (.04) the law prohibiting smoking in bars and taverns went into effect. Likewise, there is a negative correlation between the pre-/post-ban change variable and pre-ban bar blocks (−.12).⁷

Table 1.

Correlations, Means, and Standard Deviations of Dependent and Independent Variables.

	1	2	3	4	5	6	7	8	9	10
1. Assaults pre-ban	1.00
2. Assaults post-ban	.79**	1.00
3. Bar block pre-ban	.11**	.04**	1.00
4. Bar block post-ban	.09**	.04*	.90**	1.00
5. Percentage Hispanic	.08**	.07**	.07**	.04	1.00
6. Percentage African American	.11**	.14**	−.01	−.01	.04*	1.00
7. Percentage single-headed households	.11**	.11**	.00	.00	.16**	.48**	1.00
8. Percentage alone	.04	.01	.15**	.14**	−.04	.01	−.09**	1.00
9. Percentage youth	.09**	.09**	.05*	−.04*	.16**	.29**	.55**	−.38**	1.00
10. Percentage vacant	.06*	.06**	.01	−.00	.12**	.22**	.12**	.14**	.01	1.00
M	.64	.53	.02	.02	2.11	12.38	5.57	12.81	22.36	4.14
SD	2.11	1.74	.14	.13	6.30	25.38	8.55	14.88	13.00	7.95

*p < .05. **p < .01.

Table 2 shows the results for counts of reported assaults for all residential blocks in the adjoining communities. In order to show the change in the relationship between the presence of bars on blocks and changes that occurred following the implementation of the law prohibiting smoking in recreational alcohol-serving establishments, these two models show the negative binomial regression results for offenses occurring before and after the effective start date of the law. In these results, bar block is positively associated with counts of reported assaults both before and after the implementation of the clean air law. Having at least one or more bars on a block or adjoining such a block increased the expected number of assaults on that block by 110% before the law took effect. This finding is similar to research by Murray and Roncek (2008) who used data from Omaha, Nebraska, in their examination of assaults around bars. Following the implementation of the smoke-free law, having at least one or more bars on a block or an adjoining block increased the expected number of assaults by 73.2% compared to similar nonbar blocks. The findings among the other independent variables such as percentage Hispanic, percentage African American, and percentage youth performed as expected with significant and positive relationships to the number of reported offenses (Murray & Roncek, 2008; Nielsen & Martinez, 2003). While the relationship between the presence of a bar on a block or an adjacent block remains positive and significantly related to the number of assaults, the strength of that relationship diminished once the clean indoor air law was in effect.

Table 2.

Negative Binomial Results for Reported Assaults Pre- and Post-Smoking Ban Implementation for All Residential Blocks.

	Model 1: Pre-ban		Model 2: Post-ban
	B (SE)	Exp(B)	B (SE)	Exp(B)
Bar block pre	1.108** (.208)	3.027	—	—
Bar block post	—	—	.732** (.243)	2.079
Percentage Hispanic	.011* (.005)	1.011	.018*** (.006)	1.018
Percentage African American	.003 (.002)	1.003	.010*** (.002)	1.010
Single-headed households	.009 (.005)	1.009	.009 (.006)	1.009
Percentage alone	.008** (.003)	1.008	.004 (.003)	1.004
Percentage youth	.013*** (.004)	1.013	.011** (.004)	1.011
Percentage vacant	.005 (.004)	1.005	.008 (.005)	1.008
Spatial lag	.965*** (.107)	2.626	.205*** (.036)	1.144
Intercept	−1.716*** (.114)	0.180	−1.463*** (.111)	.232

*p < .05. **p < .01. ***p < .001.

To further explore the change in the relationship between the presence of one or more bars on blocks and reported assaults, a change variable was calculated. This variable measures the difference between the number of reported crimes before and after the smoke-free law with positive figures indicating a higher number assaults taking place on a given block following the ban. To measure whether the change is associated with bar blocks, an independent variable, bar block pre-ban, is included, indicating whether the block had one or more bars before the law was implemented. To account for changes in presence of bars for some blocks by 2009, a new variable, bar block change, was also created. This variable ranged from −1 for blocks that changed from a bar block to a nonbar block, to 0 for blocks that did not change, to 1 for nonbar blocks that changed to bar blocks. Results from the OLS regression model are presented in Table 3.⁸

Results in Table 3 show a negative and significant relationship (−.95) between blocks with one or more bars and changes in the number of reported assaults following the implementation of a smoke-free law. This finding indicates that of the blocks with bars, on average, one fewer assault was reported after the law was implemented than before when compared against nonbar blocks. This finding indicates that changes to policies impacting facility functioning may result in changes in criminal opportunities in and around a type of institution, which also suggests that this category of facility functions as a crime generator (Kubrin et al., 2011). Likewise, the variable indicating changes in block status is also related to changes in the frequency of reported offenses. Nonbar residential blocks that changed to bar blocks had more reported assaults, while bar blocks that changed to nonbar blocks had fewer offenses.

Table 3.

OLS Results for Difference in Number of Reported Assaults 18 Months Pre- and Post-Smoking Ban for All Residential Blocks.

	Regression Coefficient (B)	SE	Standardized Coefficients (B)
Bar block pre-ban	−.952***	.217	−.100
Bar block change	1.027*	.481	.048
Percentage Hispanic	−.007	.004	−.032
Percentage African American	.002	.001	.037
Single-headed households	−.004	.004	−.024
Percentage alone	−.004*	.002	−.047
Percentage youth	−.004	.003	−.037
Percentage vacant	.000	.004	−.003
Constant	.056	.073
R ²	.02

Note. OLS = ordinary least squares. *p < .05. **p < .01. ***p < .001.

To further explore the change in the spatial association between the presence of one or more bars on blocks and the change in frequency of reported assaults before and after clean indoor air law was implemented, we calculated monthly rates of offending. Rather than simply displaying the number of reported offenses, rates are used as the number of bar and nonbar blocks changed over the duration of the data collection period. As can be seen in Table 4, blocks with bars had higher rates of reported offenses both before and after the law was implemented. Also of note, while there was a slight increase in the rate of reported offenses on nonbar blocks, a substantial drop among bar blocks occurred. Indeed, the monthly rate of reported offenses per block declined by over 20% in the 12 months following the smoking ban when compared against the 12 months prior to implementation.

Table 4.

Average Rates of Reported Offenses on Bar Blocks and Nonbar Blocks.

Block Type	Average Rate Reported Offenses Per Block Per Month 12 Months Pre-ban (×100)	Average Rate Reported Offenses per Block Per Month 12 Months Post-ban (×100)	Percentage Change in Average Rate Reported Offenses Per Block 12 Months Pre- and Post-ban (%)
Bar blocks	6.37	4.98	−21.80
Nonbar blocks	2.94	3.01	2.40

Given the seasonal nature of offending, one potential explanation for the findings presented previously could be routine variation (Rotton & Cohn, 2004)⁹. Figure 1 presents the rates of reported offenses by block type or the total count of reported offenses divided by the number of blocks with each category, specifically bar blocks including adjacent blocks for capturing displaced offenses and nonbar blocks, for each month using 3-month moving averages. While seasonal variation is present, it is also evident that the peak in reported offending the year before the law was implemented was far higher than in either of the following years.

Figure 1.

Rate of reported assaults per Block by Block Type.

Finally, in order to examine whether offending displaced to adjacent blocks, first-order nearest blocks were identified around each establishment. This buffer included all blocks adjoining the blocks neighboring bar blocks. Results from negative binomial regression models indicate no significant displacement. Likewise, results from an OLS regression model indicate no significant change in the number of reported assaults on the buffer blocks.¹⁰

Discussion

During the previous decade, many states, counties, and municipalities within the United States have enacted statutes and ordinances restricting smoking within bars, taverns, restaurants, and other public facilities in order to improve public health. Given the magnitude of people harmed either by secondhand smoke or by directly smoking tobacco products themselves, seeking to reduce smoking is a laudable goal. However, policy changes impacting entire categories of facilities often produce unanticipated consequences (Merton, 1936).

Laws prohibiting smoking within bars and taverns may also alter the social functioning both within and around the impacted establishments, resulting in changing the ecological structure to criminal opportunities (Cohen & Felson, 1979). Previous research conducted in times and locations where consumption of tobacco products within recreational alcohol-serving establishments was less restricted finds higher frequencies of assaults clustering around bars and taverns (Roncek & Bell, 1981; Roncek & Maier, 1991). A discussion has arisen along with these findings regarding recommended policy solutions, specifically whether policies should be applied across entire categories of facilities (Kubrin et al., 2011; Wilcox & Eck, 2011). Using calls for service data from a pair of adjoining communities, the current study examines whether the frequency of reported assaults around bars changes following the implementation of a law prohibiting smoking within bars and taverns.

This study finds the spatial association between bars and assaults was diminished following a smoke-free law taking affect. While models continue to show a relationship between the presence of one or more bars on a block and the frequency of assaults, this relationship was diminished once the law prohibiting smoking in such facilities went into effect. This suggests a significant change in the criminal opportunity structure in and around bars and taverns may have occurred following the implementation of the clean indoor air law.

Based on the expectations derived from routine activities theory, there are several explanations for the findings from the current study beginning with motivated offenders. While the current study was not able to establish whether the composition of bars and taverns changed or whether population of bar customers changed, perhaps, the changes to facility functioning disproportionately drove more problem-prone establishments out of business. Likewise, the composition of bar customers could have changed to attract fewer motivated offenders. This area of research would benefit from future studies using direct observations of the number of patrons in and around bars and taverns both before and after a smoke-free law is impacted to more precisely determine the level of risk around this type of establishment.

The number or quality of suitable targets may have also changed. Perhaps the clean indoor air act reduced the degree of intoxication among bar patrons, reducing opportunities for victimization. While the current study did not have measures of risk at individual establishments, other research has found that smoking bans may increase total bar revenue as well as higher liquor sales revenue (Collins et al., 2010). Indeed, retail sales tax figures from the county encompassing both communities in the current study show increases in sales at bars and restaurants between 2007 and 2009, which suggests that the degree of intoxication may not have declined at bars and taverns but rather may have increased (Department of Revenue, 2007, 2009). Until measures of the number patrons or the distribution of alcohol consumption among those patrons become available both before and after a smoking ban is implemented, researchers will remain unable to determine whether the availability of targets produced the change observed in the current study.

As noted previously, patrons wishing to smoke tobacco products in locations with clean indoor air laws must often do so outside the impacted institutions, and they may increase guardianship around the facilities (Moore et al., 2009). Perhaps following the implementation of the clean indoor law in the current study, a greater number of patrons began to gather outside the impacted establishments.

While the findings from the current study provide an important contribution to the research examining the relationship between types of facilities and crime together with the impact of policies altering facility functioning, these findings must be considered in light of several limitations. Direct measures of risk at individual establishments, such as the number of patrons at individual establishments before and after the smoke-free law or the level of intoxication among patrons, were not available. Similarly, the current study does not use direct measures of patron behavior in and around impacted establishments. In addition, the findings from the current study may have been the result of changes in reporting behaviors by victims and witnesses or recording behaviors by dispatchers. Finally, the current study does not account for changes in police behavior.¹¹ However, informal inquiries to the police departments serving these communities yield no formal process or organization for supporting hot spot or problem-oriented policing though individual officers may take steps to address the most problem-prone businesses on an informal basis. In light of these limitations, the findings from the current study could be used to cautiously inform policy makers when deliberating increasing restrictions on smoking tobacco products within recreational alcohol-serving establishments.

When considering enacting laws restricting smoking within and around bars and taverns, policy makers are often confronted with arguments focusing on impacts on public health and financial success of affected establishments. The findings from the current study suggest a new dimension, diminished criminal opportunities, maybe warranted in these deliberations. However, these findings should be interpreted with caution at this time. The findings in the current study originate from a pair of adjoining smaller cities in the Midwest. Replications of the current study across varying contexts could bolster these findings, further warranting discussions by policy makers of this benefit when considering such restrictions.

Footnotes

Acknowledgment

The authors give their special thanks to Bernard Conrad, Tesfay Russell, and Tiffany Koss for their time and effort and Tony Thompson for making this study possible. Also, a special thank-you to Dennis Roncek for his enthusiasm and dedication to mentoring students.

Authors’ Note

Opinions expressed within this article are solely those of the authors.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for this study was provided by a Project Grant from the College of Social and Behavioral Sciences of the University of Northern Iowa.

Notes

References

Adhikari

Kahende

Malarcher

Pechacek

Tong

(2008). Smoking-attributable mortality, years of potential life lost, and productivity losses—United States, 2000—2004. Morbidity and Mortality Weekly Report, 57, 1226–1228.

Anselin

(2004). GeoDa 0.9.5-i Release Notes. Urbana: Center for Spatially Integrated Social Science, University of Illinois, Urbana-Champaign.

Anselin

Bera

(1998). Spatial dependence in linear regression models with an introduction to spatial econometrics. In Ullah

Giles

D. E.

(Eds.), Handbook of applied economic statistics (pp. 237–289). New York, NY: Marcel Dekker.

Anselin

Cohen

Cook

Gorr

Tita

(2000). Spatial analysis of crime. In Duffee

(Ed.), Measurement and analysis of crime and justice (pp. 213–262). Rockville, MD: National Institute of Justice.

Block

R. L.

Block

C. R.

(1995). Space, place and crime: Hot spots areas and hot places of liquor-related crime. In Eck

Weisburd

(Eds.), Crime and place (pp. 145–183). Monsey, NY: Criminal Justice Press.

Brantingham

(1995). Criminality of place: Crime generators and crime attractors. European Journal on Criminal Policy and Research, 3, 1–26.

Briggs

Opsal

(2012). The influence of victim ethnicity on arrest in violent crimes. Criminal Justice Studies: A Critical Journal of Crime, Law and Society, 25, 177–190.

Britt

Carlin

Toomey

Wagenaar

(2005). Neighborhood level spatial analysis of the relationship between alcohol outlet density and criminal violence. Environmental and Ecological Statistics, 12, 411–426.

Cohen

Felson

(1979). Social change and crime rate trends. American Sociological Review, 44, 588–608.

10.

Collins

N. M.

Shi

Forster

J. L.

Erickson

D. J.

Toomey

T. L.

(2010). Effects of clean indoor air laws on bar and restaurant revenue in Minnesota cities. American Journal of Preventive Medicine, 39, S10–S15.

11.

Cowling

D. W.

Bond

(2005). Smoke-free laws and bar revenues in California-the last call. Health Economics, 14, 1273–1281.

12.

Department of Revenue, Tax Research and Fiscal Analysis Section. (2007). Iowa Retail Sales & Use Tax Report. Des Moines: State of Iowa.

13.

Department of Revenue, Tax Research and Fiscal Analysis Section. (2009). Iowa Retail Sales & Use Tax Report. Des Moines: State of Iowa.

14.

Eck

J. E.

Clarke

R. V.

Guerette

R. T.

(2007). Risky facilities: Crime concentrations in homogeneous sets of establishments and facilities. Crime Prevention Studies, 21, 225–264.

15.

Federal Bureau of Investigation. (2004). Uniform crime reporting handbook. Washington, DC: U.S. Department of Justice.

16.

Federal Bureau of Investigation. (2005). Crime in the United States, 2004. Washington, DC: U.S. Department of Justice.

17.

Federal Bureau of Investigation. (2008). Crime in the United States, 2007. Washington, DC: U.S. Department of Justice.

18.

Federal Bureau of Investigation. (2009). Crime in the United States, 2008. Washington, DC: U.S. Department of Justice.

19.

Federal Bureau of Investigation. (2010). Crime in the United States, 2009. Washington, DC: U.S. Department of Justice.

20.

Federal Bureau of Investigation. (2011). Crime in the United States, 2010. Washington, DC: U.S. Department of Justice.

21.

Felson

(1995). Those who discourage crime. In Eck

J. E.

Weisburd

(Eds.), Crime and place (pp. 53–66). Monsey, NY: Criminal Justice Press.

22.

Felson

R. B.

Burchfield

K. B.

(2004). Alcohol and the risk of physical and sexual assault victimization. Criminology, 42, 837–859.

23.

Felson

R. B.

Staff

(2010). The effects of alcohol intoxication on violent versus other offending. Criminal Justice and Behavior, 37, 1343–1360.

24.

Frisbie

D. W.

Fishbine

Hintz

Joelson

Nutter

J. B.

(1977). Crime in Minneapolis: Proposals for prevention. St. Paul, MN: Community Crime Prevention Project, Governor’s Commission on Crime Prevention and Control.

25.

Gorman

Speer

Gruenwald

Labouvie

(2001). Spatial dynamics of alcohol availability, neighborhood structure and violent crime. Journal of Studies on Alcohol, 62, 628–636.

26.

Gruenwald

P. J.

Freisthler

Remer

LaScala

E. A.

Treno

(2006). Ecological models of alcohol outlets and violent assaults: Crime potentials and geospatial analysis. Addiction, 101, 666–677.

27.

Heinz

A. J.

Beck

Meyer-Lindenberg

Sterzer

Heinz

(2011). Cognitive and neurobiological mechanisms of alcohol-related aggression. Nature Reviews: Neuroscience, 12, 400–413.

28.

Hindelang

M. J.

Gottfredson

M. R.

Garofalo

(1978). Victims of personal crime: An empirical foundation for a theory of personal victimization. Cambridge, MA: Ballinger.

29.

Homel

Clark

(1995). The prediction and prevention of violence in pubs and clubs. In Clark

(Ed.), Crime prevention studies (pp. 1–46). Monsey, NY: Criminal Justice Press.

30.

Homel

Tomsen

Thommeny

(1992). Public drinking and violence: Not just an alcohol problem. Journal of Drug Issues, 22, 679–697.

31.

Jones

Holl

(2006, April 14). New Jersey joins 10 states in banning indoor smoking. New York Times, p. B6.

32.

Klinger

D. A.

(1997). Negotiating order in patrol work: An ecological theory of police response to deviance. Criminology, 35, 277–306.

33.

Kubrin

C. E.

Squires

G. D.

Graves

S. M.

Ousey

G. C.

(2011). Does fringe banking exacerbate neighborhood crime rates? Criminology & Public Policy, 10, 435–466.

34.

Livingston

(2008). Alcohol outlet density and assault: A spatial analysis. Addiction, 103, 619–628.

35.

McClelland

G. M.

Teplin

L. A.

(2001). Alcohol intoxication and violent crime: Implications for public health policy. The American Journal on Addictions, 10, 70–85.

36.

Merton

(1936). The unanticipated consequences of purposive social action. American Sociological Review, 1, 894–904.

37.

Moore

Annechino

Lee

(2009). Unintended consequences of smoke-free bar policies for low-SES women in three California counties. American Journal of Preventive Medicine, 37, 138–143.

38.

Murray

R. K.

Roncek

D. W.

(2008). Measuring diffusion of assaults around bars through radius and adjacency techniques. Criminal Justice Review, 33, 199–220.

39.

Nielsen

Martinez

(2003). Reassessing the alcohol-violence linkage: Results from a multiethnic city. Justice Quarterly, 20, 445–469.

40.

Peterson

Krivo

Harris

(2000). Disadvantage and neighborhood violent crime: Do local institutions matter? Journal of Research in Crime and Delinquency, 37, 31–63.

41.

Polk City Directories. (2007a). Cedar Falls, IA. Livonia, MI: Polk City Directories.

42.

Polk City Directory. (2007b). Waterloo, IA. Livonia, MI: Polk City Directories.

43.

Polk City Directory. (2009a). Cedar Falls, IA. Livonia, MI: Polk City Directories.

44.

Polk City Directory. (2009b). Waterloo, IA. Livonia, MI: Polk City Directories.

45.

Rand

Robinson

(2011). Criminal victimization in the United States, 2008—Statistical tables. Washington, DC: U.S. Department of Justice, Office of Justice Programs.

46.

Roncek

Bell

(1981). Bars, blocks, and crimes. Journal of Environmental Systems, 11, 35–47.

47.

Roncek

Faggiani

(1985). High schools and crime. The Sociological Quarterly, 26, 491–505.

48.

Roncek

Maier

(1991). Bars, blocks, and crimes revisited: Linking the theory of routine activities to the empiricism of “hot spots.” Criminology, 29, 725–753.

49.

Room

(2004). Smoking and drinking as complementary behaviours. Biomedicine and Pharmacotherapy, 58, 111–115.

50.

Rotton

Cohn

E. G.

(2004). Outdoor temperature, climate control, and criminal assault: The spatial and temporal ecology of violence. Environment & Behavior, 36, 276–306.

51.

Sampson

R. J.

Raudenbush

S. W.

Earls

(1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918–924.

52.

Scollo

Lal

Hyland

Glantz

(2003). Review of the quality of studies on the economic effects of smoke-free policies on the hospitality industry. Tobacco Control, 12, 13–20.

53.

Shaw

C. R.

McKay

H. D.

(1942). Juvenile delinquency in urban areas. Chicago, IL: University of Chicago Press.

54.

Sherman

Gartin

Buerger

(1989). Hot spots of predatory crime: Routine activities and the criminology of place. Criminology, 27, 27–55.

55.

Smirnov

Anselin

(2001). Fast maximum likelihood estimation of very large spatial autoregressive models: A characteristic polynomial approach. Computational Statistics and Data Analysis, 35, 301–319.

56.

Smokefree Air Act, Iowa Code §142D. (2008).

57.

Speer

Gorman

Labouvie

Ontkush

(1998). Violent crime and alcohol availability: Relationships in an urban community. Journal of Public Health Policy, 19, 303–318.

58.

Townsley

(2009). Spatial aurocorrelation and impacts on criminology. Geographical Analysis, 41, 452–461.

59.

Truman

(2011). Criminal victimization, 2010. Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice.

60.

Tynan

Babb

MacNeil

Griffin

(2011). State smoke-free laws for worksites, restaurants, and bars—United States, 2000–2010. Morbidity and Mortality Weekly Report, 60, 472–475.

61.

U.S. Census Bureau. (2011). State and county quickfacts. Retrieved December 11, 2011, from http://quickfacts.census.gov/qfd/states/19000.html

62.

Varano

Schafer

Cancino

Swatt

(2009). Constructing crime: Neighborhood characteristics and police recording behavior. Journal of Criminal Justice, 37, 553–563.

63.

Weisburd

Hinkle

Famega

Ready

(2011). The possible “backfire” effects of hot spots policing: An experimental assessment of impacts on legitimacy, fear and collective efficacy. Journal of Experimental Criminology, 7, 297–320.

64.

Wilcox

Eck

J. E.

(2011). Criminology of the unpopular: Implications for policy aimed at payday lending facilities. Criminology & Public Policy, 10, 473–482.