Abstract
The purpose of this study was to assess citizens’ perceptions of police performance in a small, rural community that is experiencing rapid population growth. This unique setting differs from other cities and towns because of the pressure that it places on local police to respond quickly and effectively to rapid social changes. A multi-stage random sampling technique was used to gather 380 surveys completed by residents living in Williston, North Dakota during the fall of 2015. The findings revealed that, in general, residents believed that the police were doing a good job addressing community concerns; however, the analysis also revealed that such views were impacted by previous victimization experiences and fear of victimization. In addition, the results indicated that participant sex was significantly related to the perceived ability of police to maintain order on the streets, whereas marital status and the frequency of reading the regional newspaper were significantly related to the perceived ability of police to control crime in the neighborhood.
Perceptions of police performance in a rapid-growth community
In the United States, the police perform a variety of functions (Bittner, 1970; Kelling and Moore, 1988; S Walker, 1977). Traditionally, police are expected to control crime, maintain order, enforce the law, and provide service to the community (Bittner, 1970; Trojanowicz and Bucqueroux, 1990; S Walker, 1977). Although this list can vary and/or encompass other responsibilities with shifts in the policing paradigm, these core functions continue to be regarded with the utmost importance by policing officials (Ortiz et al., 2007; Zhao et al., 2003).
To determine whether the police are meeting expectations, researchers have utilized surveys to assess citizens’ perceptions of police performance. However, many of these studies have been conducted in large, urban settings; thereby, telling us little about this topic in the context of smaller communities (Fyfe, 2010; Pelfrey, 2007; Prine et al., 2001). It is unfortunate that more research is not available on non-urban places given that approximately half of police departments in the United States employ only 10 or fewer sworn police officers (Banks et al., 2016).
There are clear differences between police departments in large cities and departments located in small, rural towns. For instance, urban police agencies have greater access to financial support compared with rural police departments, which are often underfunded (Ashcroft et al., 2004; Falcone et al., 2002; Oliver and Meier, 2004). This means that urban agencies are more likely than small agencies to have access to new technology, equipment, computers, vehicles, and cameras for improving their efficiency and outcomes (Ashcroft et al., 2004; Falcone et al., 2002). In short, the paucity of research on police in non-urban spaces suggests more work is needed to determine the perceived functions of police in other types and sizes of communities.
In addition, one particular type of setting that is largely missing from the literature is communities that experience rapid population growth. Whether the result of an oil boom, a flourishing job market, or net migration (more people entering the city than leaving), rapidly growing communities are unique because of the ongoing shift in the town's social and physical dynamics (Stebbins, 2019). Heavier traffic flow, lack of affordable housing, and, in some cases, limited access to healthcare and education are a few examples of the challenges faced by residents living in these types of communities (Archbold, 2015; Borchert, 2020).
Within rapid-growth communities, town residents must adjust to sudden changes. Simultaneously, local police officers are forced to respond to sudden rises in crime and disorder (Archbold, 2015). As such, local police are placed under extreme pressure to perform and with limited access to resources. This raises the question: How do citizens perceive the ability of police to carry out their functions in rapidly growing communities? The current study responds to this question using survey data from a small, rural town in western North Dakota that grew in population as a result of increased energy development.
What do we know about perceptions of police performance in small, rural communities?
Individual factors
To date, only a handful of studies have examined the perceived performance of police in small, rural areas and these studies have generally included individual factors such as citizen age, sex, race/ethnicity, marital status, home ownership, and education. For instance, Benedict et al. (2000) found both men and women living in a small town reported the police were attentive to problems occurring in their community; however, women were more likely than men to report being concerned about their personal safety. Both men and women also believed that the police could have done a better job interacting with members of the community in more positive ways.
In a survey of residents living in 14 small towns, Nofziger and Williams (2005) found that younger individuals, men, African Americans, and renters expressed lower levels of confidence in their local police when compared with residents who were older, female, White, and homeowners. When considering only those who had previously experienced a police contact, only race was significant; more specifically, levels of confidence were reduced for African American residents when compared with all other races, an outcome that is consistent with findings from research in large, urban communities on perceptions of the police (Lai and Zhao, 2010; Reisig and Parks, 2000; Weitzer and Tuch, 2005; Wu, 2014).
In a more recent study conducted in a small town in Iowa, older, married residents rated police protection less favorably than individuals who were younger and unmarried (Holmes et al., 2017). However, older, White residents reported having greater trust in their local police than residents who were younger and not White. Research regarding education has been mixed with some studies reporting a positive relationship between levels of education and positive views of police responsiveness to neighborhood concerns (Wentz and Schmligen, 2012), whereas others failed to find any such relationship (Holmes et al., 2017).
An additional factor worth including is length of residency. This variable was found to be significant in the small body of research centered on fear of crime in communities experiencing rapid population growth. For example, some research on citizens’ fear of crime victimization suggests that fear levels tend to be higher for long-time residents compared with residents who have less residential longevity (Cao et al., 1996; Eschholz et al., 2002; Hunter et al., 2002; Mrozla et al., 2018). This difference could translate into long-time residents viewing police performance differently from residents who are relatively new to the area.
Media exposure
It has become especially important to examine the role of media given recent high-profile incidents involving the police. For instance, there has been extensive coverage surrounding the death of George Floyd while in custody of several Minneapolis police officers, sparking outrage from citizens from across the country. This tragic incident is in addition to others, including the death of Breanna Taylor, all of which have raised numerous questions about the role and functions of police and whether they have been meeting their obligations to protect and serve the American people. Unfortunately, there is a paucity of research on small and rural towns regarding the impact of media on ratings of police performance. Nevertheless, information can be gathered from studies that examined urban/mixed communities.
In an early study, Weitzer (2002) found that publicized incidents of police misconduct generated unfavorable attitudes toward the police and police performances in two major US cities—New York and Los Angeles. More recently, a systematic review of 42 studies by Graziano (2019) revealed that negative perceptions of police were related to both exposure to high-profile incidents involving the police and consumers’ awareness of negative media coverage of police.
It is unclear what role media play in residents’ perceptions of police performance in non-urban settings. Mrozla et al. (2018) surveyed residents from a small but growing town in western North Dakota and found no relationship between the amount of time spent watching local television news and residents’ fear of crime. Lytle et al. (2020) also examined fear of crime but among residents living in a semi-rural community and found media consumption was less important than other factors such as gender and perceived physical disorder. These findings are surprising given past studies have shown fear of crime is an important predictor of citizens’ perceptions of police effectiveness (Hawdon et al., 2003; Nofziger and Williams, 2005; O’Connor, 2008).
The current study includes the examination of two media sources—the local television news and regional newspaper. Although social media and other online news outlets can provide information about police, residents in small towns rely more heavily on local news sources for details about recent changes occurring in their community (Miller et al., 2012).
Crime victimization and fear of victimization
Prior crime victimization or fear of becoming a victim of crime have been found to be important predictors of perceptions of police services in small cities and towns. Hawdon et al. (2003) surveyed residents from 41 neighborhoods in western South Carolina and found that crime victims had more negative perceptions of police effectiveness. However, in Nofziger and Williams’s (2005) examination of 14 small towns, crime victimization did not play a role in the perceived ability of local police to control crime and respond to major crises.
The effects of crime victimization on the perceived performance of police in rapidly growing communities has yet to be determined. Nevertheless, there is some evidence to suggest that crime victimization matters within this context. In their study of a growing small town, Mrozla et al. (2018) found that crime victims expressed greater fear of crime. Although this study did not examine residents’ perceptions of police effectiveness, past research has indicated that a relationship exists between these two variables (Hawdon et al., 2003; Nofziger and Williams, 2005; O’Connor, 2008).
Small-town residents who are fearful of becoming victims of a crime or who perceive crime to be a problem in their neighborhood have also been shown to hold less-favorable views of police functions (Nofziger and Williams, 2005; O’Connor, 2008; Prine et al., 2001; Reisig and Parks, 2000). Hawdon et al. (2003) reported lowered fear of criminal victimization was associated with more favorable views of police effectiveness among residents from larger and smaller communities. Nofziger and Williams (2005) also reported that small-town residents who perceived less crime in their community expressed greater confidence in the ability of their local police to control crime. Finally, in a study of residents living in one Midwestern city with a population of fewer than 150,000, Wentz and Schlimgen (2012) reported that residents who feared for their safety gave lower ratings of police services, further suggesting the importance of fear of victimization when examining perceptions of police performance.
Police functions in rapidly growing communities
Rapid increases in population can result in significant changes to the social and physical dynamics of a community. For instance, Midland, Texas recently experienced mass influxes of new residents as people sought employment in the oil fields. In Collin County, Texas, the oil boom has led to the town doubling in size within the past two decades and it was estimated that its population in 2,050 will reach 2.4 million, a figure that is three times its size in 2010. In addition to such growth and at a rate that far exceeds what the town's infrastructure can support, access to basic services like healthcare, housing, and education have become major challenges for residents living within these areas (Crutchfield et al., 2020).
The current study examines perceptions of police performance but in a much smaller community that also experienced a surge in population as a result of increased oil production. The population of Williston, North Dakota increased by approximately 50% within a five-year period from 2010 to 2015 (United States Census Bureau, 2015). As the town grew, so did the demands for police services. Archbold (2015) interviewed 101 patrol officers and sheriff's deputies from eight police agencies located in western North Dakota and found that these communities have become more physically complex with the addition of newly built apartments and homes, as well as heavier traffic flow, and growing businesses.
Police organizations and their method of operation were also severely impacted as police personnel expressed concerns over the lack of resources for addressing the increased demand. Many patrol officers and sheriff's deputies in this study reported the challenge of having to attend to the large volume of citizens’ calls for service with inadequate resources. Some of the police agencies were able to hire a few more police officers when the population growth began; however, it was not enough to address the growing demands of the community. Consequently, the sheer amount of work not only translated into fewer interactions with residents, but also transformed the nature of their interactions (Archbold, 2015).
Having knowledge of the local police's perspective on their own ability to perform during this demanding period is important because it could be indicative of residents’ views. Under such circumstances, it is possible that residents may begin to question the efficacy of their local police and whether they are equipped to perform their primary functions successfully amid drastic changes to the community. As such, the following research question is addressed: Which factors are related to residents’ perceptions of police performance in a community experiencing rapid population growth?
Method
Research site
Data for this study were drawn from a larger project on residents’ perceptions of the changes that occurred in their community during an oil boom that began in 2008. The site of interest—Williston, North Dakota—is in the western region of the state and is situated within the Bakken shale region where mass production of crude oil was taking place. Associated with this high volume of oil extraction was the creation of jobs within the oil fields, along with various other employment opportunities in the service industry. As a result, Williston became one of several counties to experience rapid population growth as people from all over the nation flooded the state in search of work (United States Census Bureau, 2015). For instance, in 2010 Williston had an estimated 14,716 residents. Contrast this number with the 22,051 residents recorded in 2015. In fact, Williston was identified as one of the fastest growing micropolitan areas in the United States for the period between 2014 and 2015. Williston experienced a population increase of about 9.9% during that same one-year period (United States Census Bureau, 2015; Williston Economic Development, 2016). Prior to this time, the growth rate was only 4.7% from 2000 to 2010 (Mackun and Wilson, 2011).
Changes to the demographic profile of Williston can also be observed (Table 1). The median age of residents declined from 35.5 years in 2010 to 31.7 years in 2015. In 2010, 93.35% of the population were White, and in 2015, this dropped to 86.97%. The change to the gender ratio is also apparent with women making up 48.99% of the population in 2010 and 45.51% in 2015, a decline that is likely due to the large number of men who moved to the area to find work in the oil fields. The number of housing units occupied by renters increased from 2325 (37.62%) in 2010 to 3549 (38.99%) in 2015. Finally, the median household income increased from $56,836 in 2010 to $90,171 in 2015 (United States Census Bureau, 2015).
Williston demographic characteristics, 2000, 2010, 2015.
Data source
A multi-stage random sampling technique was used to administer door-to-door surveys to households in Williston during fall 2015. Using a map of Williston that was provided by the planning and zoning department, the city was first divided into four quadrants, then neighborhoods were identified within each quadrant. A research team of five distributed surveys to residents based on randomly chosen city blocks. Quadrant 1 (located on the northern part of the city) consisted primarily of industrial and commercial businesses, although there were few areas dedicated to single and multi-family households. Quadrant 2 is considered the central part of the city, and is comprised of several types of residential dwellings (e.g. single family homes, multi-family households, and mobile parks). The local airport is also located within this section of the city. On the southeastern region of Williston is Quadrant 3. This quadrant is not dominated by any particular type of structure; instead, it contains a mix of industrial, commercial, and residential areas. The last quadrant (Quadrant 4) is in the southwestern part of town and consists of largely agricultural areas along with several industrial businesses.
Team members approached each household by first introducing themselves and then explained the purpose of the study. Household members had to be at least 18 years of age to complete the survey. Whenever an adult was unavailable, team members would revisit that household. On occasions, a copy of the survey was left with respondents to be completed at their convenience, and then retrieved by the research team later. A total of 75 surveys were collected from Quadrant 1, 137 from Quadrant 2, 100 from Quadrant 3, and 68 from Quadrant 4. The larger number of surveys obtained from the second and third quadrants correspond to the relatively large number of residential homes located within these two regions. By contrast, Quadrants 1 and 4 comprised mostly agricultural land or industrial areas.
According to the United States Census Bureau (2015), there were 9120 households in 2015. The research team visited 1,823 households, of which 1,288 had residents who either chose not to answer the door or were not home to answer the door. The team undertook 535 face-to-face interactions with household residents with 155 people refusing to complete the survey. This resulted in a total of 380 completed surveys. Thus, the response rate for face-to-face interactions was 71% (380/535). The overall response rate was 20.8% (380 completed surveys after knocking on the doors of 1,823 households).
Variables
Dependent variables
Although police are known to perform a variety of functions, controlling crime and maintaining order are two of their main functions, and are examined in the current study. First, two items were combined to form a perception of order maintenance variable: “How good a job are the police doing in maintaining order on the streets and sidewalks in your neighborhood?” and “How good a job are the police doing in controlling traffic-related issues in your neighborhood?”. Responses ranged from 1 = very bad to 4 = very good. Cronbach's alpha reliability coefficient revealed an acceptable degree of internal consistency (α = .72).
The second dependent variable measures perceptions of crime control by the police. This variable was created by combining two survey questions: “How good a job are the police doing in controlling violent crime in your neighborhood?” and “How good a job are the police doing in controlling property crime in your neighborhood?”. Responses ranged from 1 = very bad to 4 = very good. Cronbach's alpha suggests a high degree of internal consistency (α = .84). The rise in both violent and property crime was believed to be associated with increases in the town's population (Komer, 2019).
Citizen characteristics
The literature has uncovered several individual characteristics that have been included in previous studies on perceptions of police performance. Studies generally find that older people are more likely than younger individuals to view the performance of police in a favorable way (Brown and Benedict, 2002; Hurst et al., 2000; Schafer et al., 2003; Wu, 2014). Age was coded as a continuous variable measured in years. Research findings on the influence of citizens’ sex has been mixed. Some studies find that females hold more positive views of police service than males do (Schafer et al., 2003), whereas others find the opposite to be true (Schuck et al., 2008; Wu, 2014). In the current study, participant sex was coded “0” for males and “1” for female. For all dichotomous variables, those coded as “1” were treated as the reference category.
Research examining racial differences in the perception of police functions has consistently reported that African Americans are least likely to hold positive views (Cochran and Warren, 2012; Weitzer and Tuch, 2005; Wu, 2014). For race/ethnicity, White residents were coded as “0”, and all other races were coded as “1”. Although it is desirable to have more than two racial and ethnic categories, the small number resulting from a lack of diversity within our sample makes it challenging to make appropriate comparisons. Likewise, the small numbers associated with the different categories of marital status resulted in residents being grouped into two categories where single, widowed, or divorced individuals were included in one category (coded “0”) and married individuals were included in the other (coded “1”). Some studies observed marital status to have an impact on citizens’ perception of police responsiveness (Holmes et al., 2017), whereas others did not (Schafer et al., 2003).
The literature on education is also mixed. Holmes et al. (2017) found no relationship between levels of education and perceptions of police services, whereas others have reported this relationship exists (Wentz and Schlimgen, 2012; Wu, 2014). Education level was coded into five groups (no high school diploma = 0, high school diploma = 1, some college = 2, two years of college = 3, and four or more years of college = 4). Regarding home ownership, homeowners have been reported to have higher confidence in the ability of police to address crime than do renters (Nofziger and Williams, 2005). For this study, renters were coded “0”, whereas homeowners were coded “1”.
Finally, in the context of rapidly growing communities, length of residency may play a role in residents’ perceptions of police. Studies have shown that fear of victimization tends to be higher for long-time residents compared with residents with less residential longevity (Hunter et al., 2002), which could translate into divergent views of police functions. Residential longevity was separated into two categories—long-time and boom residents. Long-time residents refer to those who have been living in Williston prior to the rapid influx of residents resulting from the oil boom (all prior years up to 2007), whereas boom residents refer to those who arrived during the energy boom (2008–2015) (0 = boom residents, 1 = long-time residents). Although boom residents may not necessarily be new to the area, especially if they lived in Williston for the entire period between 2008 and 2015, their views of the town could still differ from those held by residents who were familiar with the town prior to its growth.
Media consumption
There is some evidence to indicate that the amount of media consumption can shape citizens’ views of police functions (Callanan and Rosenberger, 2011; Dowler and Zawilksi, 2007; Graziano, 2019; Miethe et al., 2019). Here, both the amount and source of information were included in the analyses. More specifically, respondents were asked the following two questions: “How many times per week do you watch the local news on television?” and “How many times per week do you read the local or regional newspaper?”. These media variables are continuous in nature.
Crime victimization and fear of victimization
Past research has found a relationship between prior criminal victimization and lower levels of satisfaction with police services (Hawdon et al., 2003; Maxson et al., 2003; Wu, 2014). To determine whether prior crime victimization plays a role in the current setting, respondents were asked to indicate “no” (0) or “yes” (1) to the following two questions: “Have you, a family member, or a close friend been the victim of a property crime in the past two years?” and “Have you, a family member, or a close friend been the victim of a violent crime in the past two years?”.
To determine the level of fear that citizens have toward becoming a victim of crime, respondents were asked, “How fearful are you of becoming a victim of violent crime?” and “How fearful are you of becoming a victim of property crime?” (α = .78). Respondents indicated their level of fear on a scale of 1 = not at all fearful to 4 = very fearful. This general fear of crime measure is important for understanding citizens’ assessment of their surroundings and the police (Bridenball and Jesilow, 2005; Weitzer and Tuch, 2005).
Analytical strategy
Descriptive statistics were generated to create a snapshot of the frequencies and percentages associated with all variables included in this study. Multivariate analyses were utilized to examine the relationship between the independent variables on the outcome measures. Prior to any statistical analyses, multi-collinearity diagnostics were performed and revealed all tolerance values were >0.25 and all variance inflation factor scores were <4, indicating that there were no problems with multi-collinearity in the current study (JT Walker and Maddan, 2013). Because of the ordered nature of the dependent variables, a series of ordinal regression analyses were conducted for the two outcome measures (violent and property) (JT Walker and Maddan, 2013). Model 1 and Model 2 examine the impact of the predictor variables on the perceived ability of police to address order maintenance issues and the perceived ability of police to control neighborhood crime (violent and property). In the third model (Model 3), both dependent variables were combined to form an overall perception of police functions variable. Cronbach's alpha reliability coefficient for this combined variable revealed high internal consistency (α = .84).
Findings
Descriptive statistics for the sample used in this study are included in Table 2. The average age of survey respondents was 44.51 years and more than half (58.73%) were female. Most survey respondents were White (87.78%), married (60.75%), and homeowners (69.89%). On average, residents in the sample had lived in Williston for about 19.16 years. More than half (51.88%) of the residents lived in Williston prior to the oil boom. Most residents who completed the survey (81.07%) had at least a high-school education. Of those residents, 22.70% earned some college credits, 19.08% completed two years of college, and 30.59% had four or more years of college experience.
Sample descriptive statistics.
The sample used in this study is similar to the population demographic of Williston, North Dakota regarding the distribution of race/ethnicity, homeownership, and education. For instance, there were significantly more White residents than all other races/ethnicities and more homes were owned than rented. However, the study sample deviates from Williston's population demographic regarding sex and age. There were more males than females living in Williston in 2015; however, the sample in the current study contains more women than men. One possibility for this gender imbalance is the time of day that the questionnaires were administered. The research team visited households from 09:00 until approximately 20:00. Survey distribution occurred during a period when people are likely to be at work. Moreover, it has been said that the increased employment opportunities in the oil fields have attracted mostly men (Hargreaves, 2015; Valentine, 2014). Thus, it is likely that a larger number of men than women were at their place of employment during the team's visits to households, resulting in more women than men completing the survey.
The median age recorded for the population of Williston (31.7 years) was younger than the median age in our study sample of 43 years. Similar to participant sex, it is possible that young males were at work in the oil fields during the hours of survey distribution.
On average, Williston residents read the local or regional newspapers approximately three times per week and watched the local news on television four times per week. Fewer than half of respondents (44%) have been victims of property crime and even fewer respondents have been victims of violent crime (16.5%). Approximately 20.6% of respondents reported being fearful or very fearful of becoming victims of violent and property crimes. Overall, residents perceived police functions positively. More specifically, 80.8% of respondents indicated that the police were doing a good job controlling traffic and maintaining order on the streets and sidewalks and 85.4% reported that the police were doing a good job combating property and violent crimes in their neighborhood.
Table 3 contains the results of three ordinal regression models. The first model revealed that men were less likely than women to agree that the police were doing a good job maintaining order. This model also indicates that residents who have not been victims of violent or property crimes within the past two years were more likely to report that the police were doing a good job at maintaining order when compared with people who have been victimized. More specifically, the odds of agreeing that the police were doing a good job were 2.43 times higher for people who were not victims of violent crimes and 1.79 times higher for people who were not victims of property crimes compared with individuals who have been victimized. Finally, individuals who reported being more fearful of becoming victims of crime were less likely to say that the police were doing a good job maintaining order in their community.
Ordinal regression models: citizens’ perception of police functions.
*p <.05; **p <.01.
As for the perceived ability of the police to control violent and property crime, Model 2 revealed that residents who were single/divorced/widowed were less likely to report that the police did a good job compared with people who were married. As the frequency of reading the local newspaper each week increased, the odds that citizens would agree that the police are doing a good job controlling crime in their community also increased. Like Model 1, people who were not crime victims (violent and property) were more likely to agree that the police were doing a good job controlling crime compared with those who have been victimized. Fear of victimization was also significant—residents who reported greater fear of victimization were less likely to agree that the police are doing a good job controlling crime. Finally, in the full model that combined both order maintenance and crime control functions of police (Model 3), only previous crime victimization experiences (violent and property) and fear of victimization were significant predictors. It appears that prior criminal victimization and fear of victimization are important variables regarding citizens’ perceptions of police functions. It is important to note that residential longevity was not statistically significant in any of the models; thus, it appears that there is no significant difference between new and long-term residents’ perceptions of the police ability to control crime and maintain order in this rapid-growth community.
Discussion
The purpose of this study was to provide a better understanding of how citizens perceive police performance in a small, rural community that is experiencing rapid population growth. Overall, residents reported that they believe that the police are doing a good job addressing community concerns; that is, most respondents said that the police are doing a good job maintaining order (81%) and controlling crimes (84%), suggesting that even under the current circumstances, the police are still able to perform their most basic functions successfully, at least from the perspective of residents living in the area.
In a growing town like Williston, one would expect that residents would be more critical of their local police and their ability to protect them from becoming victims of crime. Not only did the local news frequently report crime to have increased (Hargreaves, 2015), local police organizations struggled to adapt to increased demands resulting from rapid population growth (Archbold, 2015). Instead, residents’ views were generally positive, which prompts the question—why? More specifically, why did residents view police functions favorably when their community was experiencing significant changes in size and composition?
Williston is a town that is constantly changing. As such, it may be important to not only consider the context of the area, but also specific events that were occurring within such context. Although Williston grew in population, residents completed the surveys at a time when oil production was on the decline because of reductions in crude oil prices (Sankararaman and Medhora, 2015). Consequently, it is possible that oil-field workers were beginning to leave Williston in search of employment elsewhere in the country. For those who remained, it is likely that residents had a stake in the community and as a result, supported their local police and their efforts to address neighborhood problems. Future research should consider events that are unique to each context when investigating factors that could yield favorable views of police functions in other types of communities, and not just rural towns. Further, it is unknown whether this finding would persist if oil-field workers never left. Future research should explore this topic in growing communities that are unlikely to experience declines in population that are a result of losses in employment.
There are other notable findings resulting from the analyses. For instance, men were less likely than women to say that the police were doing a good job maintaining order in their community. Past researchers have suggested that more positive views held by women are due to the differential police treatment of men compared with women (Lai and Zhao, 2010; Spohn and Beichner, 2000; Weitzer and Tuch, 2006). It is also possible for men’s and women's views to be dependent on the environment. Holmes and Goodman (2010) surveyed a medium-sized southern US city and reported that women were not only less likely than men to believe that the police were doing a good job controlling crime and enforcing traffic laws in their neighborhood, they were also less likely to feel safe walking in their neighborhood and more likely to believe that crime was a problem in the area.
Compared with urban residents, people living in rural communities are more likely to know and trust their neighbors (Parker et al., 2018). Given that Williston could be considered relatively small even when it was growing in size, women (as opposed to the men who were at work during the day) may have had more opportunities to familiarize themselves with the town resulting in them feeling generally safer. More research is needed to understand gender differences better in the perceived performance of police in all types of communities, including those places that are experiencing a surge in population.
The literature regarding residents’ marital status is also mixed with some studies reporting no effect (Dowler, 2002; Hawdon et al., 2003; Schafer et al., 2003), whereas others observed it have an influence on residents’ perceptions (Holmes et al., 2017). Results from the current analysis revealed married individuals were more likely to report that the police were effective in controlling crime compared with unmarried individuals. It is possible that married residents and residents who owned their homes were simply more invested into the area; therefore, they may have actively sought to familiarize themselves with and support all aspects of their community, including their local police (Correia, 2000). By contrast, unmarried persons may have less investment in the community and, as a result, were less familiar with the efforts of the local police in tackling crime in their neighborhood. Most of the people who moved to Williston looking for work in the oilfields were single men who either rented living space or stayed in temporary housing that was subsidized by their employer.
In a study that used a nationwide sample of residents in the United States, media consumption (e.g. crime drama, crime news, number of hours watching television) was found to have no relationship with public perceptions of police effectiveness (Dowler, 2002); however, this does not appear to be the case in the current study. Williston residents who reported that they frequently read the regional newspaper were more likely to say that the police were doing a good job controlling crime, a finding is unexpected given the frequently reported link between the town's growth and the rise in crime.
Nevertheless, this finding could also be attributed to the reported layoffs and outflow of oil-field workers. In 2015, headlines such as the following surfaced: “Quitting the Bakken: One Oil Worker's Story” (Guerin, 2015); “Slowdown and Out in Williston, ND: Layoffs, Reduced Hours and Smaller Paychecks Drive Oil Boomers Into Hard Times or Out of Town” (Dizard, 2015); and “Halliburton Confirms layoffs in Williston” (Jean, 2015). The significance of this finding is further realized when considering that news reports often associated the influx of male workers to “man camps” (i.e. spaces dedicated to housing oil workers) to drastic increases in the crime rate, including murder, rape, assault, and drug-related problems (Hargreaves, 2015; Valentine, 2014). Because workers were reportedly leaving, perhaps too were the perceived problems that had accompanied them. Thus, it is possible that residents who read the regional newspaper may have been influenced by such reports, resulting in the belief that local police are now able to effectively address the crime problems occurring in their neighborhood.
Next, the analyses revealed that residents who have not been victims of violent or property crimes within the past two years were more likely to report that the police were doing a good job compared with residents who have been victimized, a finding that was present in all three statistical models of police functions. This outcome is consistent with the broader literature. For instance, Wu (2014) surveyed Seattle, Washington residents and found crime victims were less satisfied with police services and held more negative views of the police compared with those who had not been victimized. Similarly, Berthelot et al. (2018) used data from the 2013 Racial Attitudes Survey and found violent and non-violent crime victims expressed lowered confidence in the ability of police to fight crime.
Perceptual differences between victims and non-victims could be attributed to crime victims’ belief that the police have been ineffective in preventing crimes against them; or to put it another way, those who have not been victimized may be inclined to believe that the police (thus far) have been an effective force in preventing crimes against them. It is also possible that lowered ratings among crime victims are due to levels of distress. Avery et al. (2020) reported that crime victims who experienced greater distress because of their victimization rated police effectiveness lower than those who experienced minimal distress, although such differences were accounted for by the context of the neighborhood. Within small towns like Williston where people are generally familiar with and trusting of one another (Parker et al., 2018), residents may experience great distress when they find that they have been victimized. Regardless of the reason, it appears that prior criminal victimization is an important predictor of police functions in rapidly growing communities. Finally, residents who reported higher levels of fear of victimization were less likely to rate police functions positively, a finding that is in accord with the fear of crime literature showing individuals who were fearful of becoming crime victims were also more critical of the police (Hawdon et al., 2003; Nofziger and Williams, 2005; O’Connor, 2008).
Overall, the current analyses yielded some findings that were consistent with those that utilized data from large and small communities such as the impact of crime victimization and fear of victimization on residents’ perceptions of police performance. However, unlike previous research that did not observe a relationship between the amount of media consumed and perceived effectiveness of police (Dowler, 2002), here the role of media is underscored. Because Williston is a small town, it is unsurprising that residents would rely on their local newspaper for information about their changing community and, depending on the nature of the report, could have a positive or negative influence on people's perceptions. In short, more research is needed to determine what factors are most important for understanding police functions across communities with vastly different needs.
Limitations
The current study is not without limitations. First, women and older residents are over-represented in the study sample, which could have affected the findings. For instance, the variable age was not significant even though past studies typically find more favorable views of police among older residents compared with younger individuals (Berthelot et al., 2018; Gau, 2010; Holmes et al., 2017; McNeeley and Grothoff, 2016). Supplementing the door-to-door survey technique with an online version could help address the observed disparity in gender and age between the town's population and sample population. Moreover, the lack of racial diversity among Williston residents prevented further analyses from being made. Oversampling racial and ethnic minorities in smaller, less-diverse areas could be beneficial to researchers seeking to uncover perceptual differences based on group membership.
Nevertheless, it has been suggested that racial differences in citizens’ views are not as important as their beliefs about police action (Cao and Wu, 2019). When the police are seen as treating people fairly and with respect, citizens are more likely to view them in a positive light (Gau, 2010; Hinds and Murphy, 2007; Tyler, 2006). The current article does not examine police use of procedural justice and police legitimacy on residents’ perception of police performance, both of which are especially important when considering that residents living in small towns are more likely than urban residents to be familiar with their local police (Payne et al., 2005). Living in a rapidly changing community, however, could have altered the dynamics between these two parties (Archbold, 2015). Thus, future research should examine how procedural justice impacts police legitimacy and ultimately residents’ judgment of police performance in communities that are rapidly growing.
Third, although several high-profile incidents involving the police and minority citizens (e.g. the death of George Floyd and Breanna Taylor) were covered extensively in the news media, their effects on residents’ perceptions were not directly examined. It is also important to note that data for this study were collected prior to these events unfolding and the nationwide protests against police brutality that took place beginning in spring 2020. Future research should consider the effects of these occurrences and whether perceptions depended on the type of community where residents live.
Fourth, the current results are only generalizable to other small towns experiencing rapid population growth. Regardless of its initial size, any rapid population growth can be accompanied by both positive and negative effects. As previously mentioned, one potential downside is the reported increase in crime and disorder. Therefore, much can be learned by continuing to investigate police functions within this unique setting so that police departments can become better prepared when responding to rising neighborhood concerns.
Finally, this study captures residents’ views at one point in time (cross-sectional); therefore, it is unknown how or if residents’ perceptions may have changed over the years. As discussed, when the production of crude oil began to decline, residents’ views of police functions may have shifted as people continued to leave the area. Given the current study only examined residents’ perceptions amid rapid changes, it would be important to continue to survey residents over multiple time points to determine if/how the town's growth may have also shaped their views. With that said, another important question is whether cities whose growth is not attributed to mass oil production would produce similar results, for example communities that experience population increase because of local universities or during certain seasons (such as lakeside communities during the summer).
Based on the current findings, it would be important for police agencies in small rapid-growth communities to form partnerships with the local news media. Such partnerships would allow police departments to communicate with local residents about what they are doing to keep community members safe. Presenting local crime statistics could also help residents determine whether there is anything to be fearful of, rather than relying on information that may not accurately reflect the current state of their neighborhood. In addition, community leaders should advance tactics that would encourage people, especially newcomers to the town, to invest in the area. For example, offering prospective home buyers incentives to purchase a home for them and their families has the potential to shape how they view the town; which could then impact their level of engagement in the community.
Footnotes
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) received no financial support for the research, authorship, and/or publication of this article.
