Volume 4, Issue 1 • Winter 2015

Table of Contents

Foreword

Modifying Dialectical Behavior Therapy for Incarcerated Female Youth: A Pilot Study

The Impact of Child Protective Service History on Reoffending in a New Mexico Juvenile Justice Population

Social Distance Between Minority Youth and the Police:
An Exploratory Analysis of the TAPS Academy

Rural Youth Crime: A Reexamination of Social Disorganization Theory’s Applicability to Rural Areas

How to Help Me Get Out of a Gang: Youth Recommendations to Family, School, Community, and Law Enforcement Systems

Exploratory Research Commentary:
How Do Parents and Guardians of Adolescents in the Juvenile Justice System Handle Adolescent Sexual Health?

Rural Youth Crime: A Reexamination of Social Disorganization Theory’s Applicability to Rural Areas

Matthew D. Moore, Department of Sociology and Criminal Justice, Grand View University; and Molly Sween, Department of Criminal Justice, Weber State University.

Correspondence concerning this article should be addressed to Matthew D. Moore, Department of Sociology and Criminal Justice, Grand View University, Rasmussen Center 128, Des Moines, IA 50316. E-mail: mmoore@grandview.edu

Keywords: social disorganization theory; rural crime; juvenile delinquency

Abstract

Criminological theories are often developed based on studies of urban areas. The current analysis examines the applicability of social disorganization theory to youth crime in rural areas using Osgood and Chambers’ (2000) analysis. The current analysis used negative binomial regression models to test social disorganization theory based on juvenile arrest rates in 2,011 nonmetropolitan counties within 48 states in the United States. The findings indicate that social disorganization theory can be applied to understand youth crime in rural areas: residential mobility, ethnic heterogeneity, family disruption, poverty, and population density predicted higher levels of crime. Although the population at risk—those between the ages of 15 and 24—was significant, we found age was not associated with crime in rural areas, which is the opposite finding of other social disorganization theorists. The unemployment rate of the counties had no effect on crime in our study. The findings of the current study match many of the findings of Osgood and Chambers’ (2000) original analysis on social disorganization theory and rural crime, indicating that many of the components of social disorganization theory can be applied to understanding youth crime in rural areas.

Introduction

The vast majority of criminological research has been done in urban areas. Since little attention has been paid to delinquency in rural areas (Kaylen & Pridemore, 2013; Wells & Weisheit, 2004), our portrait of delinquent behavior is incomplete. This neglect of rural areas produces confusion in criminology. Kaylen and Pridemore (2013) explain that many studies treat rural areas as “miniature versions of urban areas, with similar social processes occurring on a smaller scale” (p. 170). As such, criminologists tend to falsely accept that theories and causes of crime are the same for rural and urban areas.

Although social disorganization theory—defined as the decline in the influence of existing social rules on the behavior of individuals—has been applied to rural areas in a small number of studies (Jobes, Barclay, Weinand, & Donnermeyer, 2004; Kaylen & Pridemore, 2011; Osgood & Chambers, 2000), the results of the studies examining the relation between social disorganization theory and rural crime have produced mixed results. Despite the equivocal results, the researchers who tested this relationship have argued that social disorganization theory can be applied to rural areas (Osgood & Chambers, 2000). However, Kaylen and Pridemore (2013) point out studies on rural crime have suffered from problems with data measurement and collection, making any studies of rural crime difficult to compare with those focusing on urban crime. To aid in moving research on rural crime forward, the current analysis builds on Shaw and McKay’s (1942) conclusion that rural areas experiencing a high rate of crime are socially disorganized.

Scholars have begun to acknowledge that individuals’ motivations for and environmental contributors to crime may be different in urban and rural areas (Deller & Deller, 2010). For example, Wells and Weisheit (2004) examined urban and rural areas across the United States and found that some of the predictors of crime in urban areas were not associated with crime in rural areas.

Kaylen and Pridemore (2013) pointed out that new research is emerging which is studying rural crime, but that rural crime is still an understudied area of criminology. Moreover, Kaylen and Pridemore (2013) explained that many studies treat rural areas as “miniature versions of urban areas, with similar social processes occurring at a smaller scale” (p. 170). As such, criminologists tend to falsely accept that theories and causes of crime are the same for rural and urban areas based on their view that rural areas are just miniature versions of urban centers.

The current study examines the generalizability of social disorganization theory to rural areas by building on Osgood and Chambers’ (2000) analysis. We studied 2,011 rural counties across the United States to test the theory’s applicability to crime in rural areas. Using a larger sample size than Osgood and Chambers (2000) and following the same methodological approach, the results of the current analysis attempt to provide more generalizability than previous studies using similar dependent variables (Kaylen & Pridemore, 2011; Osgood & Chambers, 2000).

Literature Review

Social Disorganization Theory

Recognizing that the city of Chicago was undergoing drastic structural changes in the 1920s and 1930s, Shaw and McKay (1942) set out to understand the relationship between place and juvenile delinquency rates. Shaw and McKay (1942) demonstrated to criminologists that social ecological factors could impact criminal patterns. After gaining access to juvenile court records, they mapped out where each youth lived within the city of Chicago. They found juvenile crime rates were drastically different from one place to the next. More specifically, they saw that the highest rates of juvenile delinquency were concentrated near the center of Chicago, and the lowest rates of juvenile delinquency were found on the outskirts of the city.

In trying to explain this phenomenon, Shaw and McKay (1942) claimed that areas with high rates of juvenile delinquency were structurally different than areas with lower rates of juvenile delinquency. Shaw and McKay (1942) illustrated that as the city center transitioned from being residential to primarily commercial in nature, the effects on the residents living there were negative. For example, during the transition, residents were either forced to find housing elsewhere or submit to living in substandard conditions. This led to residential turnover, an increased number of broken families, and an overall increase in concentrated poverty and decay (Akers & Sellers, 2009).

While the physical and structural changes that Shaw and McKay (1942) discovered were alarming, they were more concerned with what impact these changes had on the relationships among people living within these communities. These areas of transition were described as “socially disorganized” in that the residents living there experienced high rates of population turnover, were described as ethnically heterogeneous (due to an influx of immigrants), and the community suffered from many of its citizens living either in poverty or in a lowered socioeconomic status. Because of these rapid changes and the resulting stressors these changes placed upon people, social disorganization has been credited with hampering the overall levels of informal social control that people are willing to exert over one another (Bursik & Grasmick, 1993). Shaw and McKay (1942) argued that the crime-ridden areas were socially disorganized, and that this social disorganization negatively influenced community members’ willingness to intervene and prevent juvenile delinquency from occurring.

Social disorganization theory initially gained much interest from the criminological community due to its unique place-based perspective. However, social disorganization theory fell out of favor, and the theory remained relatively dormant until the late 1980s and early 1990s. Around this time, scholars began empirically testing social disorganization theory as a theoretical framework and explored the theory’s validity in a variety of different neighborhood contexts (Sampson & Groves, 1989; Bursik & Grasmick, 1993; Sampson, Raudenbush, & Earls, 1997). This growing body of research has led to a renewed interest in social disorganization theory among criminologists.

More recent studies examining social disorganization theory have found that additional features of a neighborhood, such as high population density, poverty, unemployment, and a large percentage of female-headed households, are associated with crime and delinquency (Markowitz, Bellair, Liska, & Liu, 2001; Li, 2011; Kaylen & Pridemore, 2011). Other research has demonstrated that perceptions of neighborhood disorder can be explained by social disorganization theory (Witherspoon & Ennett, 2011). Sampson (2012) illustrated that lower levels of crime in a community may be due to residents being able to effectively communicate with social control agencies. Allen and Cancino (2012) applied social disorganization theory to the Texas–Mexico border region and found many of the variables associated with social disorganization are related to crime on the Texas–Mexico border. Furthermore, Mustaine, Tewksbury, Huff-Corzine, Corzine, and Marshall (2014) demonstrated that social disorganization theory can be applied to child sexual assault. Therefore, based on the studies just mentioned, Shaw and McKay’s (1942) initial conceptualization of social disorganization theory has been expanded to explain many different areas of crime.

Social Disorganization Theory and Rural Crime

Social disorganization theory was initially developed to explain crime in urban areas. However, more recently scholars have expanded social disorganization theory to examine crime and delinquency in rural areas. Studies have been conducted using social disorganization theory not only in rural areas of the United States (Osgood & Chambers, 2000; Osgood & Chambers, 2003; Bouffard & Muftić, 2006; Li, 2011) but in rural areas in other countries (Jobes et al., 2004). Unfortunately, the number of studies that have explicitly tested social disorganization theory in the rural context is limited and their findings are mixed (Kaylen & Pridemore, 2012).

Osgood and Chambers (2000) examined social disorganization theory in rural counties across four states in the United States. In their research, they tested many of the key variables associated with social disorganization (i.e., residential instability, ethnic diversity, family disruption, low economic status, high population density, and proximity to urban areas) using arrest rates for juveniles in 264 nonmetropolitan counties in Florida, Georgia, Nebraska, and South Carolina. The authors hypothesized that rates of juvenile violence would be positively related to all of their social disorganization theory variables. They found that many of the key variables of social disorganization theory associated with crime and delinquency in urban areas were also associated with crime and delinquency in rural areas. Osgood and Chambers (2000) demonstrated that residential instability, ethnic diversity, and family disruption were significant predictors of juvenile arrest rates in rural counties in the four states they analyzed. Osgood and Chambers argued, based on the findings of their study, that the basic components of social disorganization theory could be used to explain crime in both urban and rural areas.

Kaylen and Pridemore (2011) examined social disorganization theory in rural areas of Missouri. Using hospital records from 106 rural counties in Missouri, the scholars found that only family disruption was a significant predictor of crime in rural areas. Kaylen and Pridemore (2011) concluded that “the association between traditional social disorganization variables and youth violence may not be generalizable to rural areas” (p. 987).

In other research studying the connection between social disorganization theory and rural crime, Jobes et al. (2004) examined social disorganization and crime in rural Australia. Using a cluster analysis, Jobes et al. found that communities had lower crime rates when they had more cohesive and integrated community structures. Therefore, the authors concluded that social disorganization theory is applicable to both urban and rural areas. The different conclusions drawn by Osgood and Chambers (2000), Kaylen and Pridemore (2011, 2013), and Jobes et al. (2004) have called into question our understanding of the applicability of social disorganization theory to crime in rural areas.

More recently, scholars have called into question the ability of social disorganization theory to explain crime in contexts other than cities in the United States. Examining the applicability of social disorganization to a Western European city, Bruinsma, Pauwels, Weerman, and Bernasco (2013) surveyed 3,575 residents in 86 neighborhoods of The Hague. They collected information on six different models of social disorganization, such as the classic model and collective efficacy. Bruinsma et al. (2013) concluded that social disorganization does not explain crime in The Hague. Instead, social disorganization may be better suited to explaining distinct urban processes. This has led to doubt as to whether social disorganization theory can be applied to areas other than cities in the United States.

The Current Study

The goal of the current study is twofold. First, given the limited research that has directly tested social disorganization theory in the rural context, it is our hope that this study will add to that growing body of literature. A second goal of this study is to provide more explanatory power than previous studies have been able to do. In an effort to do this, we are using Osgood and Chambers’ (2000) study as a framework for our analysis. As such, we used the same social disorganization theory variables and juvenile crimes derived from the Uniform Crime Report as Osgood and Chambers. Our study differs from that of Osgood and Chambers, however, in that we are conducting our analysis on all nonmetropolitan counties in 48 (Hawaii and Alaska not included)1as opposed to only four states. Our goal in conducting this larger analysis is to provide a greater degree of generalizability than past studies, using smaller sample sizes, were able to (Bouffard and Muftić, 2006, N = 221; Jobes et al., 2004, N = 123; Kaylen and Pridemore, 2011, N = 106; and Osgood and Chambers, 2000, N = 264).

1Alaska and Hawaii were not included because of missing and incomplete data from the U.S. Census Bureau and the Uniform Crime Report.

Drawing on the findings from previous rural crime studies, we hypothesize there will be a relationship between juvenile delinquency and social disorganization variables in rural communities. More specifically, we hypothesize that juvenile delinquency rates will be positively associated with all of our social disorganization variables (i.e., residential instability, ethnic heterogeneity, family disruption, a high poverty rate, population at risk, unemployment, and population density). In addition, we hypothesize that our findings, like those of Osgood and Chambers (2000), will lend the most support to the variables of residential instability, ethnic heterogeneity, and family disruption.

Methods

Data

As Osgood and Chambers (2000) pointed out in their analysis, most studies at that time focused on variation in crime rates in neighborhoods in the same metropolitan area. This type of analysis does not allow for generalization to other areas of the country. To correct for this type of operationalization, Osgood and Chambers (2000) used county-level data from four different states: Florida, Georgia, South Carolina, and Nebraska. While their county-level analysis and use of different states was an improvement over many past analyses, the generalizability of Osgood and Chambers’ (2000) findings does have limits. Therefore, the current analysis included more rural counties in the United States (N = 2,011) and a larger sample of states (N = 48).

More recently, questions about the validity of county-level data have been brought to light, especially in rural counties and in counties with small populations (Kaylen & Pridemore, 2011; Lott & Whitley, 2003; Maltz & Targonski, 2002; Wiersema, Loftin, & McDowall, 2000). As mentioned above, county-level Uniform Crime Report data do have limitations. Moreover, Maltz and Targonski (2002) have explained that missing data and imputed data at the county level make using the data problematic. Even with these limitations, we believe we are justified in using county-level data for the current analysis. We acknowledge the limitations of the Uniform Crime Report data and believe the findings of this study will not be the sole source of information on crime in rural areas, but will be used as one of many studies examining this issue (see Limitations section).

The current study includes all counties that were not considered part of a metropolitan statistical area (MSA) by the Census Bureau. The Census Bureau classifies counties as not being a part of an MSA when it does not have a city of 50,000 or more, as well as when less than 50% of the county population resides in a metropolitan area of 100,000 or more. The current analysis included 2,011 counties with an average population of 24,580. The counties range in population from 41 to 190,846.

The Dependent Variables

The dependent variables for the analysis were collected from the Uniform Crime Report. In total, seven dependent variables were included in the current analysis: murder, rape, robbery, aggravated assault, weapons, and simple assault. The Violent Crime Index for 2010, which is the sum of murder, rape, robbery, and aggravated assault, was also included in the analysis, as was the total number of juvenile arrests for each crime. Negative binomial regression was used to control for any spikes in crime rates that can occur due to a small increase in the number of events occurring in an area with a small population (the Analytic Strategy section describes negative binomial regression). Social disorganization theory focuses on how the environment affects juvenile delinquency rates. Therefore, Uniform Crime Report data on juvenile arrests were used in the current analysis. The Uniform Crime Report classifies juveniles as individuals aged 11 through 17. Table 1 illustrates the descriptive statistics for the variables in the analysis.

The Independent Variables

All of the independent variables were collected from the United States Census Bureau. The 5-year estimates from the American Community Survey were used for the years 2006-2010. The 5-year estimates were used because the Census Bureau does not collect information every year for all counties with small populations (see Table 1).

Table 1. Descriptive Statistics for Variables Used in the Analysis

Mean

Standard Deviation

Residential Instability

.28

.06

Ethnic Heterogeneity

.17

.15

Family Disruption

.06

.03

Poverty Rate

.12

.06

Population at Risk

.15

.26

Unemployment

.04

.02

Population Density

45.73

110.36

Violent Crime

3.05

6.64

Murder

.06

.61

Rape

.23

.81

Robbery

.47

1.61

Assault

2.27

5.08

Weapons

1.14

3.53

Simple Assault

13.27

25.66

The variables used in the current analysis were selected based on the traditional social disorganization theory model of urban areas and on the variables that Osgood and Chambers (2000) selected for their original analysis. Residential instability was measured as the proportion of the population that had moved since 2005. Ethnic heterogeneity refers to the proportion of households occupied by white versus nonwhite persons. The ethnic heterogeneity measure calculates the likelihood of two randomly selected individuals from the county having different ethnicities. Following Osgood and Chambers (2000), we calculated the ethnic heterogeneity measure as 1(pi)2, whereby pi is the proportion of households within a given ethnic group (i.e., white or nonwhite). The proportion of the households within a given ethnic group is then squared and summed across the two groups. The ethnic heterogeneity measure ranges from 0 to 0.5. A score of 0 indicates the county has only white or nonwhite residents; a score of 0.5 indicates the county has an equal number of white and nonwhite residents (see Table 2).

Table 2. Negative Binomial Regression Analysis and Type of Crime

Violent Crime

Murder

Rape

Robbery

Assault

Weapons

Simple
Assault

B

(S.E.)

B

(S.E.)

B

(S.E.)

B

(S.E.)

B

(S.E.)

B

(S.E.)

B

(S.E.)

Residential Instability

3.755***

(.590)

-1.269)

(2.375)

3.061**

(1.152)

5.914***

(1.060)

3.771***

(.615)

3.917***

(.768)

4.957***

(.505)

Ethnic Heterogeneity

2.831***

(.298)

5.826***

(1.172)

1.769**

(.609)

4.682***

(.549)

2.519***

(.312)

2.896***

(.378)

1.916***

(.245)

Family Disruption

7.011**

(2.085)

-3.487

(7.997)

1.888

(4.448)

12.865**

(3.907)

6.877**

(2.175)

7.563**

(2.711)

6.750***

(1.674)

Poverty Rate

-.306

(.859)

12.575***

(3.184)

-1.593

(1.979)

.902

(1.689)

-.584

(.909)

.175

(1.171)

-.690

(.712)

Population at Risk

-2.670***

(.140)

-2.674***

(.485)

-2.700***

(.277)

-3.449***

(.258)

-2.633***

(.146)

-3.151***

(.182)

-2.721***

(.115)

Unemployment

-2.312

(2.102)

-12.581

(7.943)

-2.623

(4.433)

-2.717

(3.918)

-2.527

(2.218)

-.148

(2.725)

.885

(1.755)

Population Density

.100

(.114)

1.331**

(.419)

-.324

(.230)

.540*

(.210)

.072

(.120)

.128

(.150)

.104

(.093)

Adjacent to Metro Area

-.139*

(.066)

-.362

(.289)

-.126

(.130)

-.081

(.123)

-.129

(.069)

-.149

(.085)

-.057

(.056)

Northeast

.788***

(.146)

-.335

(.824)

1.148***

(.241)

.716**

(.239)

.703***

(.150)

.327

(.178)

.940***

(.129)

Midwest

.539***

(.093)

.961*

(.442)

.795***

(.185)

.155

(.182)

.534***

(.097)

.389**

(.121)

.494***

(.074)

West

.599***

(.128)

2.097***

(.379)

.199

(.257)

.217

(.233)

.650***

(.134)

.641***

(.165)

.620***

(.109)

*p < .05; **p < .01; ***p < .001; (Standard Error)

Previous studies of social disorganization have used ethnic heterogeneity in their analyses. Kornhauser (1978) found support for increased ethnic heterogeneity and increased crime in urban areas. Bursik and Webb (1982) argued that increased ethnic heterogeneity may lead to groups leaving the neighborhood due to the new groups moving in. Thus, increasing residential instability has been demonstrated to increase crime. Examining both urban and rural areas, Wells and Weisheit (2004) found ethnic heterogeneity to be a consistent predictor of violent crime in rural and urban settings. When examining the research on rural crime, ethnic heterogeneity has been shown to be predictive of crime (Bouffard & Muftić, 2006; Osgood & Chambers, 2000; Wells & Weisheit, 2004). Again, an increased level of ethnic heterogeneity may indicate the inability of the neighborhood to increase informal social control.

We measured family disruption by the proportion of female-headed households in the county and the poverty rate by the proportion of families living below the poverty level. In line with the work of Osgood and Chambers (2000), we also included the unemployment rate as a second economic measure. We measured population at risk in the same way as Wells and Weischeit (2004) in their analysis of urban and rural crime: the proportion of the population between the ages of 15 and 24. This is a deviation from the work of Osgood and Chambers (2000), who focused on the proportion of the population between the ages of 10 and 17. We used the ages of 15 to 24 for two reasons. First, the ages of 15 to 24 more closely follow the age-crime curve. In this way, the age groups most likely to commit a crime could have a large influence on crime levels in a county. Second, because the U.S. Census Bureau changed the age range category in 2000, focusing on the 10- to 17-year-old age group was no longer an option. The distribution of the population between the ages of 15 and 24 was skewed; to correct for this, we used the natural logarithmic transformation in the current analysis. The natural logarithmic transformation did correct the skewed distribution of the population at risk.

We also included the population density in the model. Population density is calculated as the population of the county divided by the land area, in square miles, of the county. The distribution of the population density was skewed, and the natural logarithmic transformation was used to correct for this skewed distribution. The natural logarithmic transformation did correct the skewed distribution of population density. We created the category adjacent to the metro using Beale codes (U.S. Department of Agriculture, 2013). Beale codes categorize counties as being either adjacent to metropolitan areas or nonadjacent, with 1 being adjacent to metropolitan areas and 0 being nonadjacent. Finally, the current analysis used region codes to control for the region of the country in which the county was located. The South has higher rates of crime than other regions of the country and has been controlled for in previous studies (Blau & Blau, 1982; Gastil, 1971; Nisbett & Cohen, 1996). Taking this into account, dummy variables for the Northeast (N = 95), Midwest (N = 761), and West (N = 285) were used. The South was the reference category for the current analysis. The current study tested for multicollinearity and we did not experience any issues related to this assumption. Appendix A provides the correlations for the variables used in the analysis.

Table 3. Number of Counties That Are Adjacent to Metro, Non-Adjacent to Metro, and Region

N

Percent

Adjacent to Metro

1053

52.36

Non-Adjacent to Metro

958

47.64

Northeast

95

4.72

South

870

43.26

Midwest

761

37.84

West

285

14.17

Analytic Strategy

We used negative binomial regression to understand the effect of social disorganization theory on juvenile delinquency. Osgood (2000) demonstrated that negative binomial regression is the proper statistical method to use when examining crime in rural areas. Osgood (2000) explained that counties with small populations could have a large rise in crime rates with the occurrence of just one crime. At the same time, a more populated area would see only a small increase in its crime rate with one additional crime. Thus, areas with small populations could have significantly higher crime rates than more heavily populated areas even though more crimes had occurred in the more heavily populated area. For example, if an area had 10,000 residents and had one homicide the rate would be 10. If that same area had two homicides the next year the rate would jump to 20. An area with 500,000 residents would have 50 homicides to have a rate of 10 and would have to experience 50 more homicides to have a rate of 20. Negative binomial regression uses counts as the dependent variable to take away the large rate increases that small areas would experience due to a small increase in crime.

Findings

Results shown in Table 2 indicate that residential instability was a significant predictor of crime for the violent crime index and the crimes of rape, robbery, assault, weapons, and simple assault. Only murder was not significant in our analysis. The coefficient for residential instability and the violent crime index was 3.755. This coefficient indicates that with each unit increase of residential instability, the expected count of violent crime increases by 3.755. The coefficients for rape, robbery, assault, weapons, and simple assault were all positive and large, suggesting that residential instability is a key variable in explaining youth crime in our sample of rural counties.

When examining ethnic heterogeneity, we found this variable was significant for all crimes in our analysis. All of the coefficients were positive, indicating that an increase in ethnic heterogeneity increased the occurrence of the crimes we analyzed. For example, with each unit increase of ethnic heterogeneity, the expected count of rape increased by 1.769. Like residential instability, ethnic heterogeneity was a significant predictor of youth crime for the rural counties in our analysis.

Family disruption was positive and significant for violent crime, robbery, assault, weapons, and simple assault among youth in rural areas. Family disruption was not significant for murder or rape. The positive coefficients demonstrated that family disruption increases violent crime, robbery, assault, weapons, and simple assault among youth in rural areas. For example, with each unit increase on the scale of family disruption, the expected count of simple assault increased by 1.916.

Poverty was significant only for murder in our analysis. With each unit increase on the poverty scale, the expected count of murder among juveniles in rural areas increased by 12.575. The unemployment rate was not significant for any of the crimes in the analysis. Population density was significant and positively related to murder and robbery. For each unit increase on the scale of population density, the expected count of robbery increased by 0.540.

The variable population at risk was significant for all of the crimes in the current analysis. However, the coefficients were negative, which was not in the direction we theorized. For example, with each unit increase on the scale of population at risk, the expected count of assault decreased by 2.633. This finding indicates the population at risk may have a different effect on crime in rural than in urban areas.

The dummy variable, being adjacent to the metro, was significant for the violent crime index. However, the coefficient was negative (-.139). Being adjacent to the metro was not significant for any of the other crimes in the analysis. Violent crime was significant at the p < .05 level; the nonsignificant findings for the other crime variables suggest that a county adjacent to a metropolitan area is not at increased risk for crime.

We included the regional variables in the analysis to control for variation in juvenile crime rates across the United States. The Northeast was significant and positive for violent crime, rape, robbery, assault, and simple assault. The Midwest was significant and positive for violent crime, murder, rape, assault, weapons, and simple assault. The West was significant and positive for violent crime, murder, assault, weapons, and simple assault. These findings do not align with previous studies that found the South has higher levels of crime than other regions of the country. However, past analyses examining rural crime rates did not control for region (Barnett & Mencken, 2002; Deller & Deller, 2010; Kaylen & Pridemore, 2011; Osgood & Chambers, 2000; Wells & Weisheit, 2004). Therefore, the findings could be due to actual differences in criminal activity in the different regions, or may be evidence that the problems with Uniform Crime Report data (see the Limitations section) are more problematic in a particular region, such as the South.

Discussion

Past research on the relation between social disorganization theory and crime has produced mixed results. Some researchers have found that social disorganization theory does predict crime in rural areas (Jobes, et al. 2004; Osgood & Chambers, 2000) and others have found that social disorganization theory does not explain rural crime (Kaylen & Pridemore, 2013; Wells & Weisheit, 2004). Still other research on social disorganization theory and crime has found that certain factors of the theory are related to rural crime, whereas other variables are not (Barnett & Mencken, 2002; Kaylen & Pridemore, 2011; Li, 2011). In the current study, we found that certain factors of social disorganization theory can explain juvenile crime in rural areas (i.e., residential instability, ethnic heterogeneity, and female-headed households), whereas other factors have little to no relationship with rural juvenile crime (i.e., poverty rate, population density, and unemployment). We found that other variables influenced youth crime in rural areas, but in the opposite direction than theorized (i.e., population at risk).

Residential instability has been demonstrated to be a key factor in increasing crime rates in both urban (Kapsis, 1978; Kornhauser, 1978; Sampson, 1995; Xie & McDowall, 2008) and rural areas (Petee & Kowalski, 1993; Osgood & Chambers, 2000). The current analysis provides evidence that residential instability is a key factor predicting juvenile crime in rural counties in the United States. As social disorganization theory would predict, increased residential instability would reduce the level of informal social control within the neighborhood. As a result, residents may be less likely to watch out for and react to improper behavior in a neighborhood because of the anonymity of the residents in that neighborhood.

Ethnic heterogeneity was also a significant predictor of rural crime in our analysis. In Shaw and McKay’s original 1942 conceptualization of social disorganization theory, ethnic heterogeneity was theorized to break down social ties within the neighborhood, thus reducing the likelihood of informal social control. As different racial and ethnic groups moved into an area, the social solidarity needed to create organizations and social ties would not develop. Thus, crime would increase in areas that have a high level of ethnic heterogeneity. Shaw and McKay did not argue that race itself was the cause of crime. Instead, they contended that when new groups move into the socially disorganized neighborhood, they take on the problems of that neighborhood. The neighborhood is the cause of crime because of its disorganized structure, not the racial or ethnic composition of its community members.

Sampson and Groves (1989) argued that family disruption (e.g., female-headed households) would increase social disorganization within a neighborhood. Single parents would have less resources and time to control the behavior of their children. Studies on social disorganization and crime in urban areas have found that family disruption does increase crime (Rocque, Posick, Barkan, & Paternoster, 2014; Sampson & Groves, 1989). Our analysis found that family disruption was significant for all crime committed by juveniles with the exception of murder and rape. Other studies on rural crime have also found that family disruption increased crime levels (Bouffard & Muftić, 2006; Osgood & Chambers, 2000). Moreover, Kaylen and Pridemore (2011) found family disruption to be the strongest predictor of crime when using hospital records from rural areas in Missouri.

The population at risk was significant in our model, but in the opposite direction than theorized. That is, we found the population at risk—the proportion of the population ranging in age from 15 to 24—to be negatively associated with crime in rural areas. Wells and Weisheit (2004), who also used this measure in their analysis of social disorganization theory, similarly found the population at risk to be negatively associated with crime in both urban and rural settings. On the other hand, Osgood and Chambers (2000) found the population at risk in their study—that is, the proportion of the population ranging in age from 10 to 17—to increase crime. Since the U.S. Census Bureau changed their groupings of age ranges in 2000, the age range 10 to 17 was not available for the current analysis. The negative association we found between age and crime could be due to the current age range category capturing the beginning of the desistence process of the age-crime curve. However, the negative association may also indicate that the age of the population at risk plays little or no role in crime in either urban or rural areas.

Like Osgood and Chambers (2000), we did not find unemployment or the poverty rate to be significant predictors of juvenile crime in rural areas. Many authors have noted that these two variables may operate differently in urban and rural areas (Weisheit, Falcone, & Wells, 1994). In the current analysis the poverty rate was significant for murder, but for all other crimes both the poverty rate and unemployment were not significant. Wells and Weisheit (2004) pointed out a similar pattern in their analysis of rural and urban areas. They found that the poverty rate was a significant predictor of crime in urban areas, but not a significant predictor in rural areas. While it is hard to fully explain our findings, some have argued that social factors better predict crime in rural communities than do economic factors (Weisheit et al., 1994). According to Osgood and Chambers (2003), “it appears that—unlike in most urban areas—poverty does not disrupt the social fabric of small towns and rural communities” (p. 6). From our analysis and the previous analysis, it seems that poverty and unemployment may be more predictive of crime in urban areas than rural areas.

Population density was significantly associated with increased levels of murder and robbery in the current analysis. However, the conclusion that population density is a significant predictor of crime in rural areas is unclear. Wells and Weisheit (2004) and Osgood and Chambers (2000) found mixed results for population density. Li’s (2011) study found that population density was not related to property crime but was negatively related to violent crime. Therefore, the role of population density and social disorganization theory in rural crime is unclear. We can only speculate that perhaps people living in less densely populated areas form stronger social networks (which increases informal social control and minimizes crime), as has been posited by Wilkinson (1984a & 1984b).

Finally, being adjacent to a metropolitan area was not significant for any juvenile crime in the study, except violent crime, which was significant in the opposite-than-theorized direction. This was in line with the findings of Osgood and Chambers (2000). A rural county that is adjacent to a metropolitan area does not have a spillover of crime, but rather seems to have no bearing on crime in the county. This finding is also consistent with studies demonstrating that crime displacement does not occur (Guerette & Bowers, 2009; Weisburd et al., 2006). In the context of this study, it appears that juvenile crime does not move to other locations adjacent to crime-prone areas.

Limitations

One potential reason our findings are inconsistent with those of other studies testing social disorganization theory in rural communities is that we are using different data. After the Osgood and Chambers 2000 study was published, researchers began questioning the reliability of data from the Uniform Crime Report for small counties in the United States (Lott & Whitley, 2003; Maltz & Targonski, 2002; Wiersema et al., 2000). It is possible that residents of rural counties may be less likely to report a nonviolent crime to the police because such crimes may be handled in an informal manner. Urban areas, on the other hand, do not have as many tightly knit groups that would enable them to handle nonviolent crimes informally. However, violent crimes would be as difficult to handle informally in rural as in urban areas because of the nature of a violent act (i.e., it is difficult to hide a body).

Another potential problem with Uniform Crime Report data is the potential to underestimate the incidence and prevalence of various crimes. Because the Uniform Crime Report contains information only on crimes known to the police, there may be a large number of crimes that are not reported and, therefore, not known to the police. Klaus (2004) pointed out that there could be a number of reasons individuals may not report a crime to the police, such as fear of reprisal and the belief that the crime was not important enough to contact the police. Klaus (2004) went on to estimate that approximately 42% of crime is reported to the police. This would lead to a large gap between actual crime and crime that is reported.

To address the problems found in the Uniform Crime Report county-level data, researchers have explored using other data sets. Kaylen and Pridemore (2013) used hospital records from Missouri to examine rural crime. The scholars argued that hospital records would provide a better data set than Uniform Crime Report county-level data because hospital codes are standardized by the World Health Organization (WHO), and the codes used have been shown to be reliable (Kaylen & Pridemore, 2013). Moreover, Kaylen and Pridemore (2013) pointed out that many past studies exploring violence have used hospital records. However, it is still unclear why researchers should assume that hospital records would not suffer from some of the same problems as Uniform Crime Report data: that is, using hospital records supposes that an individual who is assaulted would go to the hospital instead of contacting the police. Kaylen and Pridemore (2013) did explain that individuals do go to the hospital when they are seriously injured; however, we cannot be sure how much crime hospital data will capture.

Other scholars have suggested using victimization surveys such as the National Crime Victimization Survey, to collect data on crime. The use of crime victimization surveys may begin to capture the unreported crime. However, victimization surveys also have a number of potential problems. Some individuals may lie about an incident because of embarrassment. Others may make up crimes that had not occurred with the belief they are helping the researcher. Respondents to a victimization survey may have problems remembering when an event took place or how many times. Victimization surveys may also have sampling errors. Thus, the use of surveys may lead to problems similar to those found when using Uniform Crime Report data.

We do acknowledge the potential problems with Uniform Crime Report data, and it is our goal to recognize these issues and not claim that our analysis is the only study that should be used to gauge social disorganization theory’s applicability to youth crime in rural areas. However, we feel justified in using the data and then comparing our findings to those of studies that use the same data, as well as data from different sources. By comparing our findings to those of others we might begin to see a pattern emerge and find new ways to measure crime in rural areas.

Conclusion

Our analysis demonstrates that certain factors of social disorganization theory do apply to rural areas. Residential instability, ethnic heterogeneity, and family disruption were all significantly associated with increased levels of crime. The poverty rate, unemployment, and population density were either nonsignificant or yielded inconsistent findings. The population at risk was associated with crime but in the opposite direction theorized. The current analysis demonstrates that social disorganization theory cannot be applied fully to youth crime in rural areas, but that parts of the theory can be applied while other parts have no association. However, given that our findings are so similar in scope to those of Osgood and Chambers (2000), we believe we can draw similar conclusions; while not directly transferable from the urban to rural context, social disorganization theory can be useful to help us start to make sense of the phenomenon of juvenile rural crime.

Rural crime does provide us with a unique opportunity to examine criminological theories. Often, theories of crime ignore rural communities. Wells and Weisheit (2004) pointed out that theories often assume rural areas are just small urban areas. The disregard for rural areas leaves criminologists in an awkward position when examining crime in rural areas. New measures may need to be developed to fully tap into the potential of using social disorganization as a theoretical framework for explaining rural crime.

In addition, our findings can help shed light on possible policy implications derived from social disorganization theory. If one were to adhere to the ideas positioned by Warner, Beck, and Ohmer (2010), then informal social control should be conceptualized as a mechanism that increases direct intervention within a community. In an effort to reach these goals, Shaw and McKay implemented Community Action Programs designed to encourage “the community to develop their own solutions to problems and were based primarily on providing social support” within the community (Warner et al., 2010, p. 355). Examples of policies they implemented include recreation and mediation outlets for delinquent youth, as well as community organized committees overseeing change from the ground up. While these efforts proved successful in various pockets of Chicago, one could question whether similar policies would be successful if replicated in rural settings.

As things stand now, little is known about what policy implications specifically focused on juvenile rural crime would look like, due mainly to the limited research on the topic. Some have suggested a grassroots approach, while others argue for more systematic responses in the way of Federal assistance to rural social service agencies (Weisheit et al. 1994; Cancino, 2005). In staying true to some of the tenets of social disorganization theory, Cancino suggests that rural communities need to work on strengthening the levels of social cohesion among its citizens (2005). In order to do so, Cancino suggests more frequent contact between law enforcement, local politicians, and citizens in the community to collaborate in crime fighting efforts (2005). Similar to the above suggestions, Wilkinson (1984a and 1984b) argues that rural communities need to focus on strengthening social ties among its citizens because this “makes these places less likely to see an increase in crime even if they exhibit high rates of family disruption, poverty and other forces” (referenced in Li, 2011, p. 67).

Perhaps, as Warner et al. (2010) claim, we should begin thinking “outside the box” about ways to increase social control. This may be an even more challenging task when trying to address juvenile crime. Rural crime poses a unique challenge from a policy perspective. With many individuals in rural America living in lower socioeconomic circumstances and potentially isolated from their neighbors, how best to tackle rural crime remains an open question. Rather than assume that criminological theories, like social disorganization, explain crime in both urban and rural areas, criminologists should continue to explore what informal social control looks like in the rural context and find the most effective ways to increase it.

About the Authors

Matthew D. Moore, PhD, is an assistant professor in the Department of Sociology and Criminal Justice, Grand View University. His research interests are in the areas of cross-national criminology, social capital, and suicide. His recent publications have appeared in Crime & Delinquency and Social Indicators Research.

Molly Sween, PhD, is an assistant professor in the Department of Criminal Justice, Weber State University. Her research interests are in the areas of juvenile delinquency, criminological theory, and social inequality.

References

Akers, R. L., & Sellers, C. E. (2009). Criminological theories: Introduction, evaluation, and application. New York, NY: Oxford University Press.

Allen, J., & Cancino, J. M. (2012). Social disorganization, Latinos and juvenile crime in the Texas-Mexico borderlands. Journal of Criminal Justice, 40(2), 152–163.

Barnett, C., & Mencken, F. C. (2002). Social disorganization theory and the contextual nature of crime in nonmetropolitan counties. Rural Sociology, 67(3), 372–393.

Blau, J. B., & Blau, P. M. (1982). The cost of inequality: Metropolitan structure and violent crime. American Sociological Review, 47(1), 114–129.

Bouffard, L. A., & Muftić, L. R. (2006). The “rural mystique”: Social disorganization and violence beyond urban communities. Western Criminology Review, 7(3), 56–66.

Bruinsma, G. J. N., Pauwels, L. J. R., Weerman, F. M., & Bernasco, W. (2013). Social disorganization, social capital, collective efficacy and the spatial distribution of crime and offenders. The British Journal of Criminology, 53, 942–963.

Bursik, R. J., & Grasmick, H. (1993). Neighborhoods and crime. New York, NY: Lexington Books.

Bursik, R. J., & Webb, J. (1982). Community change and patterns of delinquency. American Journal of Sociology, 88, 24–42.

Cancino, J. M. (2005). The utility of social capital and collective efficacy: Social control policy in nonmetropolitan settings. Criminal Justice Policy Review, 16(3), 287–318.

Deller, S. C., & Deller, M. A. (2010). Rural crime and social capital. Growth and Change, 41(2), 221–275.

Gastil, R. D. (1971). Homicide and a regional culture of violence. American Sociological Review, 36, 412–427.

Guerette, R. T., & Bowers, K. J. (2009). Assessing the extent of crime displacement and diffusion of benefits: A review of situational crime prevention evaluations. Criminology, 47(4), 1331–1368.

Jobes, P. C., Barclay, E., Weinand, H., & Donnermeyer, J. F. (2004). A structural analysis of social disorganization and crime in rural communities in Australia. The Australian and New Zealand Journal of Criminology, 37(1), 114–140.

Kapsis, R. E. (1978). Residential segregation and delinquency. Criminology, 15, 459–486.

Kaylen, M. T., & Pridemore, W. A. (2011). A reassessment of the association between social disorganization and youth violence in rural areas. Social Science Quarterly, 92(4), 978–1001.

Kaylen, M. T., & Pridemore, W. A. (2012). Systematically addressing inconsistencies in the rural social disorganization and crime literature. International Journal of Rural Criminology, 1(2), 134–152.

Kaylen, M. T., & Pridemore, W. A. (2013). Social disorganization and crime in rural communities: A first direct test of the systemic model. The British Journal of Criminology, 53, 905–923.

Klaus, P. (2004). Crime and the nation’s households, 2002. Washington, DC: U.S. Department of Justice.

Kornhauser, R. R. (1978). Social sources of delinquency. Chicago, IL: University of Chicago Press.

Li, Y. Y. (2011). Social structure and informal social control in rural communities. International Journal of Rural Criminology, 1(1), 63–88.

Lott, J. R., Jr., & Whitley, J. (2003). Measurement error in county-level UCR data. Journal of Quantitative Criminology, 19(2), 185–198.

Maltz, M. D., & Targonski, J. (2002). A note on the use of county-level UCR data. Journal of Quantitative Criminology, 18(3), 297–318.

Markowitz, F. E., Bellair, P. E., Liska, A. E., & Liu, J. (2001). Extending social disorganization theory: Modeling the relationships between cohesion, disorder, and fear. Criminology, 39(2), 293–320.

Mustaine, E. E., Tewksbury, R., Huff-Corzine, L., Corzine, J., & Marshall, H. (2014). Community characteristics and child sexual assault: Social disorganization and age. Journal of Criminal Justice, 42(2), 173–183.

Nisbett, R. E., & Cohen, D. (1996). Culture of honor: The psychology of violence in the south. Boulder, CO: Westview.

Osgood, D. W. (2000). Poisson-based regression analysis of aggregate crime rates. Journal of Quantitative Criminology, 16(1), 21–43.

Osgood, D. W., & Chambers, J. M. (2000). Social disorganization outside the metropolis: An analysis of rural youth violence. Criminology, 38(1), 81–116.

Osgood, D. W., & Chambers, J. M. (2003). Community correlates of rural youth violence. OJJDP Juvenile Justice Bulletin, May 2003.

Petee, T. A., & Kowalski, G. S. (1993). Modeling rural violent crime rates: A test of social disorganization theory. Sociological Focus, 26, 87–89.

Rocque, M., Posick, C., Barkan, S. E., & Paternoster, R. (2014). Marriage and county-level crime rates: A research note. Journal of Research in Crime and Delinquency. doi:0022427814547113.

Sampson, R. J. (1995). The community. In J. Q. Wilson and J. Petersilia (Eds.), Crime (pp. 193–216). San Francisco, CA: ICS.

Sampson, R. J. (2012). Great American city: Chicago and the enduring neighborhood effect. Chicago, IL: University of Chicago Press.

Sampson, R. J., & Groves, W. B. (1989). Community structure and crime: Testing social-disorganization theory. American Journal of Sociology, 94(4), 774–802.

Sampson, R. J., Raudenbush, S. W., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277(15), 918–924.

Shaw, C. R., & McKay, H. D. (1942). Juvenile delinquency and urban areas. Chicago, IL: University of Chicago Press.

U.S. Department of Agriculture. (2013). Rural-urban continuum codes. U.S. Department of Agriculture. Retrieved from http://www.ers.usda.gov/data-products/rural-urban-continuum-codes.aspx

Warner, B. D., Beck, E., & Ohmer, M. L. (2010). Linking informal social control and restorative justice: Moving social disorganization theory beyond community policing. Contemporary Justice Review, 13(4): 355–369.

Weisburd, D., Wyckoff, L. A., Ready, J., Eck, J. E., Hinkle, J. C., & Gajewski, F. (2006). Does crime just move around the corner? A controlled study of spatial displacement and diffusion of crime control benefits. Criminology, 44(3), 549–592.

Weisheit, R. A., Falcone, D. N., & Wells, L. E. (1994). Rural crime and rural policing. National Institute of Justice: Research in Action, September 1994, 1–15.

Wells, L. E., & Weisheit, R. A. (2004). Patterns of rural and urban crime: A county-level comparison. Criminal Justice Review, 29(1), 1–22.

Wiersema, B., Loftin, C., & McDowall, D. (2000). A comparison of supplementary homicide reports and national vital statistics system homicide estimates for U.S. counties. Homicide Studies, 4(4), 317–340.

Wilkinson, K. P. (1984a). Rurality and patterns of social disruption. Rural Sociology, 48, 23–36.

Wilkinson, K. P. (1984b). A research note on homicide and rurality. Social Forces, 63, 445–452.

Witherspoon, D., & Ennett, S. (2011). An examination of social disorganization and pluralistic neighborhood theories with rural mothers and their adolescents. Journal of Youth and Adolescence, 40(9), 1243–1253.

Xie, M., & McDowall, D. (2008). The effects of residential turnover on household victimization. Criminology, 46, 539–576.

Appendix

Correlation Tables for Variables Used in the Model

1

2

3

4

5

6

7

8

9

10

11

12

13

14

1. Violent Crime

1.00

2. Murder

.340

1.00

3. Rape

.486

.040

1.00

4. Robbery

.692

.150

.270

1.00

5. Assault

.966

.286

.385

.525

1.00

6. Weapons

.567

.132

.259

.554

.506

1.00

7. Simple Assault

.690

.175

.356

.607

.630

.695

1.00

8. Residential Instability

.198

.021

.082

.161

.194

.155

.233

1.00

9. Ethnic Heterogeneity

.280

.106

.018

.257

.171

.181

.146

.077

1.00

10. Family Disruption

.212

.096

.042

.234

.180

.191

.217

.122

.622

1.00

11. Poverty Rate

.050

.084

-.058

.100

.029

.068

.025

-.065

.509

.676

1.00

12. Population At Risk

-.188

-.039

-.113

-.128

-.181

-.137

-.213

-.150

-.108

-.233

-.078

1.00

13. Unemployment Rate

.153

.051

.051

.138

.138

.159

.166

.037

.345

.532

.517

-.247

1.00

14. Population Density

.158

.022

.087

.160

.139

.121

.169

.165

.074

.106

.045

-.095

.085

1.00

 

 

 

OJJDP Home | About OJJDP | E-News | Topics | Funding
Programs | State Contacts | Publications | Statistics | Events