<%BANNER%>

A Structural Examination of Rural Crime in the Midwest, 1980-2000

Permanent Link: http://ufdc.ufl.edu/UFE0019749/00001

Material Information

Title: A Structural Examination of Rural Crime in the Midwest, 1980-2000 What Has Changed Over Time and Why?
Physical Description: 1 online resource (144 p.)
Language: english
Creator: Hays, Stephanie Ann
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: criminology, farming, rural
Criminology, Law and Society -- Dissertations, Academic -- UF
Genre: Criminology, Law, and Society thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Recent research has documented an increase in rural crime, particularly violent crime, over the past two decades. While the relationship between community structure and crime has received a large amount of attention in criminology, most of this research has focused only on urban areas. This study builds on previous works by not only examining the relationship between structural factors and arrests in rural areas, but also by taking into account structural changes that have had a profound effect on rural life over the past few decades (i.e., the industrialization of farming). Using cross-sectional time series regression and pooled time series regression, I estimate the effects of changes in social disorganization and the industrialization of farming on criminal arrests in rural, Midwest counties from 1980 to 2000. Overall, the findings suggest that the industrialization of farming, particularly changes in the number of farms in a county, has had a significant impact on arrests.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Stephanie Ann Hays.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Lanza-Kaduce, Lonn M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0019749:00001

Permanent Link: http://ufdc.ufl.edu/UFE0019749/00001

Material Information

Title: A Structural Examination of Rural Crime in the Midwest, 1980-2000 What Has Changed Over Time and Why?
Physical Description: 1 online resource (144 p.)
Language: english
Creator: Hays, Stephanie Ann
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: criminology, farming, rural
Criminology, Law and Society -- Dissertations, Academic -- UF
Genre: Criminology, Law, and Society thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Recent research has documented an increase in rural crime, particularly violent crime, over the past two decades. While the relationship between community structure and crime has received a large amount of attention in criminology, most of this research has focused only on urban areas. This study builds on previous works by not only examining the relationship between structural factors and arrests in rural areas, but also by taking into account structural changes that have had a profound effect on rural life over the past few decades (i.e., the industrialization of farming). Using cross-sectional time series regression and pooled time series regression, I estimate the effects of changes in social disorganization and the industrialization of farming on criminal arrests in rural, Midwest counties from 1980 to 2000. Overall, the findings suggest that the industrialization of farming, particularly changes in the number of farms in a county, has had a significant impact on arrests.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Stephanie Ann Hays.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Lanza-Kaduce, Lonn M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0019749:00001


This item has the following downloads:


Full Text

PAGE 1

A STRUCTURAL EXAMINATION OF RURA L CRIME IN THE MIDWEST, 1980: WHAT HAS CHANGED OVER TIME AND WHY? By STEPHANIE ANN HAYS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007 1

PAGE 2

2007 Stephanie Ann Hays 2

PAGE 3

To my family for always being there 3

PAGE 4

ACKNOWLEDGMENTS I would like to thank my entire committee, Dr. Lonn Lanza-Kaduce, Dr. Mark Brennan, Dr. Eve Brank, and Dr. Charles Frazier for all of their help and support in completing this dissertation. Lonn was a great department chair who was always willing to help with anything and everything, and he gladly exchanged random tales of Iowa whenever I was homesick. Eve was always there as a great ment or in the classroom but also as a friend. Chuck provided lots of insight despite coming onto the committee at the last minute. I especially need to thank Mark. He deserves a lot of credit for this dissertation. It would not have been possible without him. Not only did he put up with me and provide endless amounts of assistance, but he also was the one who re-inspired me to take a second look at communities and rural areas. I will always cherish the memories of Ireland and Amsterdam. I also need to thank my parents, Larry and Glenda Hays, and my brother Bryan for supporting me in all of my decisions throughout the years and for always encouraging me to continue my education. Although they are probably as relieved as I am that I am finally finishing school and will no longer be a professional student. I also need to thank all of my fellow gr aduate students along the way that provided friendship and the much needed social breaks from our graduate studies. I would especially like to thank Amy Reckdenwald, Kristin Johnson, A ndrea Schoepfer, Dave Khey, Matt Nobles, Rohald Meneses, and Wesley Jennings. I will always have some great memories of our times together both in and out of the classroom. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES.......................................................................................................................10 LIST OF ABBREVIATIONS........................................................................................................11 ABSTRACT...................................................................................................................................12 CHAPTER 1 INTRODUCTION................................................................................................................. .13 Importance of Rural Crime.....................................................................................................1 3 Rural Crime Today.............................................................................................................. ...14 Why the Midwest?............................................................................................................... ...16 Research Questions............................................................................................................. ....16 Summary.................................................................................................................................17 2 SOCIAL DISORGANIZATION THEORY...........................................................................18 Social Disorganization in Urban Areas..................................................................................18 Urban Empirical Findings...............................................................................................20 Summary..........................................................................................................................22 Social Disorganization in Rural Areas...................................................................................23 Rural Empirical Findings................................................................................................24 Summary..........................................................................................................................26 Conclusion..............................................................................................................................27 3 THE INDUSTRIALIZA TION OF FARMING......................................................................28 The Farm Crisis and Industrialization in the Midwest...........................................................28 The Goldschmidt Hypothesis.................................................................................................30 Empirical Findings on Industria lization and Community Decay...........................................30 Conclusion..............................................................................................................................33 4 DATA AND METHODOLOGY...........................................................................................34 Sources of the Data.................................................................................................................34 Unit of Analysis............................................................................................................... .......34 Sample....................................................................................................................................35 Sub-Sample.............................................................................................................................36 Measures.................................................................................................................................36 5

PAGE 6

Dependent Variables.......................................................................................................36 Independent Variables.....................................................................................................37 Social disorganization indicators.............................................................................37 Industrialization of farmi ng indicators (scale).........................................................39 Industrialization of farming indicators (structure)...................................................41 Controls....................................................................................................................... ....42 Descriptives................................................................................................................... .........43 Multicollinearity.....................................................................................................................44 Analytical Plan................................................................................................................ ........45 Cross-Sectional Analyses................................................................................................45 Pooled Cross-Sectional Time Series Analyses................................................................46 5 RESULTS...................................................................................................................... .........51 Cross-Sectional Analyses.......................................................................................................51 Pooled Time-Series Analyses.................................................................................................52 Summary.................................................................................................................................54 6 DISCUSSION AND CONCLUSION....................................................................................64 Discussion...............................................................................................................................64 Limitations and Future Research............................................................................................66 Conclusion..............................................................................................................................67 APPENDIX A MIDWEST STATES..............................................................................................................6 8 B SUMMARY OF STUDIES CITED EXAM INING STRUCUTRAL CORRELATES OF CRIME IN URBAN AREAS.................................................................................................69 C SUMMARY OF STUDIES CITED EXAM INING STRUCTURAL CORRELATES OF CRIME IN RURAL AREAS..................................................................................................71 D SUMMARY OF STUDIES EXAMINI NG INDUSTRIALIZED FARMING AND COMMUNITY WELL-BEING..............................................................................................73 E RURAL URBAN CONTIUUM CODES (BEALE CODES)................................................82 F MIDWEST METRO AND NONMETRO COUNTIES.........................................................83 G MAPS OF FINAL SAMPLE (N = 596).................................................................................87 H CORRELATION TABLES....................................................................................................99 I SUB SAMPLE ANALYSES................................................................................................124 LIST OF REFERENCES.............................................................................................................129 6

PAGE 7

7 BIOGRAPHICAL SKETCH.......................................................................................................144

PAGE 8

LIST OF TABLES Table page 4-1 Descriptive statistics (means with standard deviations in parentheses).............................48 4-2 T-test scores for change from 1980 to 2000 ( N = 591)......................................................49 4-3 Variance inflation factor values for in dependent variables in cluded in all models...........50 5-2 1990 cross-sectional negative binomial and ZINB regression results (and z-scores).......57 5-3 2000 cross-sectional negative binomial and ZINB regression results (and z-scores).......59 5-4 Means and overall, between and within sa mple standard devia tions for indicators included in the time-series models ( N = 1788, n = 596, T = 3).........................................61 5-5 Pooled cross-sectional time series negative binomial regression results (and zscores), 1980-2000 ( N = 1788, n = 596, T = 3).................................................................62 B-1 Summary of articles cited examining structural correlates of crime in urban areas..........70 C-1 Summary of articles cited examining structural correlates of crime in rural areas...........72 D-1 Summary of studies examin ing the industrialization of farming and community wellbeing by year of publication..............................................................................................74 E-1 2003 Beale Codes........................................................................................................... ...82 E-2 1983 and 1993 Beale Codes...............................................................................................82 F-1 Number of metro and nonme tro counties by state and year..............................................83 H-1 Correlation matrix of variab les total sample, all years.................................................100 H-2 Correlation matrix of 19 80 variables total sample........................................................103 H-3 Correlation matrix of 19 90 variables total sample........................................................106 H-4 Correlation matrix of 20 00 variables total sample........................................................109 H-5 Correlation matrix of variab les sub sample, all years...................................................112 H-5 Continued.................................................................................................................. .......113 H-6 Correlation matrix of 19 80 variables sub sample.........................................................115 H-7 Correlation matrix of 19 90 variables sub sample.........................................................118 8

PAGE 9

9 H-8 Correlation matrix of 20 00 variables sub sample.........................................................121 I-1 Descriptive statistics (mean s with standard deviations in parentheses) for sub sample..124 I-2 Variance inflation factor values for inde pendent variables include d in all models, sub sample......................................................................................................................... .....125 I-3 Means and overall, between and within sa mple standard devia tions for indicators included in the time-series models, sub sample ( N = 1557, n = 519, T = 3)...................126 I-4 Pooled cross-sectional time series negative binomial regression results (and zscores), 1980-2000, sub sample ( N = 1557, n = 519, T = 3)...........................................127

PAGE 10

LIST OF FIGURES Figure page 2-1 Burgesss zones for city growth.........................................................................................18 2-2 Basic theoretical model of social disorganization theory..................................................19 G-1 Indiana.................................................................................................................... ............88 G-2 Iowa....................................................................................................................... .............89 G-3 Kansas..................................................................................................................... ...........90 G-4 Michigan................................................................................................................... .........91 G-5 Minnesota.................................................................................................................. .........92 G-6 Missouri................................................................................................................... ..........93 G-7 Nebraska................................................................................................................... .........94 G-8 North Dakota............................................................................................................... .......95 G-9 Ohio....................................................................................................................................96 G-10 South Dakota.............................................................................................................. ........97 G-11 Wisconsin................................................................................................................. ..........98 10

PAGE 11

LIST OF ABBREVIATIONS Beta n Number in a sub sample N Total number in a sample OMB Office of Management and Budget p Probability PIZA Population-Interaction Zones for Agriculture SD Standard deviation SE Standard error t Computed value of t test UCR Uniform Crime Report z A standard score; difference between one value in a distributi on and the mean of a distribution divided by the SD ZINB Zero-inflated negative binomial ZIP Zero-inflated Poisson 11

PAGE 12

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy A STRUCTURAL EXAMINATION OF RURA L CRIME IN THE MIDWEST, 1980: WHAT HAS CHANGED OVER TIME AND WHY? By Stephanie Ann Hays August 2007 Chair: Lonn Lanza-Kaduce Major: Criminology, Law and Society Recent research has documented an increase in rural crime, particularly violent crime, over the past two decades. While th e relationship between commun ity structure and crime has received a large amount of attention in crimi nology, most of this research has focused only on urban areas. This study builds on previous works by not only examining the relationship between structural factors and arre sts in rural areas, but also by taking into account structural changes that have had a profound effect on rural life over the past few decades (i.e., the industrialization of farming). Using cross-sectional time series re gression and pooled time series regression, I estimate the effects of changes in social disorg anization and the industr ialization of farming on criminal arrests in rural, Mi dwest counties from 1980 to 2000. Overall, the findings suggest that the industrialization of farming, pa rticularly changes in the number of farms in a county, has had a significant impact on arrests. 12

PAGE 13

CHAPTER 1 INTRODUCTION Importance of Rural Crime Rural crime has largely been ignored by criminologists ev en though 25% of the U.S. population lives in rural areas with populations less than 2,500 (Weisheit & Donnermeyer, 2000). According to the 2000 Census, 56 m illion people reside in rural areas1 (Brown & Swanson, 2003). This exceeds the total population of all but 22 of the worlds 200 nation states (Brown & Swanson, 2003). But rural crime can no longer be ignored. Out of the 100 highest ranking counties on rates of homicide, 71 are rural (Kposowa, Breault, & Harrison, 1995). While research on rural crime is growing, most of the research has focused largely on drug use (Diala, Muntaner, & Walrath, 2004; Donne rmeyer, Barclay, & Jobes, 2002; Weisheit & Fuller, 2004), domestic violence (Davis, Tayl or, & Furniss, 2001; Krishnan, Hilbert, & VanLeeuwen, 2001; Websdale, 1998; Websdale & Johnson, 1998), and community policing (Jobes, 2003, 2002; Liederbach & Frank, 2003; OShea, 1999;Weisheit, Wells, & Falcone, 1994). Despite this increased attention, these st udies still only make up a small portion of the overall criminological research. The current study shifts the focus to rural crime. It examines the relationship of rural crime with social disorganization and changes in land use and farm structure (t he industrialization of farming). The goal is to estimate the influence of shifts in structural predictors on the increases (or declines) in violent and property crime across ru ral locales. While structural theories of crime have been prominent in the criminology liter ature, these have primarily focused on urban2 areas, 1 According to the Census, rural is territory, population and housing units not classified as urban. 2 The Census defines urban as all territory, population and housin g units in urbanized areas and in places of more than 2,500 persons outside of urbanized areas. An urbanized area (UA) is an area consisting of a central place(s) and adjacent territory with a general population density of at leas t 1,000 people per square mile of land area that together has a minimum residential population of at least 50,000. 13

PAGE 14

and it is uncertain whether such theories ade quately explain rural crime. Such macro-level theories need to be examined in rural areas to ensure that they are general theories of crime and not urban specific theories (Weisheit & Donnermeyer, 2000). Recently in criminology there has been an increa sing interest in contextual studies. Place is one example of a context, and rural areas may be a different context from urban areas. Cebulak (2004) claims the context of rura l crime, its causes and its charac teristics, are so different than for urban crime, we need a separate set of theori es to account for rural crime and justice (p. 72). Some types of crime, such as theft of farm an imals or equipment and wildlife crimes, are limited to only rural areas (Cebulak, 2004; Weisheit & Wells, 1999). Fu rthermore, there are unique features of rural communities that may influe nce rural crime such as: physical distance and isolation, more informal social control, low mobility and density, higher density of acquaintanceship, mistrust of government, reluct ance to seek outside as sistance, and factory farms and processing plants (Weisheit & D onnermeyer, 2000; Weisheit & Wells, 1999). For more on the effect of processing plants on ru ral communities and crime see Broadway (1990). Rural Crime Today The image of rural areas being free of crime ha s persisted despite evidence to the contrary throughout history. There was the la wless West at the end of the 19th Century. There were lynch mobs and the Ku Klux Klan in the South. Ther e were the moonshiners during the Prohibition (Donnermeyer, 1994). Contrary to common misconceptions, rural cr ime/deviance is an increasingly important and relevant issue. Nonmetropolitan3 crime rates have been increasing sin ce 1984 (Rephann, 1999). The Uniform Crime Report (UCR) crime rate in 1991 for rural ar eas exceeded the 1966 3 According to the Census, nonmetropolitan refers to th e area and population not located in any metropolitan area (MA). 14

PAGE 15

rate for urban areas when the war on crime was declared by Congress (Donnermeyer, 1994). Furthermore, while violent crime rates in large ci ties have declined since the early 1990s, rural violent crime rates have been increasing (Weisheit & Donnermeyer, 2000). Between 1991 and 1997 urban violent crime rates decreased by 531.8 per 100,000, while ru ral violent rates increased by 37.9 per 100,000 (Weisheit & Donnermeyer, 2000). Suicide rates are also higher in rural areas (Butterfield, 2005). In the most rural counties, the incidence of suicide with a gun is greater than the incidence of murder with guns in major cities (Butterfield, 2005). From 1989 to 1999, the ri sk of dying from a gunshot was the same in rural and urban areas. The difference was who pulled the trigger (Butterfield, 2005). The types of rural crime and deviance are also unique. Driving under the influence (DUIs) are more common in rural areas, and rura l youth are more likely to use cigarettes or smokeless tobacco (Weisheit & Donnermeyer, 2000). In addition over the past 20 years, rural youth alcohol use has matched or exceeded that of urban youth. Similarly, nonmetropolitan 12th graders in 1995 had higher use rates for crack co caine, stimulants, barbitura tes, and tranquilizers than did their metropolitan counterparts (Wei sheit & Donnermeyer, 2000). Drug manufacturing, particularly for methamphetamine, is of increas ing concern in rural areas. Missouri had more methamphetamine lab seizures than any other stat e in 1997, with most of the seizures occurring in rural areas (Weisheit & Donne rmeyer, 2000). Such findings were repeated in research that showed there were 300 times more methamphetamine lab seizures in Iowa in 1999 than in New York and New Jersey combined (Eagan, 2002). 15

PAGE 16

Why the Midwest? The current study is limited to the Midwest4 region for two primary reasons. First, from 1985-1991, violent and property crime in rural areas in the Midwest increased more so than in urban areas (Donnermeyer, 1994). During this time, aggravated assault increased nearly 50%, and the homicide rate in rural Indiana in 1991 was slightly highe r than the homicide rate for urban Indiana (Donnermeyer, 1994) In 1991, total crime index ra tes for rural areas in the Midwest ranged from a low of 954.5 per 100,000 in North Dakota to a high of 3012.5 per 100,000 in Michigan. Second, the study was also limited to the Midwes t due to the interest in looking at the relationship between industriali zation of farming and crime. Th e farm crisis of the 1980s affected certain types of farms more than othe rs (Brasier, 2005). A significant number of these farms were concentrated in the Midwest and Pl ains (Leistritz & Eckstrom, 1988). Furthermore, the Midwest states analyzed here have all thre e scales of farming as identified by Wimberley (1987) smaller family farming, larger family farming, and industrial farming (Crowley & Roscigno, 2004). Research Questions The purpose of this study is to explore in more depth the relationship between social disorganization, industrialization of farming, and offending in rura l areas. By focusing on these dynamics at the structural level, a number of ke y research questions are addressed in this work, including the following: 1. Has rural crime (as measured by arrests) changed in the Midwest over the past two decades? 4 See Appendix A for a list of the Midwest states 16

PAGE 17

17 2. What is the impact of social disorgani zation on arrests in rural counties in the Midwest? 3. What is the impact of the industrializ ation of farming on arrests in rural counties in the Midwest? Summary This study seeks to fill a gap in the existi ng criminological litera ture both by focusing on rural crime (which is understud ied) and by conducting a structural analysis that incorporates important changes in rural life (industrialization of farming). It is one of only a few empirical studies of rural crime at the struct ural level. It is one of only a few rural crime analyses that are longitudinal and informed by social disorganizati on theory. This dissertation will consist of six chapters. Chapters 2 and 3 provide the theoretical background and literature review, as well as, the basis for the hypotheses. Chapter 4 describes the data and statistical procedures utilized. Chapter 5 provides the results from the multivariate analyses. Finally, Chapter 6 provides concluding remarks and directions for future rese arch, as well as suggest s application and policy implications.

PAGE 18

CHAPTER 2 SOCIAL DISORGANIZATION THEORY Social Disorganization in Urban Areas In the past 20 years there has been a renewe d interest in social disorganization theory. Social disorganization theory has its roots in Chicago as it rapidly urbanized. Park and Burgess (1925) emphasized the importance of looking at natural areas and th eir characterist ics. Areas are considered to be functioning and changing orga nisms. Burgess (1925) suggested that cities expand and grow outward in concen tric circles from the core of th e city zone one. Zone two is the transition zone. It is gene rally the oldest and poorest zone The third zone consists of workers homes. The fourth zone is the residential zone and c onsists of single family housing, and finally the fifth zone is the commuter zone or suburbs. Mobility is constantly occurring in all of these zones as new people move to the city an d current residents try to move outward into a different zone. The zones are displayed in Figure 2-1. Zone 1 City Core Zone 2 Transition Zone Zone 3 -Workers Zone 4 -Residential Zone 5 -Commute r Figure 2-1. Burgesss zones for city growth 18

PAGE 19

In Juvenile Delinquency and Urban Areas, Shaw and McKay (1942) examined juvenile delinquency in Chicago. Their main argument was that 1) poor economic status, 2) population heterogeneity, and 3) residential mobility lead to social disorganization. So cial disorganization in turn leads to breakdowns in c onventional attachments and informal and formal control and therefore results in more crime. More recently, it has been suggested that family disruption (Sampson, 1987; Sampson & Groves, 1989) an d population size and density (Mayhew & Levinger, 1976) also contribute to social disorganizati on. The key theoretical concepts of social disorganization theory are displayed in Figure 2-2. Several studies have examined social disorganization theory and have found support for the theory in urban areas (Kposowa, Breault, & Harrison, 1995; Lee, Maume, & Ousey, 2003; Petee & Kowalski, 1993; Sampson, 1991; Sampson & Groves, 1989). These studies are discussed below.1 Residential Mobility Family Disruption Population Size / Density Population Heterogeneity Low Economic Status Social Disorganization Less Attachment And Control Crime Figure 2-2. Basic theoretical model of social disorganization theory 1 See Appendix B for a summary of urban social disorganization studies cited here. 19

PAGE 20

Urban Empirical Findings Sampson and colleagues have conducted severa l studies on social di sorganization theory. Sampson (1987) examined the relationship between family disruption and crime. Using 1980 homicide and robbery rates for 171 cities, he f ound family disruption, and housing density to be related to juvenile robbery, a nd family disruption, population size, and housing density were related to adult robber y. Family disruption was also found to be related to adult homicide. Furthermore, family disruption and population size were also related to black juvenile homicide, and region and city size were associ ated with black adult homicide. Sampson and Groves (1989) expanded the rese arch on social disorganization theory. Unlike previous studies, they measured the leve ls of social organization within communities besides using the standard structural indicators of social disorganization. Based on data from the 1982 British Crime Survey ( N =10,905 individuals, N = 238 localities), they found family disruption, urbanization, and ethnic heterogeneity to be associated with attachment to peers, specifically with more disorderly peer groups Urbanization was negatively associated with friendship networks, and socioeconomic status was positively related to organizational participation. Further analysis revealed that th e structural measures of social disorganization were mediated by the community organization measures. In other words, community disorganization (as measured by sparse local friendship networks, unsupervised teenage peer groups, and low organization participation) account s for much of the effect of socioeconomic status, residential mobility, family disruption, and ethnic/population hete rogeneity on crime. In an extension of the previous study, Sampson (19 91) analyzed data from the 1984 British Crime Survey ( N = 11,030 individuals, N = 526 polling districts) and found that residential stability has a direct effect on community-based social ties, or attachments, whic h in turn increases the level of community cohesion. 20

PAGE 21

In yet another study, Sampson, Raudenbush, and Ea rls (1997) tested whether concentrated disadvantage (low economic status) and resident ial instability/mobility decrease collective efficacy and whether collective efficacy in tu rn explains the relationship between neighborhood disadvantage and crime. Collective efficacy is defined as social cohesion among neighbors combined with their willingne ss to intervene on behalf of the common good (p. 918). They interviewed 8,782 residents of 343 neighborhood clusters in Chica go. The results indicated that the effects of concentrated disadvantage and resi dential instability on violence were mediated in a large part by collective efficacy. Consistent with social disorganization th eory, Kposowa, Breault, and Harrison (1995) examined crime in counties with a population larg er than 100,000 ( N = 408) and found poverty (low economic status) to be a significant predictor of violent crime. Church membership (attachment) and divorce (family disruption) were significant predictors of property crime. When examining crime in all counties ( N = 3,076), the strongest predictor of property crime was urbanity, but percent black and percent Hispanic (population he terogeneity), population change, and unemployment (low economic status) were also significant. Predictors of violent crime in the total county sample were percent black and per cent Hispanic (population heterogeneity), church membership (attachment), and population densit y. When examining homicide in the large counties (N = 408), they found percent black (population heteroge neity), Gini coefficient2, divorce rate (family disruption), and population change to be the strongest predictors. Contrary to social disorganization theory though, th ey found poverty (low economic status) to be significantly related to less homicide. 2 The Gini coefficient is used by the federal government to document income inequality in the United States. The measure represents the proportion of the population with different income categories. The index ranges from 0 to 1 where 0 reflects complete equality and 1 represents complete inequality (Coulter, 1989). 21

PAGE 22

Lee, Maume, and Ousey (2003) also examined the relationship between socioeconomic disadvantage and poverty on the homicide ra te average from 1990-1992 in 778 metropolitan counties. Unlike Kposowa and colleagues (1995), Lee and colleagues (2003) did find poverty concentration to be significant and positively re lated to homicide. Disadvantage was also found to be significant and positively related to homicide. Land, McCall, and Cohen (1990) also examined the structural correlates of homicide rates across time 1960, 1970, and 1980 and across space cities, SMSAs,3 and states. While not specifically testing social disorgan ization theory, they did find suppor t for the indicators of social disorganization. Resource depr ivation (low economic status) was associated with higher homicide rates across all the time periods and locations. In addition, popu lation structure an index of population size and densit y and percent divorced (family disruption) were also related to homicide across most of the models. Summary Overall, there is much support for social disorganization theor y. There are, however, inconsistencies in terms of which social disorgan ization indicators are significant and for which types of offenses. Furthermore, some studi es have findings opposite of what social disorganization theory would suggest. For instance when looking at the predictors of homicide in metropolitan counties, Kposowa and colleagues (1995) found poverty (low economic status) to be significant but in the negative direction; however, Lee an d colleagues (2003) found poverty concentration (low economic status) to have a significant positive effect on homicide in metropolitan counties. 3 This acronym has changed over time. It was SMA beginning in 1949, changed to SMSA in 1959, and then changed to MSA in 1983. MSA refers to a metropolitan statistical area. It is a geographic entity defined by the federal Office of Management and Budget for use by federal statistical ag encies, based on the concept of a core area with a large population nucleus, plus adjacent communities having a high degree of economic and social integration with that core. 22

PAGE 23

Social Disorganization in Rural Areas Most research on social disorganization has focused on urban areas; however, social disorganization theory can also be applied to rural areas. Th e structural conditions usually associated with social disorganization are not unique to urban areas. In fact, nonmetropolitan poverty rates have exceeded metropolitan poverty rates every year since poverty was first officially measured in the 1960s (Rural poverty at a glance, 2004). Throughout the 1980s, the rural poverty rate exceeded the urban rate by over five percent (Brown & Swanson, 2003). In fact, 500 rural counties have consis tently had poverty rates in excess of 20% over the past four decades (Brown & Swanson, 2003). In addition, the 2003 unemployment rate was 5.8% in nonmetropolitan areas and 6.0% in metropolitan areas (Rural America at a glance, 2004). Albrecht and colleagues (2000) examined how pove rty levels in rural America have been affected by industrial transformation. Using 1990 census data on 2,390 nonmetropolitan counties, they argue that Wilsons (1987) model for the inner city underclass can be used to understand increased levels of rura l poverty and the growth of th e rural underclass (for more on rural ghettos see Davidson, 1996). Population mobility is also an issue in rural areas. While 74% of the 2,303 nonmetropolitan 1993 counties gained in populat ion from 1990 to 2000, there were still widespread losses in the Great Plains4 and Western Corn Belt5 regions, and large segments of the Heartland continue to lose people and institutions (Brown & Swanson, Ch 1, 2003) Over 1,000 nonmetropolitan counties have lost population since 2000, prim arily counties in the Great Plains, but there are also fast growing nonmetropolitan recrea tional counties in the South and 4 The Great Plains is a geographically and environmentally defined region covering parts of ten states: Montana, North Dakota, South Dakota, Nebraska, Wyoming, Kansas, Colorado, Oklahoma, Texas, and New Mexico. 5 The Corn Belt is an agricultural region of the central Unite d States primarily in Iowa and Illinois but also including parts of Indiana, Minnesota, South Dakot a, Nebraska, Kansas, Missouri, and Ohio. 23

PAGE 24

West and the growth of the Hispanic populat ion has contributed to nonmetropolitan county population growth in the West, South, and Midwest (Rural America at a glance, 2004). Population turnaround in rural counties has been li nked to a variety of so cial problems including problems in education, community solidarity, heath care, social welfar e, and crime (Price & Clay, f eaded by a alered w ith those living in central cities and suburban areas. Rura pact cial 1980). Other features of social disorg anization are also present in ru ral areas. Seventeen percent o nonmetropolitan residents are minorities, and 15 % of nonmetropolitan families are h single female (Jolliffe, 2003). Snyder and McLaughlin (2004) found that poverty in nonmetropolitan areas closely resemb les that in central cities. Th e risk of poverty for fem headed families and subfamilies with children is significantly higher for those living in nonmetropolitan areas compa l Empirical Findings A few empirical studies of so cial disorganization in rural areas have provided some support for social disorganization theory6 (Barnett & Mencken, 2002; Kposowa, Breault, & Harrison, 1995; Lee, Maume, & Ousey, 2003; Osgood & Chambers, 2000; Petee & Kowalski, 1993). Petee and Kowalski (1993) te sted social disorganization th eory on violent crime rates in 630 rural counties from 1979-1980. They found resident ial mobility to have the greatest im on violent crime followed by single-parent house holds (family disruption), and then ra heterogeneity (population heterogeneity). Ba rnett and Mencken (2002) applied social disorganization theory to vi olent and property crime rates circa 1990 in 2,254 nonmetropolitan counties. They found resource disadvantage (low economic status) to have a significant positive 6 See Appendix C for a summary of rural social disorganization studies cited here. 24

PAGE 25

effect to property hen ulation as not ey did, however, find poverty (low e hen exam theory they did find re sidential instability (mob n on violent crime, and population change wa s significant and positively related crime. While Kposowa, Breault, and Harrison (1995) did not specifically test social disorganization theory, they did incorporate standard measures of social disorganization w examining the structural correlates of crime in rural counties ( N = 1,681). They found pop change to be significant and pos itively related to both violen t and property crime. Church membership (attachment), divorce rate (fam ily disruption), and percent Native American (population heterogeneity) were also found to be significant predictors of violent crime, and percent Hispanic (popula tion heterogeneity) was found to be a significant predictor of property crime. Contrary to social disorganization th eory though, poverty (low economic status) w found to be a significant predictor of violent or property crime. Th conomic status), divorce rate (family disruption), and populati on change all to be significant when only examining hom icide in the small counties. Lee, Maume, and Ousey (2003) found poverty (low economic status) not significant w ining the average homicide rate from 1990-1992 in 1,746 nonmetropolitan counties. They did find a significant and positive relationship between disadvantage and homicide though. Osgood and Chambers (2000) also found no meani ngful relationships between indicators of economic status, poverty, or unemployment on th e juvenile violent crim e arrest rate in 264 rural counties. Consistent with so cial disorganization ility), female headed households (family disruption), and ethnic heterogeneity (populatio heterogeneity) to be associat ed with juvenile arrests. 25

PAGE 26

Jobes, Barclay, Weinand, and Donnermeyer (2004) also found none of their economic measures to be significant when examining crime rates in 123 rural LGAs7 in Australia. In support of social disorganization theory, though, they did find residential instability (mobility) and fa ult e perc ent of families receiving aid, unemployment) black (population hetero geneity) predict both violent and property crime. Contrary to most fi gs ral icide rates. When Lee and colleagues (2003) examined rural homicide rates, they ty to be nonsignificant. Osgood and mily instability (disruption) to be associ ated with higher rates of crime. Higher proportions of indigenous people (population he terogeneity) were also associated with higher rates of assa and with break and enter crimes. Arthur (1991) examined the socioeconomic pred ictors of violent and property crime in 13 rural Georgia counties from 1975-1985. He found that indicators of low economic status (the percent of the population below poverty, th and percent ndings at the urban level, however, he found the variables to be be tter predictors of rural property crime than of rural violent crime. Summary Based on the review of the literature, there are numerous inconsistencies in the findin when examining social disorganization in urban areas. Likewise, it appears these inconsistencies in the findings persist when applying social disorg anization theory to rural areas, especially when examining economic disadvantage indicators. Barnett and Mencken (2002) found resource disadvantage to have a significant positive imp act on violent crime. Kposowa and colleagues (1995), however, did not find poverty to be a signifi cant predictor of violent crime arrests in ru areas, but they did find poverty to be predictive of hom found pover 7 LGA is a local government area. It is a term used in Australia to refer to areas co ntrolled by each individual local government. 26

PAGE 27

27 Consistent with social disorganization theo ry, I propose that the measures of social disorgaization will have a positive effect on rural ar rests for crime. As indicated in this chapter, umerous studies have found some support for social disorganization indicators in predicting crime in both urban cities and in nonmetropolitan counties. H1: Increases in social disorganization will be related in increases in arrests in rural areas. Chambers (2000) also found no meaningful rela tionships between the indicators of economic status, poverty, or unemploym ent on juvenile arrests. Conclusion n n

PAGE 28

CHAPTER 3 THE INDUSTRIALIZATION OF FARMING The industrialization of farming refers to th e transformation whereby farms have become larger in scale and have declined in nu mber (Drabenstott & Smith, 1996; Lobao, 2000; Stofferahn, 2006) Due to technology changes in th e 1950s and 1960s, the number of farms decreased from 5.8 million in 1945 to only 2.3 m illion in 1974 (Albrecht, 1997; National Agricultural Statistics Service, 2002). Industrialization accelerated again during the farm crisis of the 1980s. By 1992, the number of farms across the country had declined to 1.9 million (Brasier, 2005). Industrial restructuring in urban ar eas has been argued as a mechanism for the concentrated disadvantage and violence found in urban ar eas (Massey et al., 1993, 1994; Wilson, 1987, 1996). Likewise, similar relationships may be found be tween industrialized farming and rural crime. While several scholars (Ousey, 2000; Parker, 2004; Shilhadeh & Ousey, 1998) have examined the relationship between industria l restructuring and crime in ur ban areas, little research has examined this relationship in rural areas.1 The Farm Crisis and Industrialization in the Midwest The farm crisis of the 1980s was the result of a multitude of factors including: a worldwide recession that reduced demand for farm exports, low farmland values, lower incomes, and large amounts of debt (Bro wn & Swanson, Ch 11, 2003). Farmers and communities in the Midwest were hit particularly ha rd. By 1989, 39% of Midwest farmers faced serious financial 1 Lee and Ousey (2001) examined the relationship between small manufacturing, economic deprivation, and crime rates in nonmetropolian counties. They found that in co unties where there were more small manufacturing firms, crime rates were lower. 28

PAGE 29

problems (Lansley et al, 1995). Between 1982 and 1992, the feed and grain region2 lost 17.3% of its farms (Brasier, 2005). Over the last two decades farms in the Heartland states3 have declined by one-fourth while in creasing in average size by on e-fourth to about 750 acres (Barkema & Drabenstott, 1996). Iowa was the hardest hit of the Midwest Stat es. Net farm income in Iowa went from $17,680 in 1981 to $7,366 in 1982 to $-1,891 in 1983 (Davidson, 1996, p. 17). Between 1982 and 1987, 27% of Iowa hog farmers went out of business. The economic shock of the farm crisis trickled down to the communities that depended on the farmers: retail sales declined by 25% in the 1980s; bankruptcies among Iowa businesses ro se 46% in 1985 alone; and poverty more than doubled in Iowa from 1979-1985 (Davidson, 1996). Perh aps Jacobsen said it be st: Its not really a farm crisis at all. Its a rural community crisis, and if you understand it in that way, its even scarier. (quoted in Davidson, 1996, p. 53). While the term ghetto has historically been associated with urban areas and minorities, the farm crisis and industrializati on led to the rise of the rural ghetto in the Midwest. The word ghetto speaks of the rising poverty rates, the chronic unemployment, and the recent spread of low-wage, dead-end jobs. It speaks of the relentless deterioration of health-care systems, schools, road, buildings, and of the emergence of ho melessness, hunger, and poverty (Davidson, 1996, p. 158). 2 The feed and grain region is a Land Resource Region defined by the Natural Resources Conservation Service and includes 514 counties in Minnesota, South Dakota, Ne braska, Kansas, Oklahoma, Iowa, Illinois, Wisconsin, Missouri, Indiana, Michigan, and Ohio (Brasier, 2005). 3 Heartland states include Colorado, Iowa, Kansas, Minn esota, Missouri, Montana, Nebraska, New Mexico, North Dakota, Oklahoma, South Dakota, and Wyoming (Barkema & Drabenstott, 1996; Lobao, 2000). More than twothirds of the countrys farming-dependent coun ties are located in these states (Lobao, 2000) 29

PAGE 30

The Goldschmidt Hypothesis Goldschmidt (1947) suggested that changes in fa rm structure, particul arly a decline in the number of farms and an increase in farm size, has an adverse effect on communities. Likewise, it has been suggested that large, corporate farming4 has a negative effect on communities (Goldschmidt, 1978; MacCannell, 1988). In his classic study of two agricultural to wns in California in 1944 Arvin and Dinuba Goldschmidt (1978) found lower socio-economic conditions in the community dominated by large-scale industrial farms. On the other hand, the family-farming community had higher levels of community participation, economic well-being, and business activity. Goldschmidt concluded that quality of social conditions is [negatively] associated with scale of operations; that farm size is in fact an important causal factor in th e creation of such differences, and that it is reasonable to believe that farm size is the most important cause of these differences (Goldschmidt, 1946, p. 114). Hayes and Olmstead ( 1984) criticize, however, that the two towns of Arvin and Dinuba were not as closel y matched research sites as intended. Essentially the Goldschmidt hypothesis (1978) cl aims that levels of social, economic and political well being are lower in localities with more large scale, non-family farming. In its most rudimentary form, the hypothesis pos its that the emergence of larg e-scale farming is related to the decline of family farming and associated measurable consequences for community decay (Durrenberger & Thu, 1996, p. 410). Empirical Findings on Industria lization and Community Decay Industrialized farming has been linked to a number of consequences for communities including: greater income inequality, greater pov erty, higher unemployment, declines in local 4 It should be noted though that some family farms may deci de to incorporate because of family farmers interests in estate planning, limited liability, and income tax advantages (Lobao, 2000). 30

PAGE 31

population, increases in crime rates, less civic pa rticipation, less democratic decision making, etc (for example see Heady & Sonka, 1974; Ma rousek, 1979; Skees & Swanson, 1988). While several studies have found support for the Go ldschmidt hypothesis (see for instance Buttel & Larson, 1979; Crowley & Roscigno, 2004; Durrenberger & Thu, 1996; Goldschmidt, 1978; Lyson, Torres, & Welsh, 2001; Peters, 2002), so me have only found mixed support (Gilles & Dalecki, 1988; Harris & Gilber t, 1982; Lobao-Reif, 1987), and others have found no support for the hypothesis (Barnes & Blevins, 1992; Heat on & Brown, 1982). Some of these more recent studies are discussed in more detail below.5 One of the most comprehensive and recent st udies on the effects of industrialization on community well-being was that by Lobao (1990). Sh e examined the relationships between farm scale and farm organization on community outcomes in 3,037 counties in 1970 and 1980. Industrialized farming was related to higher inco me inequality at both time frames. It was also related to lower family income and higher poverty. These are all va riables that social disorganization theory posits relate to crime. Another comprehensive study is that by Crowle y (1999). She studied th e effects of farm concentration on community well-being in 1,053 counties in the North Central region6. She found that in counties where farm sector concentration7 is higher, so too are poverty and income inequality, and education is lower. Similar re sults were found in Crow ley and Roscignos (2004) study. 5 Refer to Appendix D for a complete list of studies on industrialized farming and community decay. 6 The North Central region includes the states of Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, Ohio, North Dakota, South Dakota, and Wisconsin (Crowley, 1999). 7 Farm sector concentration means that a few large farms hold a disproportionate share of farm property in a county (Crowley, 1999). 31

PAGE 32

Peters (2002) also found support for the notion that the industrializa tion of farming has negative consequences. In his study of 278 nonmetropolitan counties in Iowa, Kansas, and Missouri in 2000, he found that great er employment in meat pro cessing and greater employment in industrial agriculture results in wo rse socio-economic conditions for children8. Similarly, Durrenberger and Thu (1996) also fo und negative consequences. They examined the industrialization of hog farming in all 99 Iowa counties. Their results indicated that the industrialization of hog farming is indeed related to a decline in well-bei ng. In this case, industrialization was linked to an increase in food stamp usage. Gilles and Dalecki (1988) examined industrialization in nonmetropolit an counties in the Corn Belt and Great Plains regions9. Using 1950 and 1970 census data and 1949 and 1969 census of agriculture data, they found mixed suppor t for the Goldschmidt hypothesis. Contrary to expectations, they found that increa ses in the percentage of large (Class 1) farms were associated with high levels of socio-economic well-being rather than low levels. Consistent with the Goldschmidt hypotheses though, an increase in th e number of hired farm laborers (corporate farming structure) was related to negative consequences (i.e. lower socio-economic status). One of the only studies to include crime rate as an indicator of community welfare was conducted by Lyson and colleagues (2001).They an alyzed the relationshi p between scale of agriculture10, civic engagement (attachment) and community welfare in 433 agriculture dependent counties. The four dependent community welfare variables were: 1) the percent of 8 His index for children at risk included: the percent of children enrolled in free and reduced lunch programs; the percent of infants born with low birth weight; births to female teenagers; and the high school dropout rate (Peters, 2002). 9 The Corn Belt and Great Plains are tw o of the land-resource regions of the Un ited States. Both the Great Plains and the Corn belt are characterized by single-fam ily farms that produce grain and livestock. 10 Large scale farming was measured by combining three va riables: 1) the percentage of agricultural sales in a county accounted for by farms with sales of $500,000 or more, 2) the percentage of farm operators in a county that reside off their farms, and 3) the percen tage of tenant farmers in a county. 32

PAGE 33

33 families in poverty, 2) unemployment rate, 3) per centage of low birth weight babies in a county, and 4) violent crime rate. They found large scal e farming was related to well-being; however, the relationship was mediated by leve l of civic engagement (attachme nts) and the strength of the middle class. Large scale farming wa s also related to violent crime. Violent crime rates were significantly higher in the counties dominated by large scale farming and significantly lower in the counties with a strong independent middle class. Conclusion Consistent with the Goldschmidt hypothesis, I propose that the industrialization of farming will have a positive effect on arrests for crime in ru ral communities. As indicated in this chapter, numerous studies have linked i ndustrialized farming to a variet y of community consequences including: greater income inequality, greater poverty, higher unemployment (Skees & Swanson, 1998; Welsch & Lyson, 2001), decline in local population (Goldschmidt, 1978; Heady & Sonka, 1974), increases in crime rates, less civic participation (Goldschmidt, 1978), etc. H2: The industrialization of farming will be rela ted to an increase in arrest levels in rural areas.

PAGE 34

CHAPTER 4 DATA AND METHODOLOGY Sources of the Data In order to achieve the goals of this study, and to address the complexity of the issue, data were obtained from multiple sources. Uniform Crime Report (UCR) County Level Arrest Data from 1981-1983, 1991-1993, and 2001-2003 were used to measure the dependent variables. Summary Files One1 and Three2 from the 1980, 1990 and 2000 US Census, as well as, the Census of Agriculture3 from 1978, 1987, and 1997 were the sources of data for the independent variables. In addition, the 1983, 1993 and 2003 Beale Codes were employed to identify the rural sample and the 1980, 1990, and 2000 Population-Intera ction Zones for Agriculture (PIZA) were used to identify a rural sub-sample. Both th e Beale Codes and the PIZA Codes are provided through the Economic Research Service of the U.S. Department of Agriculture. Unit of Analysis The current study uses rural coun ties as the unit of analysis.4 Most tests of the Goldschmidt hypothesis use county level data.5 The definitions of rural are complex and o lacking consistency. There is a debate on whether definitions ba sed on population or those ba ften sed 1 Summary File 1 presents 100% population and housing figures for the total population and for race categories. This includes age, sex, households, household relationship, housing units, and tenure. 2 Summary File 3 presents data on the population and housing long form subjects such as income and education. It includes population totals for ancestry groups. It also included selected characteristics for a limited number of racial/ethnic categories. 3 The Census of Agriculture is the leading source of statistics about the nations agricultural production and the only source of consistent, comparable data at the county, state, and national levels. The first agriculture census was conducted in 1840. 4 Gilles (1980) suggested counties as units of analysis because they are more homogenous than are states with respect to agricultural production systems (Gilles, 1980, p. 338). Counties represent the smallest administrative units for which secondary data is available but it is important to note that 1) counties may vary considerably in their dependence on agriculture and 2) the im pacts of agrarian change in an agri culture community may be masked by industrial development in a nearby town (Gilles & Dalecki, 1988). 5 See for instance Barnes and Blevins (1992), Crowley an d Roscigno (2004), Durrenberger and Thu (1996), Gilles & Dalecki (1988), Lobao-Reif (1987), Lyson et al (2001), and Peters (2002). 34

PAGE 35

on culture should be used. There is no consen sus on either, however. As a result general population size estimates are generally used. The US Census simply defines rural as territory, population and housing units not cla ssified as urban. In order to be considered urban by the census the area has to contain at least 2,500 people. Previous studies that have examined crim rural areas have used counties as the unit of analysis, and have also relied on the rural-urban continuum codes, otherwise known as the Bea e in le codes. Sample The sample is limited to non-metropolitan coun ties in the Midwest. The sample is limited to counties in the Midwest due to the interest in looking at changes in land use, particularly changes in farming practices a nd the industrialization of farming. The farm crisis of the 1980s affected certain types of farms more than othe rs (Brasier, 2005). A significant number of farms were concentrated in the Midwest and Plains (Leistritz & Eckstrom, 1988). In addition, limiting the region to the Midwest provides some cont rol for regional variati on in historical and contemporary structural and ecolo gical conditions that may influence the precise way in which processes examined here play out at the local level (Crowley & Roscigno, 2004, p. 140). Whether a county is classified as non-metropolitan is based on the 1983, 1993, and 2003 Beale Codes provided by the USDA (see Appendix E). There are 1,055 counties in the Midwest. Of those, 860 were non-metropolitan in 1983, 834 in 1993, and 770 in 2003.6 In order to be included in the current study, the county needed to be classified as non-metropolitan across all three time periods. This reduced the sample si ze to 757 counties. Due to limited UCR reporting by Illinois across the years, Illinois counties we re omitted from the sample. This reduced the sample from 757 rural counties to 692. Another 96 rural count ies across the Midwest were 6 For a breakdown of the number of metropolitan and non-met ropolitan counties by state and year see the table in Appendix F. 35

PAGE 36

omitted in order to create a balanced data set7 resulting in a final samp le size of 596 counties at each of the three decade time points (1980, 1990, a nd 2000). To see which counties are included in the final sample size of 596, please refer to the maps in Appendix G. Sub-Sample The Population-Interaction Zones for Agricultur e (PIZA Codes) were employed to create a sub-sample to control for the interactions be tween urban-related population and farm production activities. These codes consist of a four-category classification: (1) rural, little or no urbanrelated population interaction, (2) low population interaction, (3) medium population interaction, and (4) high population interaction. To crea te the sub-sample, the final sample ( N = 596) was limited to only those counties that were classi fied as (1) rural, little or no urban-related population interaction across a ll three time periods (1980, 1990, and 2000). This created a subsample of 519 counties that was used for some analyses in this study.8 Measures Dependent Variables The current study has four dependent variables derived from the UCR County Level Arrest Data.9 The four dependent variables are the UCR to tal crime index, the Part 1 crime index, the 7 In a balanced data set each time series is the same length and has the same set of points. In other words, the 596 counties included in the final sample had complete data for all three time points, not just one or two time points. 8 The 77 counties not included in the sub sample are as fo llows: (Indiana) Blackford, Cass, Decatur, Fayette, Grant, Henry, Jay, Jefferson, Jennings, Koscuisuko, Larange, Marshall, Montgomery, Noble, Randolph, Ripley, Rush, Steuben, Wabash, and Wayne; (Iowa) Muscatine; (Kan sa) Barton, Cowley, Dickinson, Geary, McPherson, Pottatomie, Reno, Riley, and Saline; (Michigan) Shiawa ssee; (Minnesota) Goodhue, Le Sueur, McLeod, Norman, Pennington, Rice, and Wilikin; (Missour i) Cape Girardeau, Johnson, and St Francois; (Nebrask a) Adams, Hall, Saline, and Scotts Bluff; (North Dakota) Richland, Rolette Traill, and Wand; (Ohio) Ashland, Athens, Champaign, Clinton, Darke, Fayette, Hardin, Highland, Hocking, Holmes, Huron, Logan, Marion, Muskingum, Ross, Sandusky, Scioto, Seneca, Shelby, Tuscarawas, Wayne, and Williams; (South Dakota ) Brown, Davison, and Lawrence; (Wisconsin) Dodge, Jefferson, and Walworth. 9 The limitations of using UCR data are well known: it only represents crimes which come to the attention of law enforcement; agency participation is voluntary; etc. In addition, there were two major changes to the UCR county36

PAGE 37

violent crime index, and the prope rty crime index for each year. The dependent variables consist of three year averages10 in arrests for each county as reported in 1981-1983, 1991-1993, and 2001-2003 UCR arrest data.11 Using lagged three year averages is consistent with the work of Lyson et al. (2001). Counts of rare events often fluctuate substant ially from year to year. Using an average over multiple years helps to provide a more reliable estimate (Lee, Maume, & Ousey, 2003). UCR total crime index count is a grand total of the number of arrests. It includes non-index crimes such as fraud, gambling, forgery, and prostitution. The part one crime index count includes arrests for: murder, rape, robbery, aggr avated assault, burglary, larceny, motor vehicle theft, and arson. The violent crime total is an index of arrests for murder, rape, robbery, and aggravated assault. The property crime total is an index of arrests for burglary, larceny, motor vehicle theft, and arson. In all the models, I o ffset for the log of th e population (Agresti, 1996; Allison, 1999) thereby essentially creating a rate for the dependent variables. Independent Variables Social disorganization indicators Standard measures of social diso rganization are us ed in this study.12 They include (1) residential mobility, (2) the percent living below poverty (low economic status), (3) the percent level files that were implemented starting in 1994, so co mparisons before and after 1994 must be interpreted with caution. 10 Iowa data are a two year average for 1992-1993 time period because Iowa did not report UCR data in 1991 due to switching to NIBRS. In addition, Wisc onsin data are a two year average for 2002-2003 because no data were reported in 2001. 11 Common to use multiple years to creat e crime averages when doing macro-le vel research (Lee & Ousey, 2001). 12 All of the social disorganization indicators were calcula ted from Summary Files One and Three of the US Census from 1980, 1990, and 2000. 37

PAGE 38

black (population heterogeneity), and (4) the percent Hispanic (population heterogeneity). The measures of social disorg anization utilized in this study are defined below. Residential mobility. The first measure of social disorganization is residential mobility. Consistent with numerous other studies (for example see Lee & Ouse y, 2001; Lee, Maume, & Ousey, 2003; Miethe, Hughes, & McDowall 1991; Osgood & Chambers, 2000), residential mobility is calculated by dividing the number of people ages 5 and over that lived in a different house in a given year by the total number of peopl e ages 5 and over. This is then multiplied by 100 to obtain the percent. For example: 100 age) of years 5 Population( 1995)in house different ain lived that age of years 5n (Populatio Percent living below poverty. For the current study, the per cent of the population living below poverty13 is based on the number of individuals for whom poverty status was determined14 in a given year with incomes below the poverty line divided by the total populatio for whom poverty status was determined in a gi ven year. The result is then multiplied by 100 to obtain the percent. For n example: 100 1999)in ed deterermin wasstatus poverty for whom population Total( )level poverty below incomes with 1999in determined wasstatus poverty for whomn (Populatio Percent black. Several studies have measured racial heterogeneity as the percent of the population that is black (Kposowa, Breault, & Harrison, 1995; Lee & Bartkowski, 2004). This is 13 The Census Bureau uses a set of money income thresholds that vary by family size and composition to detect who is poor. If the total income for a family or unrelated indi vidual falls below the relevant poverty threshold, then the family (and all members of the family) or unrelated individual is classified as being below the poverty level. 14 Poverty status was determined for all people except in stitutionalized people, people in military group quarters, people in college dormitories, and unrelated individuals under 15 years old. 38

PAGE 39

calculated by dividing the black population for a given year by th e total population for a given year. This is then multiplied by 100 to obtain the percent. For example: 100 Population 2000( )Population Black (2000 Percent Hispanic. While most studies on urban areas include percent black or percent minority as a measure of social disorganization, the current study is more interested in the percent Hispanic. Many Hispanic immigrants have settled in th e rural Midwest to work in meatpacking and food processing due to the restruct uring of the meat pro cessing industry in the 1980s (Baker & Hotek, 2003). Including the Hispan ic population is consistent with the work of Kposowa and colleagues (1995). It is important to look at the Hispan ic population because Hispanics account for 25% of the nonmetropolit an population growth between 1990 and 2000. In addition, around 90% of all nonmetropolitan coun ties experienced Hispanic population growth in the 1990s, and the Hispanic population is growing fast er than all other ethnic and racial groups in rural America (Kandel & Newman, 2004). Wit hout the growth of the Hispanic population during the 1990s, many rural c ounties would have lost popula tion (Kandel & Cromartie, 2004). This measure is computed by dividing the Hi spanic population for a gi ven year by the total population for that year. This is then multiplied by 100 to obtain the percent. For example: 100 Population 2000( )Population Hispanic (2000 Industrialization of farming indicators (scale) Goldschmidt (1947) suggested that changes in fa rm structure, particularly a decline in the number of farms and an increase in size, has an adverse effect on community well-being. Likewise, it has been suggested that large, corporate farming has a negative effect on 39

PAGE 40

communities (Goldschmidt, 1978; MacCannell, 1988). Because of this the following measures of land use have been included in the current study. Number of farms. Goldschmidt (1947) suggested that a decline in the number of farms has negative effects on well-bei ng. When farms decrease in number and increase in size, employment opportunities are reduced this then affects the viability of communities (Marousek, 1979). This variable is simply measured as the number of farms in a county in a given year as reported in the Census of Agriculture.15 Average farm size.16 Average farm size refers to th e average farm size in acres for a given year (1978, 1987 and 1997) in each county as reported in the Census of Agriculture. The average number of acres has been used tradit ionally to measure farm size (Buttel & Larson, 1979; Flora & Flora, 1988; Gr een, 1985; Heaton & Brown, 1982; Skees & Swanson, 1988; Van Es et al, 1988). This variable is simply measured as the average size of farm in acres in a given county in a given year as reported by the Census of Agriculture. 15 The Census of Agriculture defines a farm as any place fr om which $1,000 or more of agricultural products were produced and sold, or normally would have been sold, during the census year. The definition has changed 9 times since 1850. This current definition has been in place since the 1974 Census. 16According to the Census of Agricultu re, The acreage designated as lan d in farms consists primarily of agricultural land used for crops, pasture, or grazing. It also includes woodland and wasteland not actually under cultivation or used for pasture or grazing, provided it was pa rt of the farm operators total operation. Large acreages of woodland or wasteland held for nonagricultural purposes were deleted from individual reports during the processing operations. Land in farms includes acres in the Conservation Reserve and Wetlands Reserve Programs. Land in farms is an operating unit concept and includes land owned and operated as well as land rented from others. Land used rent free was to be reported as land rented from others. All grazing land, except land used under government permits on a per-head basis, was included as land in farms provided it was part of a farm or ranch. Land under the exclusive use of a grazing association was to be re ported by the grazing association and included in land in farms. All land in American Indian reservations used for growing crops or grazing livestock was to be included as land in farms. Land in reservations not reported by individual American Indians or non-Native Americans was to be reported in the name of the cooperative group that used the land. In many instances, an entire American Indian Reservation was reported as one farm. (Census of Agriculture, 1997, definition of Land in Farms). 40

PAGE 41

Industrialization of farming indicators (structure) Scale alone does not capture th e organizational features of industrialized farming since family farms may now be large scale due to advances in technology. Most large farms are operated by owners who are sole proprietors (R eimund et al, 1987). Therefore, organizational measures of industrial farming structure are also needed (Lob ao, 2000). They are as follows: Percent hired labor. Hired labor is considered a measur e of corporate farming structure. Barnes and Blevins (1992) and Br asier (2005) measured this by us ing the percent of farms with hired labor.17 Similarly, the current study also measures this by using the percent of farms with hired labor.18 This is calculated by dividing the number of farms that use hired labor in a given year by the total number of farms for that year. This is then multiplied by 100 to obtain a percent. For example: 100 1997)in Farms ofNumber Total( Labor) Hired with 1997in Farms of(Number Percent corporate farms.19 Another measure of industrial stru cture is the percent of farms that are corporate farms. This includes both family held and non-family held corporations. This is calculated by dividing the total number of farms identified as co rporations in a given year (as listed in the Census of Agriculture) by the total number of farms for that year. This is then multiplied by 100 to obtain a percent. For example: 17 Others have used the number of hired farm workers (Gilles & Dalecki, 1988; Lobao-Reif, 1987; Wimbereley, 1987). 18 According to the Census of Agriculture, hired farm and ranch labor includes regular workers, part-time workers, and members of the operators family if the received payments for labor. 19 All farms in the Census of Agriculture are classified by four type of organization: 1) individual or family (sole proprietorship), excluding partnership and corporation, 2) partnership, including family partnership, 3) corporation, includes family corporations, and 4) other, cooperative, estate or trust, institutional, etc. 41

PAGE 42

100 1997)in Farms ofNumber Total( 1997)in ns Corporatio as Classified Farms ofNumber (Total Controls The current study controls for the elderly population and for the state. In addition, a control for time is also included in the time-series models.20 Percent elderly. In several Midwest counties, the percent of the populatio n over the age of 65 represents a significant porti on of the total population. Rura l areas are aging rapidly as young adults leave. For example, in 1987 Iowa ranked 4th in the country in terms of the percent of the population over the age of 65 (Davidson, 1996). In 1998, the elderly constituted 15% of the rural population nationwide resulting in declining populations and tax bases while increasing demand for medical and social services (Rogers, 2000). For the current study, the percent elderly is based on the number of individuals 65 and over in a given year divided by the total population in a given year. This result is then multiplied by 100 to obtain the percent. For example: 100 )Population Total( age) of years 65 n (Populatio State. A dummy variable was created for the Midw est states included in the sample. This is done to capture scale effect s caused by state-level farm a nd economic policies (Brasier, 2005, p. 549). States have different policies regarding anti-corporate farming laws. Dummy variables were created for 10 states (k-1 states), with Kansas being the omitted state (LewisBeck, 1990). 20 The current study does not control for government subsidies because the Census of Agriculture did not start collecting that information until 1987. 42

PAGE 43

Time. Consistent with previous research m odeling change (Cohen & Cohen, 1983; Gilles & Dalecki, 1988), time is included as a control variable. This is because observed change may be the result of unreliable measures (Gilles & Dalecki, 1988). Dummy variables were created for 1990 and 2000 (k-1 time periods), with 1980 bein g the reference period (Lewis-Beck, 1990). Descriptives The means and standard deviations for the va riables are presented in Table 4-1 for the three different time points (1980, 1990, and 2000). When examining the dependent variables, we notice that both UCR total arre sts and violent crime arrests in creased from 1980 to 1990 to 2000. UCR total arrests had a mean of 499.5850 (SD = 632.874) in 1980, 678.3659 (SD = 948.565) in 1990, and 875.2448 (SD = 1166.573) in 2000. Violent crime arrests had a mean of 9.2405 (SD = 13.223) in 1980, 14.9221 (SD = 23.874) in 1990 and 19.2811 (SD = 28.031) in 2000. In regards to the social disorganization i ndicators, the percent black and the percent Hispanic both increased from 1980 to 1990 to 2000; however, residential mobility and the percent living below poverty fluctuated across the thre e time periods. In addition, there is also a decline in the number of farms across the three time periods. At the same time, there is an increase in the average size of farms and the pe rcent of farms that are corporations across the three time points. Table 4-2 reports the paired sample t-test values to show whether the variable means are significantly different between 1980 and 2000. All of the variables have significantly different means except for one. Beginning with the key depende nt variables in the study, the mean arrests significantly increased for total UCR crimes and for violent crimes. However, mean arrests for property crime significantly declined between 1980 and 2000, and there was no significant change in the means of Part 1 arrests. 43

PAGE 44

When examining the indicators of social diso rganization, the means for percent black and the percent Hispanic significantly increas ed from 1980 to 2000; whereas the means for residential mobility and poverty significantly decr eased. Furthermore, the mean number of farms significantly decreased from 1980 to 2000; while the average size of farms and the percent of farms that are corporations significantly increased. Multicollinearity The bivariate correlation matrices for both the sample and the sub-sample are presented in Appendix H for 1980, 1990, 2000, and for all three time points combined. Generally if a correlation is .5 or greater multicollinearity is a problem, but there is no definitive rule for this. When examining the correlation matrices for the fu ll sample there are some correlations that are in fact greater than .5. For example, residential mobility and the percent of the population that is elderly have correlation values of -. 503 in 1980, -.549 in 1990, and -.697 in 2000 To verify that multicollinearity is not a problem, variance inflation factor values were calculated for the variables included in the mode ls. Variance inflation factors indicated how much the variance of a coefficient is increa sed due to collinearity (Ott & Longnecker, 2001, p. 652). The variance inflation factor is designated by VIF = 1 / (1 R2) where R2 refers to how much of the variation in one independent variable is explained by the others (Ott & Longnecker, 20001, p. 709). It is generally accepted that a variance inflation factor greater than four indicated multicollinearity (Fisher & Mason, 1981, p. 109); however, some suggest that a value of 10 or greater indicates multi collinearity (Pindyck & Rubinfeld, 1998; Ott & Longnecker, 2001, p. 652). Table 4-3 displays the variance infl ation factor values for the independent variables included in the models. None of the vari ables have a variance inflation 44

PAGE 45

factor greater than four. This is also true wh en looking at the variance inflation factors for the sub-sample (Appendix I). Analytical Plan Cross-Sectional Analyses Cross-sectional regression models will be utilized to capture the impacts of social disorganization and industrializat ion of farming on crime in the rural Midwest at three different time points 1980, 1990, and 2000. Poisson based regression will be used to examine the structural correlates of offe nding in the Midwest counties.21 Low arrest counts may be common in the data since the study is looking at counts in rural counties. Poisson based techniques take into account the problem of the relatively low arrest counts (Osgood, 2000). Negative binomial regression models will be used because they allow for overdispersion (Gardner et al, 1995; Osgood, 2000). Count data often have overdi spersion, with the variance exceeding the mean (Agresti, 2002). Poisson forces the variance to equal the mean whereas the negative binomial distribution has )/( )Var( )(2k where is called the dispersion parameter. Comparing the sample mean and variance of the dependent count variable provide s a simple indication of overd ispersion (Cameron & Trivedi, 1998). k /1The negative binomial model, however, usually under predicts the amount of zero counts in the dependent variable. Due to the large number of zero counts, zero-inflated negative binomial regression models will also be employed. Zero-inflated models account for overdispersion due to excess zero counts (Cameron & Trived i, 1998; Min & Agresti, 2002). 21 While most prior studies on aggregate crime rates have us ed OLS regression models, this is problematic when low crime counts are common in the data (Lee & Ousey, 2001; Osgood, 2000). 45

PAGE 46

Since the zero-inflated Poisson model (ZIP) assumes that the m ean equals the variance, zeroinflated negative binomial models (ZINB) are more likely to be appropriate (Min & Agresti, 2002). While uncommon in criminology (for an example see Robinson, 2003), zero-inflated negative binomial regression (ZIN B) is common in public health research (for an example see Chin & Quddus, 2003). In all the models, I offset for the log of the population (Agresti, 1996; Allison, 1999) thereby essentially creating a rate for the dependent variables. Pooled Cross-Sectional Time Series Analyses Multivariate models are also utilized to capture the imp acts of changes in social disorganization and industriali zation of agriculture on changes in crime in the rural Midwest across time (from 1980 to 2000). The current study employs pooled cr oss-sectional time series regression as the primary techniqu e for modeling change in this re search. A variety of methods for modeling change exist in the social scie nces (Allison, 1990; Firebaughh & Beck, 1994; Kessler & Greenberg, 1981). Pooled cross-sectional time series allo ws the researcher to capture variation across different units in space, as well as variation that emerges over time (Sayrs, 1989, p. 7). Pooling the data is useful when the length of the time series is abbreviated and can boost sample size (Sayrs, 1989). There are several ways for analyzing time se ries panel data. Researchers commonly use fixed effects and/or random effects methods to analyze panel data. The current study will employ a fixed effects model. Since this study seeks to evaluate the effects of change in a covariate on the change in arrests within a given county, the fixed effects estimator is appropriate because it expresses variables only in terms of change within a un it (Parker, 2004; STATA, 2003). The fixed effect model also controls for all included and omitted time invariant variables (Allison, 1994; Johnson, 1995). 46

PAGE 47

47 In addition, the Hausman specifica tion test (1978) was used to verify that the fixed effect model is the appropriate technique for the current study. This tests whether or not the coefficients estimated by the more efficient random effects model are the same as those estimated by the fixed effects model (Greene, 2000; Hausman, Hall, & Griliches, 1984; Hsaio, 1986; Maddala, 1983). If the results of the Hausman chi-square test are insignificant, then it is acceptable to use the random effects specification in th e model. If the test yields a significant p value, then fixed effects is the more appropriate m odel. As indicated in the next ch apter, all of f our of the pooled time-series models use the fixed effects spec ification. The Hausman chi-square test was significant in all four cases.

PAGE 48

Table 4-1. Descriptive statistics (means w ith standard deviati ons in parentheses) Notes: (a) 3 cases missing, (b) 3 cases missing, (c) 6 cases missing Time 1 1980 Time 2 1990 Time 3 2000 All 3 Times UCR total arrests 499.5850 (632.874) 678.3659 (948.565) 875.2448 (1166.573) 684.0874 (953.320) Part 1 total arrests 98.8258 (127.906) 114.0922 (168.191) 103.8541 (142.642) 105.5793 (147.215) Violent crime arrests 9.2405 (13.223) 14.9221 (23.874) 19.2811 (28.031) 14.4724 (22.934) Property crime arrests 89.5584 (117.883) 99.1689 (149.131) 84.5649 (119.893) 91.0948 (129.807) Number of farms 875.89 (476.603) 779.44 (414.169) 682.67 (366.072) 779.50 (428.579) Average size of farms (acres) 571.77 (740.794) 597.9 (664.958) 659.58 (698.682) 609.69 (702.802) % Corporations 2.0048 (2.075) 3.1541 (2.686) 4.7993 (3.727) 3.3172 (3.126) % Hired labor 39.8500 (9.009) 40.1373 (9.704) 35.6219 (9.718) 38.5386 (9.699) % Black .6335 (1.933) .7424 (2.043) .8878 (2.025) .7544 (2.002) % Hispanic .9033 (1.679) 1.1818 (2.413) 2.4661 (4.571) 1.5160 (3.207) Residential mobility 40.6336 (7.142) 37.4766 (6.992) 38.0765 (6.220) 38.7321 (6.931) % Poverty 13.7299 (4.616) 14.3969 (4.771) 11.3530 (4.055) 13.1609 (4.675) % Elderly 15.9182 (3.691) 18.0515 (4.017) 17.8071 (3.965) 17.2567 (4.007) N 596 593a 593b 1782c 48

PAGE 49

Table 4-2. T-test scores for change from 1980 to 2000 (N = 591) Mean 1980 Mean 2000 Mean Diff t UCR total arrests 501.9532 874.9980 -373.0448 -13.043*** Part 1 arrests 99.3799 103.9430 -4.5632 -1.602 Violent arrests 9.2414 19.2484 -10.0071 -10.896*** Property arrests 90.1120 84.658 5.4261 2.174** Number of farms 883.13 684.90 198.23 33.185*** Average size of farms 575.00 661.09 -86.10 -9.956*** % Corporations 2.0218 4.8156 -2.7938 -27.338*** % Hired labor 39.9966 35.6672 4.3294 12.349*** % Black .6376 .8880 -.2504 -8.512*** % Hispanic .9065 2.4684 -1.5619 -11.265*** Residential mobility 40.6210 38.0892 2.5317 14.207*** % Poverty 13.7293 11.3222 2.4072 16.553*** % Elderly 15.9198 17.8216 -1.9018 -19.677*** p < .10 ** p < .05 *** p < .01 49

PAGE 50

50 Table 4-3. Variance inflation f actor values for independent va riables included in all models Time 1 1980 Time 2 1990 Time 3 2000 All 3 Times Number of farms 1.619 1.677 1.598 1.653 Average size of farms 2.338 1.901 2.119 1.969 % Corporations 2.506 2.122 2.181 2.380 % Hired labor 1.460 1.542 1.681 1.555 % Black 1.327 1.368 1.342 1.339 % Hispanic 1.637 1.636 1.541 1.540 Residential mobility 1.814 2.066 2.225 2.020 % Poverty 1.608 1.461 1.451 1.527 % Elderly 2.416 2.430 2.476 2.424 Indiana 1.876 1.771 1.732 1.725 Iowa 2.325 2.353 2.281 2.217 Michigan 2.183 2.106 2.001 2.008 Minnesota 2.199 2.103 1.970 1.998 Missouri 1.949 1.986 2.024 1.952 Nebraska 2.044 1.988 2.018 1.958 North Dakota 2.101 1.991 1.836 1.866 Ohio 1.921 1.921 1.938 1.876 South Dakota 1.928 1.698 1.615 1.676 Wisconsin 1.940 1.903 1.925 1.856 1990 dummy 1.525 2000 dummy 2.127

PAGE 51

CHAPTER 5 RESULTS This chapter presents the results from th e regression models for the cross-sectional analyses and for the pooled time-series analyses. In all the models, I offset for the log of the population thereby essentially creating a rate for the dependent variable. Cross-Sectional Analyses Tables 5-1 through 5-3 present the negative binomial and zeroinflated negative binomial regression coefficients, Z scores and standard errors for the 1980, 1990, and 2000 arrest models. The social disorganization variables are group ed at the top of the tables, followed by the industrialization of farming indicators, and then the controls. Those m odels that used ZINB instead of negative binomial regr ession are indicated in the table notes. As discussed in Chapter 4, zero-inflated models account for overdisp ersion due to excess zero counts (Cameron & Trivedi, 1998; Min & Agresti, 2002) The ZINB model is more appr opriate if the Vuong test is significant. For example, when looking at the 1980 cross sectional results (T able 5.1), the violent crime model uses the ZINB regression and had a Vuong test value of 6.32 (p < .01). When examining the social disorganization indi cators, the percent black is significant and positively related to: all four arrest models in 198 0; Part 1 arrests and violent crime arrests in 1990; and Part 1, violent, and property arrests in 2000. In other words, in 1980 for example, a one unit increase in the percent bl ack resulted in a 2.54% increase in UCR total arrests, a 4.68% increase in Part 1 arrests, a 6.83% increase in violent arrests, and a 4.21% increase in property arrests.1 Percent Hispanic is also positive and signi ficant in several models: Part 1 arrests and violent arrest in 1980; violent arrests in 1990; and all four models in 2000. As expected, 1 Exponentiating the parameter estimates, subtracting one, and multiplying this result by one hundred returns the percentage increase in the arrest rate for a unit increas e in a given independent variable (Lee et al, 2003). 51

PAGE 52

residential mobility is also positively significant in all but two of the models (violent arrests 1980 and violent arrests 1990). While the percent living below poverty is significantly related to arrests in several of the models, it is in the opposite direc tion as expected (i.e., greater am ounts of poverty resulted in less crime, instead of in more crime as social di sorganization theory would predict). For example when examining property crime arrests in 1980, a one unit increase in pove rty results in a 3.80 percent decrease in property arre sts. While this finding is contrary to social disorganization theory, it is not surprising if one looks at prior research. There is mixed support for the notion that the industrialization of ag riculture results in more crime in a county. As expected, a decline in the number of farms resu lts in an increase in criminal arrests in two of the models (violent arrests 1980 and violent a rrests 2000). Likewise the percentage of farms that are corporate farms is positively related to arrests (i.e. more corporate farms, more crime), in three of the models (violent arrests 1990, part 1 arrest 2000, and property arrests 2000). For instance, in 1990, a one unit increase in the percentage of corporate farms corresponds with a 3.64 pe rcent increase in violent crime arrests. The percent of farms that use hi red labor is significant in several of the models, but it is in the opposite direction as expected. As the percen t of farms with hired labor increases, arrests significantly decrease. This may be because this measure is more a proxy of economics or employment in a county than it is a proxy of co rporate farming structure, especially since the measure of hired labor in cludes both part-time work ers and family members. Pooled Time-Series Analyses Table 5-4 presents the overall means, as we ll as the overall, between, and within sample standard deviations for the indi cators included in the time-series models. For example, there was 52

PAGE 53

an overall mean of 776.939 farms from 1980 to 2000. The overall mean size of those farms was 609.688 acres. Table 5-5 presents the findings from the pool ed time-series multivariate models. These results are discussed below. Results from the sub-sample are available in Appendix I, and are similar to those presented here on the full samp le. As discussed in ch apter four, all models employ a fixed-effects specification since this study seeks to evaluate the e ffects of change in a covariate on the change in the dependent variable within a given county. In addition, results from the Hausman Test (also provided in Table 5-5) indicate that the fi xed-effects model is the more appropriate technique than the random-effects model. There is no support for the first hypothesis th at increases in soci al disorganization indicators from 1980 to 2000 are related to incr eases in arrests over the past two decades. Contrary to expectations, as the black population increased over time within a county, it was significantly related to a decline in crime in tw o of the models (total UCR arrests and violent crime arrests within a county). Likewise, the increase in the Hi spanic population from 1980 to 2000 within a county was also signifi cantly related to a decline in one model (total UCR arrests) and was insignificant in the other models. Also contrary to social disorganization theory, residential mobility is inversely related to Part 1 arrests and property arrests (i.e., as residential mobility increased within a county over time, arre sts declined), and the percent below poverty is insignificant across all four arrest models. There is some support for the second hypothesis that increases in the industrialization of farming from 1980 to 2000 will be related to an incr ease in arrests over the past two decades. As expected, based on the Goldschmidt hypothesis, a decline in the number of farms over the past two decades within a county is si gnificantly related to an increase over time in total UCR arrests, 53

PAGE 54

Part 1 arrests, violent crime arrests, and prope rty crime arrests within a county. In addition, increases in the average farm size over the past two decades are related to increases in total UCR arrests; however, average farm size is insignificant in the other three models. Changes in the indicators of industrial structur e (percent of corporate farms and the percent of farms with hired labor) within a county from 1980 to 2000 though have no significant effect on changes in arrests over time across any of the four models. Summary Overall, I find mixed support for the hypotheses presented in Chapters 2 and 3. There is support for social disorganization theory when examining cross-sectional data. However, when examining the pooled time-series data, all of th e social disorganizati on indicators that are significant are in the opposite direct ion of what the theory posits. In terms of the i ndustrialization of farming, the decline in the number of farms is significantly related to increases in arrests in both the cross-sectional and pool ed time series analyses. 54

PAGE 55

Table 5-1. 1980 cross-sectiona l negative binomial and ZINB re gression results (and z-scores) UCR Total Part 1 Violenta Property % Black SE .0250933* (1.74) .014425 .0457475*** (3.43) .0133293 .0660362*** (4.88) .0135437 .0412246*** (3.02) .0136546 % Hispanic SE .0107253 (0.59) .0181989 .0298051* (1.70) .0175532 .0566827*** (2.99) .0189863 .0214435 (1.19) .0179563 Residential mobility SE .0287212*** (6.20) .0046309 .0317195*** (7.21) .0043996 .0080715 (1.56) .0051706 .0333057*** (7.38) .0045134 % Below poverty SE -.0189387*** (-2.99) .0063357 -.0340157*** (-5.57) .0061049 -.0010628 (-0.14) .0075486 -.0387894*** (-6.19) .0062692 Number of farms SE -.0000694 (-1.11) .0000625 -.0000779 (-1.32) .0000588 -.0003423*** (-5.16) .0000664 -.0000551 (-0.91) .0000603 Average farm size SE .0000633 (1.23) .0000514 .0000283 (0.55) .0000515 .0001255 (1.51) .0000832 .0000233 (0.44) .0000526 % Corporate farms SE -.0138931 (-0.75) .0186021 -.0193353 (-1.14) .0170279 -.0163732 (-0.77) .0212093 -.0184973 (-1.07) .0173618 % Hired labor SE -.00417 (-1.34) .0031077 -.0028328 (-0.95) .0029901 -.0113462*** (-3.35) .0033829 -.0010906 (-0.35) .0031001 % Elderly SE -.0324052*** (-3.28) .0098752 -.0139942 (-1.45) .009662 -.0282118** (-2.42) .0116421 -.0124708 (-1.25) .0100098 Indiana SE -.0447772 (-.038) .1467518 -.1210811 (-0.87) .1384799 -.7255734*** (-4.61) .1573106 -.04694 (-0.33) .141822 Iowa SE .1001402 (0.94) .1066709 .1707421* (1.68) .1014366 -.3428275*** (-2.85) .1201236 .232111** (2.23) .1038772 Note: (a) ZINB regression model, 42 zero observations p < .10 ** p < .05 *** p < .01 55

PAGE 56

Table 5-1. Continued UCR Total Part 1 Violenta Property Michigan SE .2098717* (1.74) .120291 .2199783* (1.89) .11644 -.2220428* (-1.66) .1336461 .2646281** (2.21) .1195077 Minnesota SE .0198837 (0.17) .1138318) .2701722** (2.47) .1095618 -.4883436*** (-3.76) .1299759 .3568799*** (3.18) .1122905 Missouri SE -.0502783 (-0.43) .116745 .0681471 (0.62) .1095643 -.1103503 (-0.90) .1227679 .1091818 (0.97) .1123728 Nebraska SE .132694 (1.27) .1044673 -.0117065 (-0.12) .1002855 -.5377586*** (-4.14) .1298453 .0670898 (0.65) .1027791 North Dakota SE .4591505*** (3.632) .1267906 -.0500258 (-0.40) .123539 -.8469451*** (-4.72) .1794992 .0521982 (0.41) .126729 Ohio SE -.1230941 (-0.89) .1379608 -.1353041 (-1.04) .130319 -.6459574*** (-4.41) .1463151 -.0600823 (-0.45) .1333017 South Dakota SE .5606135*** (3.94) .1421578 .2548966* (1.92) .1325453 -.1452041 (-0.92) .1581768 .2868504** (2.11) .1359079 Wisconsin SE .5798815*** (4.78) .1213114 .6261943*** (5.45) .1149615 -.0013974 (-0.01) .1319821 .6884086*** (5.85) .1176409 Constant -4.19807*** -6.118062*** -6.659407*** -6.392144*** Log Likelihood -3832.4292 -2803.6708 -1505.243 -2745.3832 N 596 596 596 596 Maximum Likelihood R2 0.307 0.353 0.771 0.348 Vuong Test 6.32*** Note: (a) ZINB regression model, 42 zero observations p < .10 ** p < .05 *** p < .01 56

PAGE 57

Table 5-2. 1990 cross-sectiona l negative binomial and ZINB re gression results (and z-scores) UCR Total Part 1a Violentb Propertyc % Black SE .0160837 (0.97) .0165502 .0296571* (1.79) .0165594 .0540349*** (3.34) .0162886 .0221252 (1.29) .0171746 % Hispanic SE .0119532 (0.88) .0135176 .0160838 (1.10) .0145569 .0271923* (1.68) .0161515 .0133176 (0.91) .0147016 Residential mobility SE .0321487*** (5.84) .0055017 .0426218*** (7.03) .006066 .0101884 (1.50) .0068036 .0459214*** (7.27) .0063149 % Below poverty SE .0005098 (0.08) .006365 -.0147774** (-2.09) .0070736 -.0019339 (-0.25) .0076473 -.0167824** (-2.27) .0074045 Number of farms SE .0000637 (0.80) .0000798 .0000326 (0.38) .0000854 -.0001183 (-1.35) .0000874 .0000314 (0.36) .0000883 Average farm size SE -.0001238** (-2.18) .0000567 -.0000854 (-1.18) .0000725 .0000467 (0.53) .0000885 -.0001005 (-1.31) .0000769 % Corporate farms SE .0175035 (1.30) .0134393 .0133845 (0.90) .0148367 .0357684** (2.18) .0164239 .0109586 (0.71) .0154099 % Hired labor SE -.0062207* (-1.90) .0032772 -.0054329 (-1.54) .0035197 -.0079184** (-2.11) .0037615 -.0048024 (-1.31) .0036643 % Elderly SE -.0397267*** (-3.94) .0100956 -.0319535*** (-2.87) .0111395 -.0332394*** (-2.65) .0125487 -.0301594*** (-2.60) .0116052 Indiana SE -.5763356*** (-3.77) .1530174 -.7456477*** (-4.60) .1620066 -.9016025*** (-5.25) .1717994 -.7209942*** (-4.30) .1676722 Iowa SE -.2122415* (-1.86) .1139715 -.5159315*** (-4.21) .1226892 -.718623*** (-5.37) .1337438 -.4618421*** (-3.61) .1280505 Notes: (a) ZINB regression model, 11 zero obs ervations; (b) ZINB re gression model, 45 zero observations, (c) ZINB regressi on model, 16 zero observations p < .10 ** p < .05 *** p < .01 57

PAGE 58

Table 5-2. Continued UCR Total Part 1a Violentb Propertyc Michigan SE -.0387671 (-0.30) .1287002 -.1561611 (-1.14) .1375001 -.4141667*** (-2.90) .142736 -.1216386 (-0.85) .1429338 Minnesota SE .1425462 (1.21) .1176641 .2142951* (1.70) .1259467 -.103461 (-0.78) .1332107 .2592412** (1.98) .1307571 Missouri SE -.5292951*** (-4.26) .1241697 -.5146341*** (-3.91) .1317162 -.6062202*** (-4.38) .1384507 -.4942074*** (-3.62) .1365464 Nebraska SE .0131272 (0.12) .1084458 -.2004512* (-1.71) .1170202 -.8256106*** (-6.09) .1356763 -.1024361 (-0.84) .1215456 North Dakota SE -.3623152*** (-2.78) .1303431 -.2512866* (-1.71) .146607 -1.474432*** (-6.75) .2185654 -.0665682 (-0.44) .1525086 Ohio SE -.545606*** (-3.70) .1472986 -.8251643*** (-5.25) .1572415 -1.431436*** (-8.62) .1660642 -.6808622*** (-4.15) .1642018 South Dakota SE .3805642*** (2.68) .1430652 -.0407468 (-0.27) .1488112 -.5477135*** (-3.28) .1672262 .0544558 (0.35) .1552585 Wisconsin SE .6256112*** (4.84) .1292432 .4610953*** (3.35) .1377935 -.2568768* (-1.77) .1450342 .5644702*** (3.95) .1430194 Constant -3.859042*** -5.914251*** -6.345573*** -6.255604*** Log Likelihood -3939.8617 -2847.904 -1724.54 -2756.862 N 593 593 593 593 Maximum Likelihood R2 0.376 0.813 0.784 0.810 Vuong Test 2.84*** 4.94*** 3.62*** Notes: (a) ZINB regression model, 11 zero obs ervations; (b) ZINB re gression model, 45 zero observations, (c) ZINB regressi on model, 16 zero observations p < .10 ** p < .05 *** p < .01 58

PAGE 59

Table 5-3. 2000 cross-sectiona l negative binomial and ZINB re gression results (and z-scores) UCR Total Part 1a Violentb Propertyc % Black SE .0183299 (1.42) .0128977 .0372741*** (2.63) .0141532 .0498169*** (3.99) .0124749 .0352867** (2.33) .0151287 % Hispanic SE .0062945 (1.07) .0058897 .0132311** (1.99) .0066414 .0169453*** (2.56) .0066116 .0129998* (1.87) .0069509 Residential mobility SE .037127*** (7.13) .0052099 .0418651*** (6.97) .0060102 .0180735*** (3.00) .0060173 .0441429*** (6.87) .0064288 % Below poverty SE -.0024032 (-0.38) .0063605 -.0089238 (-1.19) .0075102 .0107717 (1.44) .007468 -.014825* (-1.85) .0080016 Number of farms SE -.0000589 (-0.80) .000074 -.0000447 (-0.55) .0000817 -.0001425* (-1.87) .0000763 -8.93e-06 (-0.10) .0000874 Average farm size SE -.0000348 (-0.73) .0000475 -9.73 e -06 (-0.16) .000061 .000065 (0.84) .0000778 -9.17e-06 (-0.14) .000066 % Corporate farms SE .0093193 (1.10) .0084702 .0163152* (1.68) .0097062 0143156 (1.45) .0098431 .0173596* (1.68) .0103511 % Hired labor SE -.0092552*** (-3.25) .0028467 -.0106454*** (-3.31) .0032202 -.0161049*** (-5.06) .0031827 -.0096868*** (-2.81) .0034509 % Elderly SE -.0162742** (-1.93) .008449 -.0124079 (-1.25) .0099279 -.0151966 (-1.46) .0103982 -.010138 (-0.95) .0106574 Indiana SE .1145506 (0.91) .1263315 .4137682*** (2.94) .1405581 .0752734 (0.55) .1378586 .5075558*** (3.40) .1492795 Iowa SE .0300274 (0.32) .0950975 .3300102*** (3.00) .1098655 .3713339*** (3.30) .1124568 .32844*** (2.80) .1174247 Notes: (a) ZINB regression model, 10 zero obs ervations; (b) ZINB re gression model, 28 zero observations, (c), ZINB regressi on model, 14 zero observations p < .10 ** p < .05 *** p < .01 59

PAGE 60

Table 5-3. Continued UCR Total Part 1a Violentb Propertyc Michigan SE .304954*** (2.87) .1062637 .2175444* (1.81) .1198635 .1673453 (1.40) .1195726 .2339188* (1.82) .1282095 Minnesota SE .3167465*** (3.25) .097591 .6323428*** (5.68) .1112658 .2459167** (2.17) .1134081 .7417164*** (6.23) .1190155 Missouri SE .0839006 (0.79) .1064908 .5390233*** (4.52) .1191321 .5949907*** (5.15) .1154466 .5298716*** (4.17) .1270982 Nebraska SE .2217144** (2.36) .0941129 .18977* (1.74) .1088991 -.184996 (-1.52) .1214146 .3114538*** (2.66) .1168814 North Dakota SE .4254756*** (3.97) .1071415 .4792039*** (3.77) .1272027 -.6717096*** (-3.92) .1714066 .6851137*** (5.05) .135647 Ohio SE .0292239 (0.24) .1242349 .2162578 (1.56) .1383167 -.5696397*** (-4.16) .1369203 .386704*** (2.63) .1470037 South Dakota SE .3712066*** (3.29) .1129348 .2294772* (1.79) .1284451 -.1756508 (-1.23) .142996 .361983*** (2.62) .1381194 Wisconsin SE .938421*** (8.46) .1109036 1.00269*** (8.05) .1245745 .8226698*** (6.66) .123463 1.03885*** (7.85) .1323966 Constant -4.416545*** -6.959965*** -7.393848*** -7.363348*** Log Likelihood -4019.9741 -2747.193 -1777.335 -2635.211 N 593 593 593 593 Maximum Likelihood R2 0.350 0.844 0.845 0.833 Vuong Test 2.68*** 3.97*** 3.27*** Notes: (a) ZINB regression model, 10 zero obs ervations; (b) ZINB re gression model, 28 zero observations, (c), ZINB regressi on model, 14 zero observations p < .10 ** p < .05 *** p < .01 60

PAGE 61

Table 5-4. Means and overall, between and with in sample standard deviations for indicators included in the time-series models (N = 1788, n = 596, T = 3) Overall mean Overall SD Between SD Within SD UCR total 683.1834 952.263 883.0524 357.6257 Part 1 105.3833 147.0369 138.9387 48.3476 Violent 14.48145 22.95122 19.00135 12.88841 Property 90.88963 129.6461 122.7206 42.00746 Number of farms 776.939 430.1281 418.194 101.5857 Average size of farms 609.688 702.8018 694.0328 108.9772 % Corporations 3.309681 3.126718 2.688989 1.598059 % Hired labor 38.4711 9.813698 8.382787 5.110402 % Black 0.7582758 2.006605 1.971166 .3812051 % Hispanic 1.513976 3.202232 2.772108 1.605711 Residential mobility 38.71674 6.933031 6.395852 2.68436 % Poverty 13.18587 4.747422 4.156911 2.297269 % Elderly 17.25987 4.016864 3.7633331 1.410101 61

PAGE 62

Table 5-5. Pooled cross-sectional time series negative binomial regressi on results (and z-scores), 1980-2000 (N = 1788, n = 596, T = 3) UCR Total Part 1 Violent Property % Black SE -.0384619** (-2.47) .0155577 -.0275986 (-1.50) .0184092 -.0446019** (-2.02) .0220354 -.021807 (-1.15) .0189964 % Hispanic SE -.0109168** (-2.22) .0049071) -.0077867 (-1.46) .0053385 -.007803 (-1.06) .007334 -.0074412 (-1.33) .0055897 Residential mobility SE -.0047676 (-1.17) .0040694 -.010506** (-2.28) .0046076 -.0090004 (-1.41) .0064059 -.0107304** (-2.23) .0048059 % Below poverty SE -.0038192 (-0.80) .00474843 -.0001653 (-0.03) .0056515 -.0092298 (-1.16) .0079519 .0024384 (0.41) .0059016 Number of farms SE -.0004975*** (-5.56) .0000894 -.0006321*** (-6.11) .0001034 -.0008568*** (-6.57) .0001305 -.0005653*** (-5.36) .0001054 Average farm size SE .0001968*** (2.86) .0000689 .0001467 (1.41) .0001041 .0001468 (0.90) .0001624 .0001335 (1.28) .0001046 % Corporate farms SE -.0093096 (-1.25) .0074573 .0018302 (0.20) .0090768 -.0036575 (-0.29) .0124186 .0030338 (0.32) .0095969 % Hired labor SE -.0005691 (-0.31) .0018155 -.0011087 (-0.52) .0021145 .0008163 (0.28) .0029352 -.00124 (-0.56) .0022145 % Elderly SE -.0077299 (-0.86) .008961 -.015158 (-1.44) .0105212 .0192787 (1.32) .014599 -.0193308* (-1.77) .010904 1990 dummy SE .2753487*** (8.28) .0332741 .0648994* (1.74) .0372084 .2882677*** (5.55) .0519842 .0281988 (0.74) .0382662 2000 dummy SE .4110019*** (10.25) .0401147 -.2354687*** (-4.99) .0471642 .3155274*** (4.93) .0639827 -.3192851*** (-6.53) .0488635 Indiana SE -1.619943*** (-7.23) .2240569 -1.877463*** (-7.69) .2441663 -1.48973*** (-5.10) .291956 -1.820965*** (-7.41) .245731 p < .10 ** p < .05 *** p < .01 62

PAGE 63

63 Table 5-5. Continued UCR Total Part 1 Violent Property Iowa SE -.6964233*** (-4.06) .171429 -1.30432*** (-6.57) .1985452 -.7701236*** (-3.19) .2414295 -1.314879*** (-6.57) .2002462 Michigan SE .0002102 (0.00) .196434 -.1848774 (-0.83) .2219363 .5454879 (1.63) .3347629 -.2547243 (-1.14) .2226815 Minnesota SE -.5435243*** (-3.02) .18024 -.458451** (-2.17) .2111221 .1057826 (0.37) .2836133 -.4552677** (-2.12) .2150322 Missouri SE -.7893498*** (-4.27) .1848205 -.9032593*** (-4.32) .2088467 -.5498497** (-2.20) .2497546 -.7903809*** (-3.72) .2122836 Nebraska SE .2703297 (1.55) .1746884 .1883918 (0.89) .2106554 .4137071 (1.18) .3498637 .1895065 (0.89) .2129819 North Dakota SE -1.016781*** (-5.31) .1915442 -.1732043 (-0.67) .2567837 -.0936378 (-0.20) .4699943 -.0138508 (-0.05) .2620215 Ohio SE -.9948173*** (-4.54) .2191897 -1.365956*** (-5.55) .2463197 -1.158445*** (-3.89) .2975563 -1.350391*** (-5.40) .2500489 South Dakota SE -.227901 (-1.11) .2060755 -.1040513 (-0.43) .2398184 .374723 (0.97) .385849 -.0662275 (-0.27) .245903 Wisconsin SE .0169701 (0.09) .1969752 .1277944 (0.58) .2212581 -.7724992*** (-2.94) .2625513) .152574 (0.68) .2231384 Constant -6.867052*** -6.294809*** -7.438934*** -6.367896*** Log Likelihood -7130.389 -4998.8519 -3010.8737 -4835.8573 N Observations/Groups 1781 / 595 a 1778 / 594 b 1763 / 589 c 1778 / 594 d Hausman Test 325.94*** 373.69*** 205.22*** 333.38*** Notes: (a) 1 group (1 observation) was dropped be cause of only one observation per group; (b)1 group (1 observation) was dropped because of only one observation per groups and 1 group (3 observations) were dropped due to all zero outcomes; (c) 1 group (1 observation was dropped because of only one observation per group and 6 groups (18 observations) were dropped due to all zero outcomes; (d) 1 group (1 observation) was dropped because of only one observation per group and 1 group (3 observations) were dropped due to all zero outcomes. p < .10 ** p < .05 *** p < .01

PAGE 64

CHAPTER 6 DISCUSSION AND CONCLUSION The purpose of this research was to explore th e dynamics of social disorganization, the industrialization of farming, a nd criminal arrests across rural counties in the Midwest. This included examining changes in structural conditions and in arrests across rural Midwest counties from 1980 to 2000. Cross-sectional time series analys es were presented in order to evaluate the utility of social disorganizati on theory and the Goldschmidt hypothe sis at explaining arrest rates at three individual points of time (1980, 1990, and 2000). In addition, pooled-time series analyses were also presented in order to evaluate the effectiveness of social disorganization theory and the Goldschmidt Hypothesis at explaining changes in arrests within rural counties over time (1980-2000). Discussion In regards to the first research question, ther e have been changes in crime (as measured by arrests) in the rural Midwest over the past two d ecades. Most notably, violent crime arrests in the rural Midwest have significantly in creased over the past two decades. This is consistent with the claims of other scholars (Donnermeyer, 1994; Rephann, 1999; Weisheit & Donnermeyer, 2000). This continued increase in rural violent crime i ndicates the necessity to continue studying crime in non-metropolitan areas. The second research question posed was: What is the impact of social disorganization on arrests in rural counties in the Midwest? The re sults presented in chapter five find mixed support for social disorganization theory when trying to explain crime at one point in time (1980, 1990, and 2000). The percent black, the percent Hispanic, and reside ntial mobility operated as expected. However, the percent of the population below poverty was related to a decline in crime in several of the models. 64

PAGE 65

There are several possible explanations for th ese findings in regards to poverty. The informal economy of the poor may be dominated mo re by licit activit ies such as hunting, fishing, selling home-grown vegetables, and repairing farm machinery, than by illicit activities (Lee et al., 2003, p. 126). It may also be that the poor in rural areas ar e better suited to deal with economic crises because they have access to mo re kinship and friendship networks than their urban counterparts. When looking at the pooled time-series results, increases in the i ndicators of social disorganization from 1980 to 1990 within a given county do not explain increases in rural crime over time. This may highlight a problem with soci al disorganization theorys ability to explain trends over time. Other scholars have also found less support for social disorganization theory when examining changes over time (for exampl e, see Miethe, Hughes, & McDowall, 1991). The current findings may also be due to the inability of the theory to explain crime outside of the large, urban context. Several researchers have qu estioned the applicability of urban theories at explaining rural crime and have suggested the need for the development of criminological theories that explore the rural contex t (Cebulak, 2004; Weisheit & Wells, 1996). The third research question was: What is the impact of the industrialization of farming on arrests in rural counties in the Midwest? More particularly, the Goldschmidt hypothesis asserts that a decline in the number of farms and in creases in farm size adversely affect rural communities. Overall, there is some indication that the indus trialization of farming does influence crime rates both cross-sectionally, and over time. This is especially true when looking at the number of farms within a county. From 1980 to 2000, a decline in the number of farms was significantly related to an increase in crimin al arrests within a county. Numerous studies (as mentioned in chapter three) have showed that th e decline of family farming has resulted in a 65

PAGE 66

variety of consequences for rural communities. But since it is unlikely that we will see a resurgence in family farming anytime in the near future, perhaps the better question is to explore what factors help mediate the negative conseque nces of the industrialization of agriculture. One possible mediator may be attachment s. Lyson et al. (2001) found that civic engagement (i.e. the percentage of the population belonging to a church and the percentage of the population voting) mediated the effects of la rge-scale farm operations. Criminologists have also found that a civic engagement is related to a decline in some types of crime in rural areas (Lee & Bartkowski, 2004a, 2004b) and several other researchers have high lighted the importance of religious institutions to ru ral civic life (Bartkowski & Regi s, 2003; Ellison & Sherkat, 1995; Paerisi et al, 2002). Civic engageme nt may also enhance the social networks and trust within a community. Limitations and Future Research Like all studies, this one is not without lim itations. There are numerous limitations with using UCR data. Most notably, th e data only present crimes which come to the attention of law enforcement. Under-reporting may be especially tr ue in rural areas where people have less trust in the government, are more reluctant to seek outside assistance, and generally handle problems informally (Cebulak, 2004; Wells & Weisheit, 1 996). Also, there was a change in the way UCR county-level files were implem ented beginning in 1994, so anal ysis comparing previous and subsequent years must be interpreted with caution. In addition, the current study only examined the total counts for crime, as well as the total counts for violent and property offenses. Ther e has been a recent trend in criminology to disaggregate both by specific offense and by gender. One concern w ith disaggregating by specific offense in rural areas though is that doing so results in a la rger number of either very low counts or even zero counts. While zero-inflated st atistical models do help, they are not magic 66

PAGE 67

67 (Osgood, 2000; Pridemore, 2005). These models depe nd on trustworthy data and, the validity of rural crime data at the county leve l is questionable (Pridmore, 2005). Finally, the current study was limited to only the Midwest from 1980 to 2000. It would be worth expanding the sample to in clude other regions as well as ear lier data. Changes in land use and the industrialization of fa rming have been occurring throughout the past century. While the industrialization of farming accelerat ed during the farm crisis of the 1980s, it also accelerated in the 1950s and 1960s due to advances in technology. Conclusion The findings from this study i ndicate that rural crime can no longer be ignored. Violent crime in rural areas is on the rise. We need to move beyond the myth that rural areas do not matter, there is no crime there, and consequent ly they do not deserve much consideration (Cebulak, 2004, p. 80). Furthermore, in the time series analyses, increases in the percent black and the percent Hispanic were related to decreases in arrests. This finding is contrary to most criminological research and further highlights the importance of why we need to continue exploring crime within the rural context. Until we can understand ru ral crime, the usefulness of national crime control policies will be limited (Weisheit & Wells, 1996, p. 395).

PAGE 68

APPENDIX A MIDWEST STATES INDIANA IOWA KANSAS MICHIGAN MINNESOTA MISSOURI NEBRASKA NORTH DAKOTA OHIO SOUTH DAKOTA WISCONSIN Note: Illinois was omitted from the final sample due to limited UCR reporting 68

PAGE 69

APPENDIX B SUMMARY OF STUDIES CITED EXAMINING STRUCUTRAL CORRELATES OF CRIME IN URBAN AREAS 69

PAGE 70

Table B-1. Summary of articles cited examining structural correlates of crime in urban areas Study Methods Measures Sampson (1987) 171 cities Family Disrupti on, region, population size, housing density Sampson & Groves (1989) 238 localities & 10,905 individuals (Britain) Family disruption, urbanization, ethni c heterogeneity, socioeconomic status Sampson (1991) Britain 526 districts; 11,030 individuals Residential stability Sampson et al (1997) Chicago 8782 individuals; 343 neighborhoods Concentrated disadvantage, residential instability, collective efficacy Kposowa et al (1995) 408 counties with populations > 100,000 Poverty, church membership, divorce rate, urbanity, % black, % Hispanic, population change, unemployment, popul ation density, gini coefficient Lee et al (2003) 778 metro counties Poverty concentration, disadvantage Land et al (1990) Cities, SMSAs, & states Resource de privation, population size, population density, % divorced 70

PAGE 71

APPENDIX C SUMMARY OF STUDIES CITED EXAMINING STRUCTURAL CORRELATES OF CRIME IN RURAL AREAS 71

PAGE 72

Table C-1. Summary of articles cited examining structural correlates of crime in rural areas Study Methods Measures Findings Petee & Kowalski (1993) 630 rural counties Residential mobility, single-parent households, racial heterogeneity, poverty (% households with annual income less than $7500), population density Residential mobility has the greatest impact on violent crime, followed by single parent households, and racial heterogeneity Barnett & Mencken (2002) 2,254 nonmetro counties in continental US Index of Resource disadvantage, population change, % nonwhite, Resource disadvantage and population change positively related to viol ent crime. Population change positively relate d to property crime. Kposowa et al (1995) 1681 rural counties Poverty, economic inequality, divorce rate, population change, % unemployed, % black Poverty, % black, divo rce rate, and population change related to homicide. Divorce and population change relate d to violent crime. Population change related to property crime. Lee et al (2003) 1746 nonmetro counties Disadvantage index, poverty concentration, % divorced, population density, Disadvantage related to homicide. But poverty was not significantly related. Osgood & Chambers (2000) 264 rural counties Economic status, poverty, unemployment, residential instability, female headed households, ethnic heterogeneity, population size No relationship between economic status, poverty, or unemploymen t on juvenile violent crime rate. Residential instability, female headed households and ethnic heterogeneity were significant. Jobes et al (2004) 123 rural LGAs in Australia Residential instability, family instability, proportion indigenous None of the economic indicators significant. Residential instability and family instability were significant. Arthur (1991) 13 rural counties in Georgia Percent below poverty, percent families receiving aid, unemployment, percent black Variables predict both property and violent crime. Predict property crime better than violent. 72

PAGE 73

APPENDIX D SUMMARY OF STUDIES EXAMINING INDUSTRIALIZED FARMING AND COMMUNITY WELL-BEING 73

PAGE 74

Table D-1. Summary of studies examining the industriali zation of farming and community well-being by year of publication Study Methodology Region Measures of Industrialized Farming Community Well-Being Indicators Results Goldschmidt (1968, 1978a)(1944 original) Comparative case study, 2 communities California Scale/organization General socioeconomic indicators (class structure, services, population, politics, retail trade) Detrimental effects Heady & Sonka (1974) 150 producing areas Continental U.S. Scale Socioeconomic indicators (income, employment generation) Some detrimental effects; large farms lower food costs but generate less total community income Flora et al. (1977) 105 counties Kansas S cale/organization General socioeconomic indicators (class structure, services) Some detrimental effects; industrialized farming is related to greater income inequality but other relationships not clearly supported Fujimoto (1977) 130 towns California Scale Social fabric (community services) Detrimental Goldschmidt (1978b) States All but Alaska Scale Social fabric (agrarian class structure) Detrimental effects 74

PAGE 75

Table D-1. Continued Study Methodology Region Measures of Industrialized Farming Community Well-Being Indicators Results Wheelock (1979) 61 counties Alabam a Scale General socioeconomic indicators class structure, population size Some detrimental effects; rapid increases in farm scale related to decline of population, income, and white collar labor force; other relationships mixed Marousek (1979) Regional economic impact, one community Idaho Scale Socioeconomic indicators (income, employment generation) Some detrimental effects; large farms result in greater regional income but produce less employment than small farms Buttel & Larson (1979) State-level data Entire U.S. Scal e/organization Environment (energy use) Detrimental effects Heaton & Brown (1982) Counties Continental U.S Scale/or ganization Environment (energy use) No detrimental effects Swanson (1980) 27 counties Nebraska Scale General socioeconomic indicators (population size) Detrimental effects 75

PAGE 76

TableD-1. Continued Study Methodology Region Measures of Industrialized Farming Community Well-Being Indicators Results Harris & Gilbert (1982) State level data Continental U.S. S cale/organization General socioeconomic indicators (class structure) Some detrimental effects; large farms result in more lower class farm personnel but have positive total effects on rural income Swanson (1982) 520 communities Pennsylvani a Scale / # of farms Social fabric (population) No detrimental effects Green (1985) 109 counties Missouri Scale / organization General socioeconomic indicators (services, population size) No detrimental effects Skees & Swanson (1988) 706 counties Southern US (excluding Florida & Texas) Scale / organization General socioeconomic indicators (services) Some detrimental effects: moderate sized farms produce greater employment; large and very small farms related to higher unemployment; some detrimental impacts of large farms over time 76

PAGE 77

Table D-1. Continued Study Methodology Region Measures of Industrialized Farming Community Well-Being Indicators Results MacCannell (1988) 98 counties Arizona, California, Florida, Texas Scale / organization / capital intensity General socioeconomic indicators (population size, retail trade, local government taxation, and expenditures Detrimental effects Flora & Flora (1988) 234 counties Great Plains and West Scale General socioeconomic indicators (retail trade, population size) Some detrimental effects; medium sized farms relative to large farms enhance community well being Buttel et al (1988) 105 counties Northeas t Organization General socioeconomic indicators (population, retail trade) No detrimental effects van Es et al (1988) 331 counties Corn Belt Scale / organization General socioeconomic indicators (population size) No detrimental effects 77

PAGE 78

Table D-1. Continued Study Methodology Region Measures of Industrialized Farming Community Well-Being Indicators Results Gilles & Dalecki (1988) 346 counties Corn Belt & Central Plains Scale / organization General socioeconomic indicators Some detrimental effects; counties with greater numbers of hired laborers tend to have lower well being; other relationships for scale not supported Lobao (1990) 3037 counties Continental U.S. Scale / organization Socioeconomic indicators (income, poverty, income inequality, teenage fertility, infant mortality) Some detrimental effects: moderate sized related to better socioeconomic conditions; industrialized farming related to greater income inequality and births to teenagers, and over time to greater poverty and lower family income, but not to other indicators 78

PAGE 79

Table D-1. Continued Study Methodology Region Measures of Industrialized Farming Community Well-Being Indicators Results Lobao & Schulman (1991) 2349 rural counties US and 4 regions Scale / organization Socioeconomic indicators (poverty) No detrimental effects; moderate size farms related to lower poverty, industrialized farms have little relationship to poverty if any Barnes & Blevins (1993) 2,000 rural counties U.S. Scale / organization Socioeconomic indicators (poverty, median income) No detrimental effects Durrenberger & Thu (1996) 99 counties Iowa Scale: farm size in acres, total county hog inventory, farms with hogs, farms with more than 1000 hogs, net agriculture sales Socioeconomic indicators (people living in poverty, people receiving food stamps) Detrimental: the more large scale operations, the fewer small and moderate farms and the more people who use food stamps Irwin et al (1999) 3024 counties Continental US Organization Social fabric (residential stability) No detrimental effects Crowley (1999) 1053 counties 12 North Central states Organization Socioeconomic indicators (poverty rate, income inequality) Detrimental effects 79

PAGE 80

Table D-1. Continued Study Methodology Region Measures of Industrialized Farming Community Well-Being Indicators Results Welsh & Lyson (2001) 433 agriculture dependent counties Iowa, Kansas, Minnesota, Missouri, ND, OK, SD, vs. states without anticorporate farm laws Scale / organization Socioeconomic (% families in poverty, unemployment rate, farms realizing cash gains) Detrimental effects on agriculture dependent counties in states without anti-corporate farming laws or in states with weaker laws Lyson et al (2001) 433 agriculture dependent counties Scale / organization Soci al fabric (civically engaged middle class, participation and involvement in civic affairs, community welfare) Detrimental effects are mediated by presence of civically engaged middle class Peters (2002) Agriculture dependent counties Iowa, Kansas, & Missouri Organization Socioeconomic/Children at Risk (% children enrolled in free or reduced meals, low birth weight infants, births to female teenagers, high school dropout rate) Detrimental Effects Wilson et al (2002) Census blocks in rural counties with CAFOs Mississippi CAFOS (swine) So cial fabric (whether swine cafos were located in high poverty/high black census blocks Detrimental: swine CAFOs more likely to be located in census blocks with poor African Americans 80

PAGE 81

81Table D-1. Continued Study Methodology Region Measures of Industrialized Farming Community Well-Being Indicators Results Crowley & Roscigno (2004) Counties North Central States IA, IL, IN, KS, MI, MN, MO, NE, OH, ND, SD Scale / organization Socioeconomic (% living below poverty, inequality of income distribution among families) Detrimental Adapted from Lobao (2000) & Stofferahn (2006)

PAGE 82

APPENDIX E RURAL URBAN CONTIUUM CODES (BEALE CODES) Table E-1. 2003 Beale Codes Metro Counties 1 Counties in metro areas of 1 million population or more 2 Counties in metro areas of 250,000 to 1 million population 3 Counties in metro areas of fewer than 250,000 population Nonmetro Counties 4 Urban population of 20,000 or mo re, adjacent to a metro area 5 Urban population of 20,000 or more, not adjacent to a metro area 6 Urban population of 2,500 to 19,999, adjacent to a metro area 7 Urban population of 2,500 to 19, 999, not adjacent to a metro area 8 Completely rural or less than 2,500 urban population, adjacent to a metro area 9 Completely rural or less than 2, 500 urban population, not adjacent to a metro area Table E-2. 1983 and 1993 Beale Codes Metro Counties 0 Central counties of metro areas of 1million population or more 1 Fringe counties of metro areas of 1 million population or more 2 Counties in metro areas of 250,000 to 1 million population 3 Counties in metro areas of fewer than 250,000 population Nonmetro Counties 4 Urban population of 20,000 or mo re, adjacent to a metro area 5 Urban population of 20,000 or more, not adjacent to a metro area 6 Urban population of 2,500 to 19,999, adjacent to a metro area 7 Urban population of 2,500 to 19, 999, not adjacent to a metro area 8 Completely rural or less than 2,500 urban population, adjacent to a metro area 9 Completely rural or less than 2, 500 urban population, not adjacent to a metro area (Economic Resource Service, 2004) 82

PAGE 83

APPENDIX F MIDWEST METRO AND NONMETRO COUNTIES 1055 Counties in the Midwest Table F-1. Number of metro and nonmetro counties by state and year State Designation 1983 1993 2003 Illinois (102 counties) Metro Code 0 1 2 3 26 1 12 7 6 28 8 6 8 6 36 n/a 17 10 9 NonMetro Code 4 5 6 7 8 9 76 8 7 24 24 2 11 74 6 5 26 24 3 10 66 9 6 22 20 2 7 Indiana (92 counties) Metro Code 0 1 2 3 30 1 10 9 10 37 4 8 13 12 46 n/a 21 8 17 NonMetro Code 4 5 6 7 8 9 62 3 3 36 10 9 1 55 3 2 29 10 10 1 46 8 1 27 5 5 0 Iowa (99 counties) Metro Code 0 1 2 3 11 0 0 5 6 10 0 0 5 5 20 n/a 0 9 11 NonMetro Code 4 5 6 7 8 9 88 3 6 23 39 6 11 89 3 6 24 36 8 12 79 3 5 25 25 10 11 83

PAGE 84

Table F-1. Continued State Designation 1983 1993 2003 Kansas (105 counties) Metro Code 0 1 2 3 8 1 3 2 2 9 2 2 3 2 17 n/a 6 4 7 NonMetro Code 4 5 6 7 8 9 97 1 8 11 31 5 41 96 3 7 11 29 5 41 88 3 8 11 23 4 39 Michigan (83 counties) Metro Code 0 1 2 3 22 1 7 9 5 25 5 4 14 2 26 n/a 6 12 8 NonMetro Code 4 5 6 7 8 9 61 2 2 12 24 4 17 58 2 1 11 25 3 16 57 5 7 8 22 3 12 Minnesota (87 counties) Metro Code 0 1 2 3 16 2 8 1 5 18 5 6 0 7 21 n/a 11 2 8 Non Metro Code 4 5 6 7 8 9 71 3 1 17 32 4 14 69 3 1 18 27 4 16 66 5 4 16 20 9 12 84

PAGE 85

Table F-1. Continued State Designation 1983 1993 2003 Missouri (115 counties) Metro Code 0 1 2 3 17 3 8 0 6 22 6 8 3 5 34 n/a 17 6 11 NonMetro Code 4 5 6 7 8 9 98 1 5 24 26 7 35 93 1 3 24 25 9 31 81 3 5 22 17 10 24 Nebraska (93 counties) Metro Code 0 1 2 3 5 0 0 3 2 6 0 0 4 2 9 n/a 0 4 2 NonMetro Code 4 5 6 7 8 9 88 1 5 7 26 3 46 87 1 6 7 21 4 48 84 1 6 7 21 4 48 North Dakota (53 counties) Metro Code 0 1 2 3 4 0 0 0 4 4 0 0 0 4 4 n/a 0 0 4 NonMetro Code 4 5 6 7 8 9 49 0 1 1 11 5 31 49 0 1 2 8 9 29 49 0 1 4 5 10 29 85

PAGE 86

86 State Designation 1983 1993 2003 Ohio (88 counties) Metro Code 0 1 2 3 36 3 13 14 6 39 8 9 15 7 40 n/a 18 15 7 NonMetro Code 4 5 6 7 8 9 52 15 2 22 10 1 2 49 13 1 25 7 2 1 48 20 0 20 6 1 1 South Dakota (66 counties) Metro Code 0 1 2 3 1 0 0 0 1 3 0 0 0 3 7 n/a 0 0 3 NonMetro Code 4 5 6 7 8 9 65 0 2 2 17 4 40 63 0 1 3 14 5 40 59 0 1 3 14 5 40 Wisconsin (72 counties) Metro Code 0 1 2 3 19 1 4 5 9 20 3 4 4 9 25 n/a 7 7 11 NonMetro Code 4 5 6 7 8 9 53 5 2 18 12 6 10 52 7 0 19 10 6 10 47 7 0 19 4 11 6 Note: Illinois was omitted from final analysis due to limited reporting of UCR data.

PAGE 87

APPENDIX G MAPS OF FINAL SAMPLE (N = 596) 87

PAGE 88

88 Figure G-1. Indiana

PAGE 89

89Figure G-2. Iowa

PAGE 90

90 Figure G-3. Kansas

PAGE 91

Figure G-4. Michigan 91

PAGE 92

Figure G-5. Minnesota 92

PAGE 93

93 Figure G-6. Missouri

PAGE 94

94Figure G-7. Nebraska

PAGE 95

95 Figure G-8. North Dakota

PAGE 96

Figure G-9. Ohio 96

PAGE 97

97 Figure G-10. South Dakota

PAGE 98

Figure G-11. Wisconsin 98

PAGE 99

APPENDIX H CORRELATION TABLES 99

PAGE 100

Table H-1. Correlation matrix of variables total sample, all years UCR Part 1 Violent Property # Farms Size % Corp % Hired % Black % Hisp UCR Total 1.000 Part 1 Total .908 1.000 Violent Crime .741 .792 1.000 Property Crime .899 .994 .721 1.000 # Farms .253 .277 .144 .289 1.000 Av. Farm Size -.226 -.235 -.203 -.231 -.343 1.000 % Corporations .040 -.009 .050 -.019 -.208 .364 1.000 % Hired Labor -.133 -.095 -.160 -.080 .097 .266 .264 1.000 % Black .217 .271 .388 .238 -.034 -.098 -.039 -.081 1.000 % Hispanic .147 .137 .177 .124 -.105 .129 .349 .091 .087 1.000 Residential Mob. .364 .412 .368 .403 .012 -.091 -.002 -.156 .345 .219 % Poverty -.207 -.186 -.114 -.191 -.163 .143 -.265 -.035 .143 -.115 % Elderly -.401 -.424 -.332 -.422 -.192 .099 .077 .070 -.213 -.238 Indiana .057 .064 .052 .064 .033 -.110 .041 -.109 .014 -.029 Iowa -.088 -.070 -.030 -.074 .233 -.161 .109 .150 -.084 -.054 Kansas -.085 -.052 .022 -.063 -.168 .177 .092 .038 .100 .321 Michigan .075 .050 .089 .041 -.262 -.171 -.141 -.134 .028 -.043 Minnesota -.011 .028 -.008 .034 .170 -.125 -.132 .080 -.093 -.037 Missouri -.011 .016 .103 .000 .088 -.131 -.125 -.238 .285 -.065 Nebraska -.113 -.119 -.143 -.110 -.095 .276 .390 .144 -.112 .057 North Dakota -.116 -.134 -.157 -.125 -.086 .242 -.220 .080 -.090 -.088 Ohio .131 .102 .034 .110 .094 -.141 -.133 -.213 .116 -.019 South Dakota -.076 -.097 -.096 -.093 -.138 .259 -.007 .037 -.080 -.075 Wisconsin .322 .275 .157 .284 .128 -.155 .016 .056 -.067 -.064 1980 Dummy -.137 -.033 -.162 -.008 .159 -.038 -.298 .096 -.043 -.135 1990 Dummy -.004 .041 .014 .044 .000 -.012 -.037 .116 -.004 -.074 2000 Dummy .142 -.008 .148 -.036 -.160 .050 .335 -.212 .047 .209100

PAGE 101

Table H-1. Continued ResMob Poverty Elderly Indiana Iowa Kansas Michigan Minnesota Missouri Nebraska UCR Total Part 1 Total Violent Crime Property Crime # Farms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob 1.000 % Poverty -.032 1.000 % Elderly -.570 .086 1.000 Indiana .038 -.153 -.212 1.000 Iowa -.070 -.165 .091 -.090 1.000 Kansas .099 -.064 .139 -.093 -.157 1.000 Michigan .101 .012 -.116 -.072 -.121 -.126 1.000 Minnesota -.099 -.045 .003 -.079 -.134 -.138 -.107 1.000 Missouri .213 .256 .021 -.072 -.122 -.126 -.098 -.107 1.000 Nebraska -.070 -.024 .140 -.084 -.142 -.147 -.114 -.126 -.114 1.000 North Dakota -.231 .157 .099 -.066 -.111 -.115 -.089 -.098 -.090 -.105 Ohio .040 -.034 -.264 -.058 -.097 -.101 -.078 -.086 -.078 -.091 South Dakota -.003 .223 .000 -.060 -.102 -.105 -.082 -.090 -.082 -.096 Wisconsin -.011 -.116 -.058 -.067 -.114 -.118 -.091 -.100 -.091 -.107 1980 Dummy .195 .086 -.237 -.001 -.001 -.001 .005 .000 -.001 -.001 1990 Dummy -.128 .187 .140 .000 .000 .000 -.002 .002 .000 .000 2000 Dummy -.067 -.273 .097 .000 .000 .000 -.002 -.002 .000 .000101

PAGE 102

Table H-1. Continued ND Ohio SD Wisconsin 1980 1990 2000 UCR Total Part 1 Total Violent Crime Property Crime # Farms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mobility % Poverty % Elderly Indiana Iowa Kansas Michigan Minnesota Missouri Nebraska North Dakota 1.000 Ohio -.072 1.000 South Dakota -.075 -.065 1.000 Wisconsin -.084 -.073 -.076 1.000 1980 Dummy -.001 -.001 -.001 .001 1.000 1990 Dummy .000 .000 .000 -.003 -.501 1.000 2000 Dummy .000 .000 .000 .002 -.501 -.499 1.000102

PAGE 103

Table H-2. Correlation matrix of 1980 variables total sample UCR Part 1 Violent Property # Farm s Size % Corp % Hired % Black % Hisp UCR Total 1.000 Part 1 Total .939 1.000 Violent Crime .729 .779 1.000 Property Crime .937 .998 .733 1.000 # Farms .288 .289 .124 .300 1.000 Av. Farm Size -.193 -.220 -.176 -.219 -.345 1.000 % Corporations -.031 -.031 -.048 -.028 -.188 .546 1.000 % Hired Labor -.039 -.031 -.094 -.023 .100 .283 .302 1.000 % Black .267 .297 .529 .263 -.024 -.089 -.058 -.028 1.000 % Hispanic .134 .136 .224 .122 -.118 .158 .304 .144 .119 1.000 Residential Mob .423 .419 .448 .404 -.051 -.013 .073 -.094 .305 .299 % Poverty -.339 -.347 -.202 -.354 -.211 .185 -.114 -.025 .093 -.170 % Elderly -.456 -.440 -.394 -.434 -.104 -.003 -.014 -.035 -.170 -.301 Indiana .077 .064 .032 .066 .045 -.113 .048 -.096 .015 -.043 Iowa -.038 -.016 -.064 -.010 .256 -.150 .011 .165 -.082 -.090 Kansas -.080 -.060 .089 -.074 -.180 .152 .124 .051 .102 .378 Michigan .058 .066 .116 .059 -.271 -.151 -.173 -.136 .004 -.035 Minnesota -.056 .006 -.076 .015 .194 -.119 -.140 .101 -.098 -.092 Missouri -.030 -.014 .084 -.024 .071 -.121 -.105 -.241 .283 -.054 Nebraska -.116 -.133 -.152 -.127 -.114 .280 .434 .013 -.102 .058 North Dakota -.060 -.127 -.155 -.120 -.080 .198 -.238 .101 -.082 -.096 Ohio .174 .150 .153 .146 .093 -.127 -.123 -.161 .124 .031 South Dakota -.072 -.102 -.072 -.102 -.149 .252 .032 .034 -.076 -.076 Wisconsin .232 .226 .080 .237 .136 -.136 .036 .087 -.068 -.073103

PAGE 104

Table H-2. Continued ResMob Poverty Elderly Indiana Iowa Ka nsas Michigan Minnesota Missouri Nebraska UCR Total Part 1 Total Violent Crime Property Crime # Farms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob 1.000 % Poverty -.222 1.000 % Elderly -.503 .244 1.000 Indiana .007 -.164 -.219 1.000 Iowa -.041 -.146 .069 -.089 1.000 Kansas .090 -.118 .193 -.093 -.156 1.000 Michigan .071 -.050 -.145 -.073 -.122 -.127 1.000 Minnesota -.079 .033 -.001 -.079 -.133 -.138 -.109 1.000 Missouri .154 .204 .124 -.072 -.121 -.126 -.099 -.107 1.000 Nebraska -.090 .032 .153 -.084 -.142 -.147 -.115 -.125 -.114 1.000 North Dakota -.147 .151 -.020 -.066 -.111 -.115 -.090 -.098 -.089 -.104 Ohio .038 -.111 -.270 -.058 -.097 -.100 -.079 -.086 -.078 -.091 South Dakota .014 .339 -.032 -.060 -.101 -.105 -.082 -.090 -.082 -.095 Wisconsin -.011 -.130 -.031 -.067 -.114 -.118 -.092 -.101 -.091 -.107104

PAGE 105

Table H-2. Continued ND Ohio SD Wisconsin UCR Total Part 1 T otal Violent Crime Property Crime # F arms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob % Poverty % Elderly Indiana I ow105a Ka nsas Michigan Minnesota Missouri Nebraska North Dakota 1.000 Ohio -.071 1.000 South Dakota -.075 -.065 1.000 Wisconsin -.084 -.073 -.076 1.000

PAGE 106

Table H-3. Correlation matrix of 1990 variables total sample UCR Part 1 Violent Property # Farm s Size % Corp % Hired % Black % Hisp UCR Total 1.000 Part 1 Total .934 1.000 Violent Crime .694 .824 1.000 Property Crime .943 .996 .769 1.000 # Farms .307 .263 .178 .268 1.000 Av. Farm Size -.223 -.203 -.183 -.200 -.331 -1.000 % Corporations .015 .005 .003 .006 -.142 .349 1.000 % Hired Labor -.076 -.082 -.090 -.078 .142 .159 .326 1.000 % Black .195 .249 .370 .222 -.032 -.100 -.035 -.092 1.000 % Hispanic .163 .214 .227 .205 -.109 .150 .308 .135 .110 1.000 Residential Mob .400 .429 .416 .417 -.063 -.088 .068 -.223 .393 .271 % Poverty -.166 -.160 -.100 -.164 -.230 .107 -.292 -.165 .188 -.123 % Elderly -.449 -.438 -.381 -.433 -.168 .084 -.006 .098 -.247 -.307 Indiana .009 .007 .021 .005 .036 -.086 .073 -.101 .013 -.045 Iowa -.111 -.121 -.075 -.125 .235 -.166 .120 .260 -.085 -.085 Kansas -.044 .021 .134 .002 -.169 .182 .123 -.016 .106 .385 Michigan .077 .061 .102 .052 -.281 -.174 -.160 -.149 .025 -.030 Minnesota .015 .061 .056 .059 .162 -.130 -.163 .113 -.102 -.057 Missouri -.044 -.032 .008 -.037 .083 -.134 -.127 -.253 .282 -.071 Nebraska -.105 -.100 -.131 -.092 -.078 .266 .405 .113 -.109 .040 North Dakota -.139 -.128 -.161 -.118 -.089 .252 -.256 .053 -.087 -.094 Ohio .096 .035 -.006 .040 .097 -.145 -.137 -.227 .122 -.001 South Dakota -.061 -.078 -.091 -.073 -.135 .254 -.023 .004 -.079 -.077 Wisconsin .373 .309 .129 .327 .136 -.155 .030 .080 -.069 -.066106

PAGE 107

Table H-3. Continued ResMob Poverty Elderly Indiana Iowa Ka nsas Michigan Minnesota Missouri Nebraska UCR Total Part 1 Total Violent Crime Property Crime # Farms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob 1.000 % Poverty .027 1.000 % Elderly -.549 .087 1.000 Indiana .044 -.177 -.220 1.000 Iowa -.103 -.172 .113 -.090 1.000 Kansas .099 -.074 .139 -.093 -.157 1.000 Michigan .173 .081 -.131 -.072 -.121 -.125 1.000 Minnesota -.117 -.053 .008 -.080 -.134 -.139 -.107 1.000 Missouri .224 .291 .023 -.072 -.122 -.126 -.097 -.108 1.000 Nebraska -.079 -.082 .147 -.084 -.142 -.148 -.113 -.126 -.115 1.000 North Dakota -.265 .176 .098 -.066 -.112 -.116 -.089 -.099 -.090 -.105 Ohio .049 .004 -.283 -.058 -.097 -.101 -.078 -.086 -.078 -.092 South Dakota -.001 .163 -.009 -.060 -.102 -.106 -.081 -.090 -.082 -.096 Wisconsin -.008 -.106 -.056 -.067 -.113 -.117 -.090 -.100 -.091 -.106107

PAGE 108

Table H-3. Continued ND Ohio SD Wisconsin UCR Total Part 1 T otal Violent Crime Property Crime # F arms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob % Poverty % Elderly Indiana I ow108a Ka nsas Michigan Minnesota Missouri Nebraska North Dakota 1.000 Ohio -.072 1.000 South Dakota -.075 -.066 1.000 Wisconsin -.083 -.073 -.076 1.000

PAGE 109

Table H-4. Correlation matrix of 2000 variables total sample UCR Part 1 Violent Property # Farm s Size % Corp % Hired % Black % Hisp UCR Total 1.000 Part 1 Total .944 1.000 Violent Crime .765 .845 1.000 Property Crime .944 .992 .771 1.000 # Farms .326 .326 .262 .327 1.000 Av. Farm Size -.296 -.298 -.288 -.287 -.348 1.000 % Corporations -.039 -.023 -.019 -.022 -.162 .312 1.000 % Hired Labor -.173 -.180 -.200 -.168 -.056 .404 .449 1.000 % Black .208 .277 .364 .244 -.020 -.114 -.088 -.094 1.000 % Hispanic .100 .121 .104 .119 -.049 .114 .304 .163 .058 1.000 Residential Mob .432 .451 .447 .432 .096 -.175 .033 -.212 .386 .276 % Poverty -.121 -.106 -.016 -.123 -.197 .199 -.211 -.103 .206 .003 % Elderly -.470 -.463 -.415 -.454 -.238 .199 .041 .199 -.264 -.299 Indiana .090 .133 .096 .136 .017 -.130 .023 -.136 .014 -.022 Iowa -.106 -.059 .019 -.075 .220 -.168 .183 .038 -.085 -.036 Kansas -.129 -.133 -.102 -.135 -.163 .199 .072 .083 .092 .333 Michigan .093 .023 .079 .009 -.252 -.192 -.139 -.129 .056 -.060 Minnesota -.007 .011 -.032 .021 .160 -.126 -.135 .032 -.080 -.013 Missouri .024 .100 .207 .071 .124 -.142 -.166 -.237 .291 -.079 Nebraska -.130 -.133 -.168 -.119 -.096 .284 .447 .306 -.124 .077 North Dakota -.140 -.152 -.177 -.140 -.097 .281 -.238 .093 -.101 -.100 Ohio .150 .141 .016 .164 .101 -.154 -.168 -.261 .102 -.050 South Dakota -.098 -.116 -.126 -.109 -.139 .275 -.019 .075 -.085 -.089 Wisconsin .361 .286 .238 .285 .119 -.176 -.002 .008 -.065 -.073109

PAGE 110

Table H-4. Continued ResMob Poverty Elderly Indiana Iowa Ka nsas Michigan Minnesota Missouri Nebraska UCR Total Part 1 Total Violent Crime Property Crime # Farms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob 1.000 % Poverty .104 1.000 % Elderly -.607 -.013 1.000 Indiana .070 -.135 -.218 1.000 Iowa -.069 -.202 .098 -.090 1.000 Kansas .116 .001 .101 -.093 -.157 1.000 Michigan .057 .002 -.079 -.072 -.121 -.125 1.000 Minnesota -.109 -.134 .003 -.079 -.133 -.138 -.106 1.000 Missouri .284 .310 -.076 -.072 -.122 -.126 -.097 -.107 1.000 Nebraska -.043 -.025 .131 -.084 -.142 -.148 -.113 -.125 -.115 1.000 North Dakota -.307 .163 .218 -.066 -.112 -.116 -.089 -.098 -.090 -.105 Ohio .037 .004 -.264 -.058 -.097 -.101 -.078 -.086 -.078 -.092 South Dakota -.023 .194 .038 -.060 -.102 -.106 -.081 -.089 -.082 -.096 Wisconsin -.016 -.127 -.090 -.068 -.114 -.118 -.091 -.100 -.092 -.107110

PAGE 111

Table H-4. Continued ND Ohio SD Wisconsin UCR Total Part 1 T otal Violent Crime Property Crime # F arms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob % Poverty % Elderly Indiana I ow111a Ka nsas Michigan Minnesota Missouri Nebraska North Dakota 1.000 Ohio -.072 1.000 South Dakota -.075 -.066 1.000 Wisconsin -.084 -.073 -.077 1.000

PAGE 112

Table H-5. Correlation matrix of variables sub sample, all years UCR Part 1 Violent Property # Farms Size % Corp % Hired % Black % Hisp UCR Total 1.000 Part 1 Total .894 1.000 Violent Crime .733 .780 1.000 Property Crime .882 .994 .705 1.000 # Farms .252 .271 .148 .281 1.000 Av. Farm Size -.236 -.239 -.211 -.233 -.337 1.000 % Corporations .015 -.029 .040 -.040 -.198 .374 1.000 % Hired Labor -.105 -.060 -.141 -.043 .136 .255 .269 1.000 % Black .127 .168 .304 .135 -.042 -.092 -.032 -.049 1.000 % Hispanic .121 .110 .148 .098 -.115 .138 .334 .101 .065 1.000 Residential Mob. .338 .376 .330 .367 -.018 -.072 .015 -.139 .255 .218 % Poverty -.190 -.160 -.100 -.163 -.116 .123 -.290 -.065 .168 -.115 % Elderly -.383 -.395 -.305 -.393 -.143 .046 .050 -.002 -.139 -.258 Indiana -.001 -.007 .001 -.008 -.010 -.065 .032 -.059 -.010 -.028 Iowa -.056 -.031 .003 -.036 .283 -.186 .103 .142 -.071 -.062 Kansas -.124 -.094 -.039 -.100 -.179 .174 .105 .031 .048 .332 Michigan .155 .117 .145 .106 -.269 -.191 -.151 -.156 .061 -.040 Minnesota .037 .083 .021 .090 .194 -.142 -.137 .058 -.079 -.033 Missouri .004 .029 .132 .009 .100 -.145 -.131 -.259 .344 -.062 Nebraska -.127 -.140 -.157 -.130 -.082 .276 .391 .132 -.105 .031 North Dakota -.128 -.143 -.162 -.133 -.084 .233 -.232 .060 -.090 -.087 Ohio .029 .015 -.027 .022 .008 -.095 -.092 -.180 .055 .009 South Dakota -.088 -.106 -.104 -.102 -.128 .257 -.016 .037 -.072 -.072 Wisconsin .355 .307 .180 .316 .127 -.168 -.002 .041 -.062 -.071 1980 Dummy -.145 -.030 -.167 -.004 .152 -.036 -.295 .084 -.047 -.131 1990 Dummy -.002 .046 .008 .051 .001 -.015 -.039 .124 -.010 -.070 2000 Dummy .148 -.016 .159 -.047 -.153 .051 .335 -.208 .056 .201 112

PAGE 113

Table H-5. Continued ResMob Poverty Elderly Indiana Iowa Kansas Michigan Minnesota Missouri Nebraska UCR Total Part 1 Total Violent Crime Property Crime # Farms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob 1.000 % Poverty -.032 1.000 % Elderly -.545 .053 1.000 Indiana -.006 -.079 -.122 1.000 Iowa -.042 -.203 .049 -.059 1.000 Kansas .067 -.082 .146 -.057 -.171 1.000 Michigan .145 -.002 -.167 -.047 -.139 -.136 1.000 Minnesota -.076 -.040 -.027 -.049 -.145 -.142 -.116 1.000 Missouri .236 .261 .001 -.046 -.137 -.134 -.109 -.114 1.000 Nebraska -.063 -.041 .113 -.054 -.160 -.156 -.127 -.133 -.125 1.000 North Dakota -.260 .138 .098 -.041 -.123 -.120 -.098 -.102 -.096 -.112 Ohio -.022 -.005 -.172 -.023 -.067 -.066 -.054 -.056 -.053 -.061 South Dakota -.012 .236 -.025 -.038 -.113 -.110 -.090 -.094 -.088 -.103 Wisconsin .003 -.122 -.089 -.043 -.127 -.124 -.101 -.106 -.099 -.116 1980 Dummy .204 .103 -.254 .000 -.001 -.001 .005 .001 -.001 -.001 1990 Dummy -.134 .191 .151 .000 .001 .001 -.003 .002 .000 .001 2000 Dummy -.070 -.294 .104 .000 .001 .001 -.003 -.002 .000 .001113

PAGE 114

Table H-5. Continued ND Ohio SD Wisconsin 1980 1990 2000 UCR Total Part 1 Total Violent Crime Property Crime # Farms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mobility % Poverty % Elderly Indiana Iowa Kansas Michigan Minnesota Missouri Nebraska North Dakota 1.000 Ohio -.047 1.000 South Dakota -.079 -.043 1.000 Wisconsin -.089 -.049 -.082 1.000 1980 Dummy -.001 .000 -.001 .001 1.000 1990 Dummy .000 .000 .000 -.003 -.501 1.000 2000 Dummy .000 .000 .000 .002 -.501 -.499 1.000114

PAGE 115

Table H-6. Correlation matrix of 1980 variables sub sample UCR Part 1 Violent Property # Farm s Size % Corp % Hired % Black % Hisp UCR Total 1.000 Part 1 Total .936 1.000 Violent Crime .727 .790 1.000 Property Crime .935 .998 .750 1.000 # Farms .314 .280 .121 .290 1.000 Av. Farm Size -.201 -.220 -.194 -.217 -.337 1.000 % Corporations -.047 -.042 -.063 -.039 -.187 .558 1.000 % Hired Labor -.016 .002 -.087 .011 .141 .275 .304 1.000 % Black .110 .155 .373 .130 -.027 -.082 -.055 .002 1.000 % Hispanic .074 .080 .176 .068 -.143 .177 .308 .153 .064 1.000 Residential Mob .401 .379 .431 .364 -.084 .007 .091 -.077 .190 .280 % Poverty -.345 -.333 -.210 -.338 -.154 .160 -.129 -.059 .121 -.174 % Elderly -.422 -.387 -.351 -.382 -.045 -.065 -.040 -.109 -.067 -.313 Indiana .001 -.012 -.018 -.010 -.007 -.067 .049 -.035 -.018 -.037 Iowa .030 .043 -.024 .049 .308 -.174 .000 .158 -.067 -.102 Kansas -.133 -.100 .033 -.111 -.188 .148 .137 .042 .050 .396 Michigan .149 .142 .214 .131 -.278 -.169 -.181 -.153 .031 -.030 Minnesota -.005 .067 -.046 .077 .217 -.134 -.147 .083 -.083 -.082 Missouri -.021 -.017 .132 -.031 .084 -.133 -.110 -.255 .343 -.050 Nebraska -.135 -.150 -.178 -.144 -.101 .280 .432 .004 -.093 .031 North Dakota -.071 -.138 -.171 -.131 -.079 .189 -.241 .083 -.079 -.098 Ohio .030 .022 .017 .022 .006 -.085 -.086 -.140 .062 .061 South Dakota -.079 -.105 -.081 -.105 -.138 .247 .025 .030 -.067 -.074 Wisconsin .278 .260 .118 .268 .134 -.148 .023 .072 -.061 -.079115

PAGE 116

Table H-6. Continued ResMob Poverty Elderly Indiana Iowa Ka nsas Michigan Minnesota Missouri Nebraska UCR Total Part 1 Total Violent Crime Property Crime # Farms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob 1.000 % Poverty -.234 1.000 % Elderly -.475 .211 1.000 Indiana -.033 -.101 -.111 1.000 Iowa -.015 -.190 .025 -.059 1.000 Kansas .044 -.141 .207 -.057 -.170 1.000 Michigan .111 -.075 -.198 -.047 -.141 -.138 1.000 Minnesota -.046 .038 -.041 -.049 -.145 -.142 -.117 1.000 Missouri .171 .202 .116 -.046 -.136 -.133 -.110 -.114 1.000 Nebraska -.091 .021 .127 -.054 -.159 -.156 -.129 -.133 -.125 1.000 North Dakota -.169 .137 -.041 -.041 -.122 -.119 -.099 -.102 -.096 -.112 Ohio -.018 -.078 -.173 -.022 -.067 -.065 -.054 -.056 -.052 -.061 South Dakota .003 .359 -.061 -.038 -.112 -.110 -.091 -.094 -.088 -.103 Wisconsin .009 -.141 -.060 -.043 -.127 -.124 -.103 -.106 -.099 -.116116

PAGE 117

Table H-6. Continued ND Ohio SD Wisconsin UCR Total Part 1 T otal Violent Crime Property Crime # F arms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob % Poverty % Elderly Indiana I ow117a Ka nsas Michigan Minnesota Missouri Nebraska North Dakota 1.000 Ohio -.047 1.000 South Dakota -.079 -.043 1.000 Wisconsin -.089 -.049 -.082 1.000

PAGE 118

Table H-7. Correlation matrix of 1990 variables sub sample UCR Part 1 Violent Property # Farm s Size % Corp % Hired % Black % Hisp UCR Total 1.000 Part 1 Total .944 1.00 Violent Crime .747 .850 1.000 Property Crime .949 .997 .805 1.000 # Farms .297 .250 .193 .253 1.000 Av. Farm Size -.233 -.209 -.201 -.205 -.330 1.000 % Corporations -.013 -.012 .003 -.014 -.135 .363 1.000 % Hired Labor -.034 -.045 -.051 -.043 .197 .145 .322 1.000 % Black .102 .135 .251 .114 -.042 -.094 -.027 -.048 1.000 % Hispanic .142 .189 .210 .181 -.126 .162 .294 .143 .078 1.000 Residential Mob .373 .376 .360 .369 -.106 -.072 .094 -.185 .297 .269 % Poverty -.129 -.123 -.078 -.126 -.185 .086 -.317 -.213 .224 -.132 % Elderly -.445 -.416 -.366 -.413 -.113 .035 -.036 .001 -.164 -.334 Indiana -.025 -.033 -.007 -.036 -.007 -.052 .040 -.056 -.012 -.038 Iowa -.087 -.100 -.046 -.106 .286 -.190 .113 .256 -.071 -.097 Kansas -.092 -.040 .034 -.050 -.181 .182 .142 -.017 .051 .400 Michigan .153 .120 .163 .111 -.290 -.193 -.171 -.176 .061 -.028 Minnesota .066 .112 .104 .110 .187 -.146 -.168 .081 -.089 -.050 Missouri -.040 -.028 .023 -.035 .093 -.146 -.131 -.279 .347 -.069 Nebraska -.121 -.122 -.149 -.115 -.064 .268 .417 .093 -.102 .013 North Dakota -.155 -.135 -.171 -.126 -.087 .243 -.267 .040 -.089 -.095 Ohio .000 -.013 -.055 -.006 .011 -.097 -.088 -.183 .062 .028 South Dakota -.071 -.090 -.101 -.086 -.124 .252 -.036 .007 -.071 -.075 Wisconsin .407 .340 .173 .356 .135 -.167 .008 .062 -.064 -.073118

PAGE 119

Table H-7. Continued ResMob Poverty Elderly Indiana Iowa Ka nsas Michigan Minnesota Missouri Nebraska UCR Total Part 1 Total Violent Crime Property Crime # Farms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob 1.000 % Poverty .030 1.000 % Elderly -.510 .055 1.000 Indiana .008 -.093 -.132 1.000 Iowa -.078 -.215 .071 -.059 1.000 Kansas .069 -.091 .150 -.058 -.171 1.000 Michigan .227 .074 -.187 -.047 -.139 -.136 1.000 Minnesota -.096 -.050 -.024 -.049 -.146 -.143 -.116 1.000 Missouri .248 .306 .003 -.046 -.137 -.134 -.108 -.114 1.000 Nebraska -.068 -.106 .119 -.054 -.160 -.157 -.127 -.134 -.125 1.000 North Dakota -.302 .152 .097 -.041 -.123 -.120 -.097 -.103 -.096 -.113 Ohio -.023 .038 -.190 -.023 -.067 -.066 -.053 -.056 -.053 -.062 South Dakota -.008 .174 -.036 -.038 -.113 -.110 -.089 -.094 -.088 -.103 Wisconsin .005 -.112 -.088 -.042 -.126 -.123 -.100 -.105 -.099 -.115119

PAGE 120

Table H-7. Continued ND Ohio SD Wisconsin UCR Total Part 1 T otal Violent Crime Property Crime # F arms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob % Poverty % Elderly Indiana I ow120a Ka nsas Michigan Minnesota Missouri Nebraska North Dakota 1.000 Ohio -.047 1.000 South Dakota -.079 -.043 1.000 Wisconsin -.089 -.048 -.081 1.000

PAGE 121

Table H-8. Correlation matrix of 2000 variables sub sample UCR Part 1 Violent Property # Farm s Size % Corp % Hired % Black % Hisp UCR Total 1.000 Part 1 Total .927 1.000 Violent Crime .713 .831 1.000 Property Crime .932 .989 .740 1.000 # Farms .321 .331 .264 .331 1.000 Av. Farm Size -.311 -.308 -.290 -.295 -.338 1.000 % Corporations -.074 -.047 -.044 -.045 -.146 .316 1.000 % Hired Labor -.150 -.154 -.189 -.136 -.023 .392 .455 1.000 % Black .145 .225 .339 .182 -.028 -.112 -.086 -.066 1.000 % Hispanic .077 .102 .074 .103 -.058 .121 .289 .173 .047 1.000 Residential Mob .418 .442 .429 .421 .083 -.157 .052 -.208 .330 .286 % Poverty -.100 -.084 .010 -.105 -.164 .189 -.234 -.134 .238 .011 % Elderly -.466 -.456 -.402 -.445 -.196 .147 -.001 .138 -.224 -.331 Indiana .018 .031 .015 .034 -.018 -.076 .024 -.089 .001 -.025 Iowa -.081 -.015 .052 -.032 .268 -.196 .177 .022 -.074 -.043 Kansas -.161 -.161 -.133 -.160 -.176 .196 .084 .069 .044 .346 Michigan .181 .093 .129 .078 -.257 -.215 -.150 -.152 .092 -.058 Minnesota .039 .064 -.013 .080 .183 -.146 -.140 .014 -.065 -.013 Missouri .054 .144 .238 .112 .136 -.157 -.175 -.261 .345 -.076 Nebraska -.142 -.156 -.181 -.141 -.082 .282 .440 .301 -.121 .047 North Dakota -.151 -.162 -.178 -.149 -.095 .271 -.258 .062 -.102 -.099 Ohio .056 .044 -.027 .060 .009 -.105 -.120 -.228 .042 -.018 South Dakota -.115 -.132 -.133 -.124 -.130 .274 -.029 .076 -.080 -.086 Wisconsin .392 .321 .241 .325 .119 -.191 -.026 -.007 -.062 -.083121

PAGE 122

Table H-8. Continued ResMob Poverty Elderly Indiana Iowa Ka nsas Michigan Minnesota Missouri Nebraska UCR Total Part 1 Total Violent Crime Property Crime # Farms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob 1.000 % Poverty .108 1.000 % Elderly -.585 -.039 1.000 Indiana .009 -.052 -.135 1.000 Iowa -.036 -.238 .054 -.059 1.000 Kansas .097 -.017 .100 -.058 -.171 1.000 Michigan .101 -.007 -.129 -.047 -.139 -.136 1.000 Minnesota -.094 -.129 -.020 -.049 -.145 -.141 -.114 1.000 Missouri .317 .323 -.110 -.046 -.137 -.134 -.108 -.113 1.000 Nebraska -.031 -.044 .103 -.054 -.160 -.157 -.127 -.132 -.125 1.000 North Dakota -.339 .145 .237 -.041 -.123 -.120 -.097 -.101 -.096 -.113 Ohio -.027 .029 -.170 -.023 -.067 -.066 -.053 -.056 -.053 -.062 South Dakota -.035 .205 .016 -.038 -.113 -.110 -.089 -.093 -.088 -.103 Wisconsin -.008 -.131 -.124 -.043 -.128 -.125 -.101 -.105 -.100 -.117122

PAGE 123

123Table H-8. Continued ND Ohio SD Wisconsin UCR Total Part 1 T otal Violent Crime Property Crime # F arms Av. Farm Size % Corporations % Hired Labor % Black % Hispanic Residential Mob % Poverty % Elderly Indiana I ow a Ka nsas Michigan Minnesota Missouri Nebraska North Dakota 1.000 Ohio -.047 1.000 South Dakota -.079 -.043 1.000 Wisconsin -.090 -.049 -.082 1.000

PAGE 124

APPENDIX I SUB SAMPLE ANALYSES Table I-1. Descriptive statistics (means with standard deviations in parentheses) for sub sample Notes: (a) 3 cases missing, (b) 3 cases missing, (c) 6 cases missing Time 1 1980 Time 2 1990 Time 3 2000 All 3 Times UCR total 395.3439 (470.64) 546.9163 (757.61) 706.7891 (925.46) 549.3846 (751.99) Part 1 total 80.0976 (105.20) 93.0530 (142.28) 82.4441 (112.09) 85.1884 (120.96) Violent 7.5761 (9.93) 12.3272 (19.32) 16.4218 (24.49) 12.0996 (19.22) Property 72.4987 (97.56) 80.7248 (126.27) 66.0120 (92.75) 73.0774 (106.64) % Black .5044 (1.75) .5973 (1.78) .7656 (1.84) .6222 (1.79) % Hispanic .8603 (1.67) 1.1406 (2.46) 2.4053 (4.74) 1.4676 (3.29) Residential mobility 40.0288 (6.89) 36.8340 (6.76) 37.4371 (6.01) 38.1037 (6.71) % Poverty 14.1285 (4.61) 14.7070 (4.64) 11.5247 (3.99) 13.4547 (4.63) Number of farms 840.92 (475.06) 750.30 (412.11) 657.48 (364.31) 749.75 (426.10) Average size of farms 610.89 (782.76) 632.37 (698.85) 702.04 (733.64) 648.36 (739.85) % Corporations 2.0466 (2.18) 3.2048 (2.74) 4.9029 (3.81) 3.3822 (3.21) % Hired labor 40.2113 (9.24) 40.7801 (9.84) 36.1465 (9.79) 39.0482 (9.84) % Elderly 16.4182 (3.56) 18.6370 (3.86) 18.38 (3.83) 17.8101 (3.88) N 519 516a 516b 1551c 124

PAGE 125

125 Table I-2. Variance inflation factor values for in dependent variables includ ed in all models, sub sample Time 1 1980 Time 2 1990 Time 3 2000 All 3 Times % Black 1.262 1.306 1.330 1.294 % Hispanic 1.686 1.670 1.584 1.551 Residential mobility 1.697 1.928 2.133 1.899 % Poverty 1.611 1.461 1.468 1.544 Number of farms 1.663 1.711 1.607 1.676 Average size of farms 2.367 1.951 2.149 2.002 % Corporations 2.523 2.164 2.193 2.397 % Hired labor 1.470 1.532 1.703 1.553 % Elderly 2.145 2.119 2.209 2.145 Indiana 1.336 1.304 1.301 1.282 Iowa 2.504 2.482 2.403 2.336 Michigan 2.314 2.264 2.108 2.118 Minnesota 2.234 2.093 1.969 1.994 Missouri 2.104 2.149 2.174 2.097 Nebraska 2.115 2.040 2.051 1.992 North Dakota 2.140 2.041 1.874 1.893 Ohio 1.398 1.443 1.457 1.408 South Dakota 2.022 1.744 1.650 1.717 Wisconsin 1.998 1.946 1.979 1.893 1990 dummy 1.531 2000 dummy 2.147

PAGE 126

Table I-3. Means and overall, be tween and within sample standard deviations for indicators included in the time-series models, sub sample (N = 1557, n = 519, T = 3) Overall mean Overall SD Between SD Within SD UCR total 548.8654 751.197 689.7247 298.6441 Part 1 85.04185 120.7921 113.9259 40.35211 Violent 12.11914 19.25894 16.02357 10.69963 Property 72.91115 106.4821 100.3338 35.84019 % Black .6271757 1.799923 1.759972 .3823622 % Hispanic 1.465405 3.291061 2.844824 1.657849 Residential mobility 38.08853 6.710935 6.12066 2.760844 % Poverty 13.48224 4.715358 4.070419 2.384866 Number of farms 746.9236 427.7022 416.5757 98.06577 Average size of farms 648.3604 739.8518 731.3509 109.9026 % Corporations 3.373285 3.208267 2.768152 1.6424856 % Hired labor 38.97564 9.97253 8.506606 5.213638 % Elderly 17.81166 3.892193 3.60781 1.466152 126

PAGE 127

Table I-4. Pooled cross-sectional time series negative binomial regression results (and z-scores), 1980-2000, sub sample (N = 1557, n = 519, T = 3) UCR Total Part 1 Violent Property % Black SE -.028339 (-1.35) .0209334 -.0069699 (-0.32) .0218006 -.0295467 (-1.05) .028098 -.0048255 (-0.21) .02275 % Hispanic SE -.0138078*** (-2.78) .0049679 -.0095236* (-1.75) .0054449 -.015317** (-2.13) .0071766 -.0080691 (-1.39) .0057953 Residential mobility SE -.0057653 (-1.35) .0042567 -.008893* (-1.84) .0048308 -.0056442 (-0.84) .0067256 -.0100419** (-1.98) .0050721 % Below poverty SE .0048982 (0.94) .0052219 .0065462 (1.06) .006164 -.0026987 (-0.31) .0086252 .0086312 (1.32) .0065155 Number of farms SE -.0004938*** (-4.86) .0001017 -.0007573*** (-6.53) .0001159 -.0009281*** (-6.38) .0001454 -.0007042*** (-5.92) .0001189 Average farm size SE .0002043*** (2.85) .0000717 .0001034 (0.94) .0001104 7.27e-06 (0.04) .0001859 .0001054 (0.95) .0001112 % Corporate farms SE -.0080454 (-0.97) .0082621 -.0032858 (-0.33) .0100904 .0019623 (0.15) .0133062 -.0039672 (-0.37) .0107644 % Hired labor SE -.007917 (-0.41) .0019519 .0003087 (0.14) .0022521 .00133 (0.43) .0030646 .0001099 (0.05) .0023774 % Elderly SE -.0104265 (-1.10) .0094389 -.0096103 (-0.87) .0110031 .0174607 (1.15) .0152055 -.0136853 (-1.19) .0114982 1990 Dummy SE .2708152*** (7.63) .0355057 .0404595 (1.03) .0393634 .2760142*** (5.02) .0549639 .0011978 (0.03) .040912 2000 Dummy SE .4436488*** (9.90) .0047983 -.2489798*** (-4.74) .0524789 .3330952*** (4.76) .0700337 -.3421539*** (-6.21) .0550846 Indiana SE -1.363439*** (-4.07) .3348605 -1.798624*** (-5.08) .3541247 -1.894399*** (-4.20) .4510923 -1.686553*** (-4.76) .354134 p < .10 ** p < .05 *** p < .01 127

PAGE 128

128 Table I-4. Continued UCR Total Part 1 Violent Property Iowa SE -.8378883*** (-4.66) .1799008 -1.463008*** (-6.87) .2129218 -1.222987*** (-4.18) .2924429 -1.389928*** (-6.54) .2125604 Michigan SE -.1410377 (-0.69) .2047797 -.3826761 (-1.61) .2373088 -.010079 (-0.03) .3833309 -.3529126 (-1.49) .2367441 Minnesota SE -.6754803*** (-3.50) .192904 -.6307844*** (-2.75) .2293834 -.4735073 (-1.44) .3295894 -.5478487** (-2.36) .2318657 Missouri SE -.9892079*** (-5.01) .1973163 -1.21166*** (-5.37) .2258044 -1.125353*** (-3.70) .3044096 -1.005274*** (-4.43) .2268056 Nebraska SE .1232473 (0.67) .1829879 -.0134738 (-0.06) .2237493 .1455135 (0.34) .433992 .0635809 (0.28) .2254141 North Dakota SE -1.028355*** (-5.08) .2024293 -.2525517 (-0.88) .2858342 -.4367297 (-0.82) .5339153 -.0571368 (-0.20) .2902809 Ohio SE -1.388486*** (-4.48) .3097298 -1.517225*** (-4.35) .3491161 -1.750663*** (-3.40) .5141653 -1.495179*** (-4.20) .3563562 South Dakota SE -.5029255** (-2.32) .2166606 -.0881448 (-0.32) .273484 .1393212 (0.29) .4859041 .0356834 (0.13) .2828991 Wisconsin SE -.10395 (-0.50) .2070572 -.0587255 (-0.25) .2375454 -1.273974*** (-4.07) .3133223 .0630091 (0.26) .2381917 Constant -6.763229*** -6.272453*** 7.001916*** -6.369946*** Log Likelihood -6028.5609 -4518.1725 -2449.4663 -4018.2986 N Observations/Groups 1550 / 518 a 1547 / 517 b 1532 / 512 c 1547 / 517 d Hausman Test 273.94*** 381.64*** 179.74*** 331.47*** Notes: (a) 1 group (1observation) was dropped because of only one observation per group; (b) 1 group (1 observation) was dropped because of only one observation per group, and 1 group (3 observations) were dropped due to all zero out comes; (c) 1 group (1 observation) was dropped because of only one observation per group, and 6 groups (18 observations) dropped due to all zero outcomes; (d) 1 group (1 observation) wa s dropped because of only one observation per group, and 1 group (3 observations) dropped due to all zero outcomes. p < .10 ** p < .05 *** p < .01

PAGE 129

LIST OF REFERENCES Agresti, A. (1996). An introduction to categorical data analysis. New York: John Wiley & Sons. Agresti, A. (2002). Negative binomial regression. In Categorical data analysis (2nd ed., pp. 559563). New York: Wiley. Albrecht, D. E. (1997). The changi ng structure of U.S. agriculture: Dualism out, industrialism in. Rural Sociology, 62, 474-490. Albrecht, D. E., Albrecht, C. M., & Albrecht, S. L. (2000). Poverty in nonmetropolitan America: Impacts of industrial, employment, and family structure variables. Rural Sociology, 65, 87103. Allison, P. D. (1990). Change scores as dependent variables in regression analysis. In C. Clogg (Ed.), Sociological Methodology (pp. 93-114). Allison, P. D. (1994). Using panel data to estimate the effects of events. Sociological Methods and Research, 23, 179-199. Allsion, P. D. (1999). Multiple regression: A primer. Thousand Oaks, CA: Pine Forge Press. Arthur, J. A. (1991). Socioeconomic predictors of crime in rural Georgia. Criminal Justice Review, 16(1), 29-41. Baker, P. L., & Hotek, D. R. (2003). Perhaps a blessing: Skills and contributions of recent Mexican immigrants in the rural Midwest. Hispanic Journal of Behavioral Sciences, 25, 448-468. Barkema, A., & Drabenstott, M. (1996). Consolid ation and change in heartland agriculture. In Economic forces shaping the rural heartland (pp. 61-77). Kansas City: Federal Reserve Bank. Barnes, D., & Blevins, A. (1992). Farm st ructure and the economic well being of nonmetropolitan counties. Rural Sociology, 57(3), 333-346. Barnett, C., & Mencken, F. C. ( 2002). Social disorganization theory and the contextual nature of crime in nonmetropolitan counties. Rural Sociology, 67(3), 372-393. Bartkowski, J. P., & Regis, H. A. (2003). Charitable choices: Religion, race, and poverty in the post-welfare era. New York: New York University Press. Brasier, K. J. (2005). Spatial analysis of change s in the number of farms during the farm crisis. Rural Sociology, 70(4), 540-560. Broadway, M. (1990). Meatpacking and its social and economic c onsequences for Garden City, Kansas in the 1980s. Urban Anthropology, 19(4), 321-344. 129

PAGE 130

Brown, D. L., & Swanson, L. E. (Eds.) (2003). Challenges for rural America in the twenty-first century. University Park, Pennsylvania: The Pe nnsylvania State University Press. Burgess, E.W. (1925). The growth of the city. In R. E. Park & E. W. Burgess (Eds.), The city. Chicago: University of Chicago Press. Buttel, F. H., & Larson, O. W. III. (1979). Farm size, structure, and energy intensity: An ecological analysis of US agriculture. Rural Sociology, 44(30), 471-488. Buttel, F. H., Lancelle, M., & Lee, D. R. ( 1988). Farm structure and rural communities in the Northeast. In L. E. Swanson (Ed.), Agriculture and community change in the U.S. The congressional research reports (pp. 181-257). Boulder, CO: Westview Press. Butterfield, F. (2005, February 13). Social is olation, guns and a culture of suicide. The New York Times, section 1, p. 28. Cameron, A. C., & Trivedi, P. K. (1998). Regression analysis of count data. Cambridge, UK: Cambridge University Press. Cebulak, W. (2004). Why rural crim e and justice really matter. Journal of Police and Criminal Psychology, 19(1), 71-81. Census of Agriculture. (1978). Volume 1, Geographic area series. Part 14, Indiana, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1978). Volume 1, Geographic area series. Part 15, Iowa, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1978). Volume 1, Geographic area series. Part 16, Kansas, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1978). Volume 1, Geographic area series Part 22, Michigan, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1978). Volume 1, Geographic area series Part 23, Minnesota, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1978). Volume 1, Geographic area series Part 25, Missouri, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1978). Volume 1, Geographic area series. Part 27, Nebraska, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1978). Volume 1, Geographic area series. Part 34, North Dakota, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1978). Volume 1, Geographic area series. Part 35, Ohio, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. 130

PAGE 131

Census of Agriculture. (1978). Volume 1, Geographic area series Part 41, South Dakota, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1978). Volume 1, Geographic area series Part 49, Wisconsin, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series. Part 14, Indiana, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series. Part 15, Iowa, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series. Part 16, Kansas, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series Part 22, Michigan, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series Part 23, Minnesota, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series Part 25, Missouri, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series. Part 27, Nebraska, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series. Part 34, North Dakota, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series. Part 35, Ohio, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series Part 41, South Dakota, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1987). Volume 1, Geographic area series Part 49, Wisconsin, state and county data. Washington, DC: US Department of Commerce, Bureau of the Census. Census of Agriculture. (1997). Volume 1, Geographic area series. Part 14, Indiana, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. Census of Agriculture. (1997). Volume 1, Geographic area series. Part 15, Iowa, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. 131

PAGE 132

Census of Agriculture. (1997). Volume 1, Geographic area series. Part 16, Kansas, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. Census of Agriculture. (1997). Volume 1, Geographic area series Part 22, Michigan, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. Census of Agriculture. (1997). Volume 1, Geographic area series Part 23, Minnesota, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. Census of Agriculture. (1997). Volume 1, Geographic area series Part 25, Missouri, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. Census of Agriculture. (1997). Volume 1, Geographic area series. Part 27, Nebraska, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. Census of Agriculture. (1997). Volume 1, Geographic area series. Part 34, North Dakota, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. Census of Agriculture. (1997). Volume 1, Geographic area series. Part 35, Ohio, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. Census of Agriculture. (1997). Volume 1, Geographic area series Part 41, South Dakota, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. Census of Agriculture. (1997). Volume 1, Geographic area series Part 49, Wisconsin, state and county data. Washington, DC: US Department of Agriculture, National Agricultural Statistics Service. CensusCD 1980 (1999). [Computer software]. New Brunswick N.J.: GeoLytics, Inc. Census 2000 Summary File 1 (Indiana). Prepared by the U.S. Census Bureau, 2001. Census 2000 Summary File 1 (Iowa). Prepared by the U.S. Census Bureau, 2001. Census 2000 Summary File 1 (Kansas). Prepared by the U.S. Census Bureau, 2001. Census 2000 Summary File 1 (Michigan). Prepared by the U.S. Census Bureau, 2001. Census 2000 Summary File 1 (Minnesota). Prepared by the U.S. Census Bureau, 2001. 132

PAGE 133

Census 2000 Summary File 1 (Missouri). Prepared by the U.S. Census Bureau, 2001. Census 2000 Summary File 1 (Nebraska). Prepared by the U.S. Census Bureau, 2001. Census 2000 Summary File 1 (North Dakota). Prepared by the U.S. Census Bureau, 2001. Census 2000 Summary File 1 (Ohio). Prepared by the U.S. Census Bureau, 2001. Census 2000 Summary File 1 (South Dakota). Prepared by the U.S. Census Bureau, 2001. Census 2000 Summary File 1 (Wisconsin). Prepared by the U.S. Census Bureau, 2001. Census 2000 Summary File 3 (Indiana). Prepared by the U.S. Census Bureau, 2002. Census 2000 Summary File 3 (Iowa). Prepared by the U.S. Census Bureau, 2002. Census 2000 Summary File 3 (Kansas). Prepared by the U.S. Census Bureau, 2002. Census 2000 Summary File 3 (Michigan). Prepared by the U.S. Census Bureau, 2002. Census 2000 Summary File 3 (Minnesota). Prepared by the U.S. Census Bureau, 2002. Census 2000 Summary File 3 (Missouri). Prepared by the U.S. Census Bureau, 2002. Census 2000 Summary File 3 (Nebraska). Prepared by the U.S. Census Bureau, 2002. Census 2000 Summary File 3 (North Dakota). Prepared by the U.S. Census Bureau, 2002. Census 2000 Summary File 3 (Ohio). Prepared by the U.S. Census Bureau, 2002. Census 2000 Summary File 3 (South Dakota). Prepared by the U.S. Census Bureau, 2002. Census 2000 Summary File 3 (Wisconsin). Prepared by the U.S. Census Bureau, 2002. Census of Population and Housing, 1990: Su mmary Tape File 1 on CD-ROM (Indiana) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1991. Census of Population and Housing, 1990: Summary Tape File 1 on CD-ROM (Iowa) [machinereadable data files]. Prepared by the Bureau of the Census, 1991. Census of Population and Housing, 1990: Su mmary Tape File 1 on CD-ROM (Kansas) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1991. Census of Population and Housing, 1990: Su mmary Tape File 1 on CD-ROM (Michigan) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1991. Census of Population and Housing, 1990: Summary Tape File 1 on CD-ROM (Minnesota) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1991. 133

PAGE 134

Census of Population and Housing, 1990: Su mmary Tape File 1 on CD-ROM (Missouri) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1991. Census of Population and Housing, 1990: Su mmary Tape File 1 on CD-ROM (Nebraska) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1991. Census of Population and Housing, 1990: Summary Tape File 1 on CD-ROM (North Dakota) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1991. Census of Population and Housing, 1990: Summary Tape File 1 on CD-ROM (Ohio) [machinereadable data files]. Prepared by the Bureau of the Census, 1991. Census of Population and Housing, 1990: Summary Tape File 1 on CD-ROM (South Dakota) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1991. Census of Population and Housing, 1990: Summary Tape File 1 on CD-ROM (Wisconsin) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1991. Census of Population and Housing, 1990: Su mmary Tape File 3 on CD-ROM (Indiana) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1992 Census of Population and Housing, 1990: Summary Tape File 3 on CD-ROM (Iowa) [machinereadable data files]. Prepared by the Bureau of the Census, 1992. Census of Population and Housing, 1990: Su mmary Tape File 3 on CD-ROM (Kansas) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1992. Census of Population and Housing, 1990: Su mmary Tape File 3 on CD-ROM (Michigan) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1992. Census of Population and Housing, 1990: Summary Tape File 3 on CD-ROM (Minnesota) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1992. Census of Population and Housing, 1990: Su mmary Tape File 3 on CD-ROM (Missouri) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1992. Census of Population and Housing, 1990: Su mmary Tape File 3 on CD-ROM (Nebraska) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1992. Census of Population and Housing, 1990: Summary Tape File 3 on CD-ROM (North Dakota) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1992. Census of Population and Housing, 1990: Summary Tape File 3 on CD-ROM (Ohio) [machinereadable data files]. Prepared by the Bureau of the Census, 1992. Census of Population and Housing, 1990: Summary Tape File 3 on CD-ROM (South Dakota) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1992. 134

PAGE 135

Census of Population and Housing, 1990: Summary Tape File 3 on CD-ROM (Wisconsin) [machine-readable data files]. Prepar ed by the Bureau of the Census, 1992. Chin, H. C., & Quddus, M. A. (2003). Modeling c ount data with excess zeroes: An empirical application to traffic accidents. Sociological Methods and Research, 32(1), 90-116. Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Coulter, P. (1989). Measuring inequality: A methodological handbook. Boulder, CO: Westview Press. Crowly, M. L. (1999). The impact of farm sector concen tration on poverty and inequality: An analysis of North Central U.S. counties. Masters Thesis, Department of Sociology, The Ohio State University, Columbus OH. Crowley, M. & Roscigno, V. (2004). Farm c oncentration, political-economic process, and stratification: the case of the north central U.S. Journal of Political and Military Sociology, 32, 133-155. Davidson, O. G. (1996). Broken heartland: The rise of Americas rural ghetto. Iowa City, IA: University of Iowa Press. Davis, K., Taylor, B., & Furniss, D. (2001). Na rrative accounts of trackin g the rural domestic violence survivors journe y: A feminist approach. Health Care for Women International, 22(4), 333-347. Diala, C. C., Muntaner, C., & Walrath, C. (2004). Gender, occupation, and socioeconomic correlates of alcohol and drug abuse among U.S. rural, metro politan, and urban residents. American Journal of Drug and Alcohol Abuse, 30(2), 409-428. Donnermeyer, J. F. (1994). Crime and violen ce in rural communities. Retrieved from http://www.ncrel.org/sdrs/areas/i ssues/envrnmnt/drugfree/v1donner.htm Donnermeyer, J. F., Barclay, E. M., & Jobes, P. C. (2002). Drug-related offenses and the structure of communities in rural Australia. Substance Use and Misuse, 37, 631-661. Drabenstott, M., & Smith, T. R. (1996). The ch anging economy of the rural heartland. In Economic Forces Shaping the Rural Heartland (pp. 1-11). Kansas City: Federal Reserve Bank. Durrenberger, E. P., & Thu, K. M. (1996). The expa nsion of large scale hog farming in Iowa: the applicability of Goldschmidts findings fifty years later. Human Organizations, 55(4), 409415. Economic Resource Service. (2003, August) Measuring rurality: New definitions in 2003. Retrieved from http://www.ers.usda.gov/Briefing/Rurality/Newdefinitions/ 135

PAGE 136

Economic Resource Service. (2004, April 28). Measuring rurality: Rural urban continuum codes. Retrieved from http://www.ers.usda.gov/brie fing/rurality/RuralUrbCon/ Egan, T., (2002, December 8). The nation: Pa storal poverty; The seeds of decline. The New York Times, section 4, p. 1. Ellison, C. G., & Sherkat, D. E. (1995). The semi-involuntary institution revisited: Regional variations in church partic ipation among black Americans. Social Forces, 73, 1415-1437. Firebaugh, G., & Beck, F. D. (1994). Does economic growth benefit the masses? American Sociological Review, 59(5), 631-653. Fisher, J. C., & Mason, R. L. (1981). The analysis of multicollinear data in criminology. In J. A. Fox (Ed.), Methods in quantitative criminology (pp. 99-125). New York: Academic Press. Flora, J. L., Brown, I., & Conby, J. L. (1977, August). Impact of type of agriculture on class structure, social well-being, and inequalities. Paper presented at the Rural Sociological Society annual meeting, Burlington, VT. Flora, C. B., & Flora, J. L. (1988). Public po licy, farm size, and community well-being in farming dependent counties of the plains. Pp. 76-129 in Agriculture and community change in the U.S.: The congr essional research reports, edited by Louis E. Swanson. Boulder, CO: Westview Press. Fujimoto, I. (1977). The communities of the San Joaquin Valley: The relation between scale of farming, water use, and quality of life. Pp. 480-500 in U.S. Congress, House of Representatives, obstacles to strengthening the family farm system. Hearings before the subcommittee on family farms, rural development, and special studies of the committee on agriculture, 95th Congress, first session. Washington, DC: U.S. Government Printing Office. Gardner, W., Mulvey, E. P., & Shaw, E. C. ( 1995). Regression analyses of counts and rates: Poisson, Overdispersed Poisson, and negative binomial models. Psychological Bulletin, 118, 392-404. Gilles, J. L. (1980). Farm size, farm structure, energy and climate: An alternative ecological analysis of United States agriculture. Rural Sociology, 45(5), 332-339. Gilles, J. L., & Dalecki, M. (1988). Rural well -being and agricultural change in two farming regions. Rural Sociology, 53(1), 40-55. Goldschmidt, W. (1946). Small business and the community: a study in central valley of California on effects of scale of farm operations, report of the Special Committee to Study Problems of American Small Business. Washington DC: 79th Congress, 2nd session, 1946. Goldschmidt, W. (1947).As you sow. Glencoe, IL: Free Press. 136

PAGE 137

Goldschmidt, W. (1978a). As you sow: Three studies in the social consequences of agribusiness. Montclair, NJ: Allanheld Osmun and Company. Goldschmidt, W. (1978b). Large-scale fa rming and the rural social structure. Rural Sociology, 45, 362-366. Green, G. P. (1985). Large-scale farming and the quality of life in rural communities: Further specification of the Goldschmidt hypothesis. Rural Sociology, 50, 262-273. Greene, W. (2000). Econometric analysis. Saddlebrook: Prentice Hall. Harris, C. K., & Gilbert, J. (1982). Large sc ale farming, rural income, and Goldschmidts agrarian thesis. Rural Sociology, 47(3), 449-458. Hayes, M. N., & Olmstead, A. L. (1984). Farm size and community quality: Arvin and Dinuba revisited. American Journal of Agricultural Economics, 66(4), 430-436. Hausman, J., Hall, B. H., & Griliches, Z. ( 1984). Econometric models for count data with an application to the pate nts-R & D relationship. Econometrica, 52(4), 909-938. Heady, E. O., & Sonka, S. T. (1974). Farm size, rural community income, and consumer welfare. American Journal of Agricultural Economics, 56(3), 534-542. Heaton, T. B., & Brown, D. L. (1982). Farm structure and energy in tensity: Another look. Rural Sociology, 47(1), 17-31. Hsaio, C. (1986). Analysis of panel data. Cambridge: Cambridge University Press. Irwin, M, Tolbert, C., & Lyson, T. (1999). Ther es no place like home: Non-migration and civic engagement. Environment and Planning, A 31, 2223-2238. Jobes, P. C. (2002). Effective officer a nd good neighbour: Problems and perceptions among police in rural Australia. Policing, 25(2), 256-273. Jobes, P. C. (2003). Human ecology and rural po licing: A grounded theoreti cal analysis of how personal constraints and community characteris tics influence strategies of law enforcement in rural New South Wales, Australia. Police Practice and Research, 4(1), 3-19. Jobes, P. C., Barclay, E., Weinand, H., & Donne rmeyer, J. F. (2004). A st ructural analysis of social disorganisation and crime in rural communities in Australia. Australian and New Zealand Journal of Criminology, 37(1), 114-140. Johnson, D. R. (1995). Alternative methods for the quantitative analysis of panel data in family research: Pooled time-series models. Journal of Marriage and the Family, 57(4):10651077. 137

PAGE 138

Jolliffe, D. (2003). Nonmetro poverty: Assess ing the effect of the 1990s. Amber Waves (September). Washington, DC: U.S. Depart ment of Agriculture, Economic Research Service. Kandel, W., & Cromartie, J. (2004). New patterns of Hispanic settlement in rural America. Rural Development Research Report Number 99. Wash ington, DC: United States Department of Agriculture. Kandel, W., & Newman, C. (2004). Rural Hispan ics: Employment and residential trends. Amber Waves, (June). Washington, DC: US Department of Agriculture, Economic Research Service. Kessler, R. C., & Greenberg, D. F. (1981). Linear panel analysis: Mode ls of quantitative change. New York: Academic Press. Kposowa, A. J., Breault, K. D., & Harrison, B. M. (1995). Reassessing the structural covariates of violent and property crime in th e USA: A county level analysis. British Journal of Sociology, 46(1), 79-105. Krishnan, S. P., Hilbert, J. C., & VanLeeuwen, D. (2001). Domestic violence and help-seeking behaviors among rural women: Resu lts from a shelter-based study. Family and Community Health, 24(1), 28-38. Land, K. C., McCall, P. L., & Cohen, L. E. (1990) Structural covariates of homicide rates: Are there any invariances across time and social space? American Journal of Sociology, 95(4), 922-963. Lansley et al. (1995). Beyond the amber waves of grain: An examination of social and economic restructuring in the heartland. Boulder, CO: Westview Press. Lee, M. R., & Bartkowski, J. P. (2004a). Love thy neighbor? Moral communities, civic engagement, and juvenile homicide in rural areas. Social Forces, 82(3), 1001-1035. Lee, M. R., & Bartkowski, J. P. (2004b). Civic pa rticipation, regional subc ultures, and violence: The differential effects of secular and reli gious participation on adult and juvenile homicide. Homicide Studies, 7, 1-35. Lee, M. R., & Ousey, G. C. (2001). Size matters: examining the link between small manufacturing, socioeconomic deprivation, a nd crime rates in nonmetropolitan counties. Sociological Quarterly, 42(4), 581-602. Lee, M. R., Maume, M. O., & Ousey, G. C. ( 2003). Social isolation and lethal violence across the metro/nonmetro divide: The effects of socioeconomic disadvantage and poverty concentration on homicide. Rural Sociology, 68(1), 107-131. Leistritz, F. L., & Eckstrom, B. L. (1988). Th e financial characteristics of production units and producers experiencing financial stress. In S. H. Murdock and F. Leistritz (Eds.), The farm financial crisis (pp. 73-95). Boulder, CO: Westview Press. 138

PAGE 139

Lewis-Beck, M. S. (1990). Applied regression : An introduction. Beverly Hills, CA: Sage Publications. Liederbach, J., & Frank, J. ( 2003). Policing Mayberry: The work routines of small-town and rural officers. American Journal of Criminal Justice, 28(1), 53-72. Lobao, L. M. (1990). Locality and inequality: Farm and i ndustry structure and socioeconomic condition. Albany, NY: State University of New York Press. Lobao, L. (2000). Industrialized farming and its relations hip to community well-being: Report prepared for the state of South Dako ta, office of the attorney general. Retrieved from http://www.agribusinessaccountability. org/pdfs/270_Industrialized%20Farming.pdf Lobao, L. M., & Schulman, M. (1991). Farming patterns, rural restructuring, and poverty: A comparative regional analysis. Rural Sociology, 56, 565-602. Lobao-Reif, L. (1987). Farm structure, industry structure, and socioeconomic conditions in the US. Rural Sociology, 52(4), 462-482. Lyson, T. A., Torres, R. J., & Welsh, R. (2001). Scale of agriculture production, civic engagement, and community welfare. Social Forces, 80(1), 311-327. MacCannell, D. (1988). Industria l agriculture and rural commun ity degradation. In L. E. Swanson (Ed), Agriculture and community change in the U.S.: The congressional research reports (pp. 15-75). CO: Westview Press. Maddala, G. S. (1983). Limited-dependent and qualitativ e variables in econometrics. New York: Cambridge University Press. Marousek, G. (1979). Farm size and rural communities: Some economic relationships. Southern Journal of Agricultural Economics, pp. 57-61. Massey, D., & Denton, N. (1993). American apartheid: Segrega tion and the making of the underclass. Cambridge, MA: Harvard University Press. Massey, D., Gross, A. B., & Shibuya, K. (1994). Mi gration, segregation, a nd the concentration of poverty. American Sociologi cal Review, 59, 425-445. Mayhew, B. H., & Levinger, R. L. (1976). Si ze and the density of interaction in human aggregates. American Journal of Sociology, 82, 86-110. Miethe, T. D., Hughes, M., & McDowall, D. (1991). Social change and crime rates: An evaluation of alternative theoretical approaches. Social Forces, 70(1), 165-185. Min, Y. and Agresti, A. (2002). Modeling nonnega tive data with clumping at zero: A survey. Journal of the Iranian Statistical Society, 1, 7-33. 139

PAGE 140

National Agriculture Statistics Service. (2002). Trends in the U.S. agriculture, A walk through the past and a step into the new millennium. U.S. Department of Agriculture. Retrieved from http://www.usda.gov/nass/pubs/trends/ Osgood, D. W. (2000). Poisson-based regressi on analysis of aggregate crime rates. Journal of Quantitative Criminology, 16(1), 21-43 Osgood, D. W., & Chambers, J. M. (2000). Social disorganization outs ide the metropolis: An analysis of rural youth violence. Criminology, 38(1), 81-115. OShea, T. C. (1999). Community policing in sma ll town rural America: A comparison of police officer attitudes in Chicago and Baldwin County, Alabama. Policing and Society, 9(1), 5976. Ott, R. L., & Longnecker, M. (2001). An introduction to statistical methods and data analysis (5th ed.). Pacific Grove, CA: Duxbury. Ousey, G. C. (2000). Deindustrialization, female headed families, and black and white juvenile homicide rates, 1970-1990. Sociological Inquiry, 70, 391-419. Parisi, D., Mclaughlin, D. K., Ta quino, M., Grice, S. M., & White, N. R. (2002). TANF/Welfare client decline and community cont ext in the rural South, 1997-2000. Southern Rural Sociology, 18, 154-187. Park, R. E. & Burgess, E. W. (1925). The city. Chicago: University of Chicago Press. Parker, K. F. (2004). Industrial shift, polarized labor markets and urban violence: modeling the dynamics between the economic transfor mation and disaggregated homicide. Criminology, 42(3), 619-645. Petee, T.A., & Kowalski, G. S. (1993). Modeling rural violent crime rates: A test of social disorganization theory. Sociological Focus, 26(1), 87-89. Peters, D. J. (2002, July). Revisiting the Goldschmidt hypothe sis: The effect of economic structure on socioeconomic conditions in the rural Midwest. Technical paper. Retrieved from http://missourifarmersunion.org/conf03/goldschmidt03.pdf Pindyck, R. S., & Rubinfeld, D. L. (1998). Econometric models and economic forecasts (4th ed.). Boston: McGraw-Hill. Price, M. L., & Clay, D. C. (1980). Structur al disturbances in rural communities: Some repercussions of the migration turnaround in Michigan. Rural Sociology, 45(4), 591-607. Pridemore, W. A. (2005). A cautionary note on using county-level crime and homicide data. Homicide Studies, 9(3), 256-268. 140

PAGE 141

Reimund, D., Stucker, T., & Brooks, N. (1987). Large-scale farms in perspective. U.S. Department of Agriculture, Economic Research Service, Agriculture Information Bulletin 505. Washington, DC: U.S. G overnment Printing Office. Rephann, T. J. (1999). Links between rural development and crime. Regional Science, 78, 365386. Robinson, A. L. (2003). The impact of police soci al capital on officer performance of community policing. Policing, 26(4), 656-689. Rural America at a glance, 2004. (2004, September). (Agriculture Information Bulletin No. 793).Washington, DC: U.S. Department of Ag riculture, Economic Research Service. Rural poverty at a glance. (2004, July). (Rural Development Research Report No. 100). Washington, DC: U.S. Department of Ag riculture, Economic Research Service. Sampson, R. J. (1987). Urban black violence: The effect of male joblessness and family disruption. American Journal of Sociology, 93, 348-382. Sampson, R. J. (1991). Linking the microand macro-level dimensions of community social organization. Social Forces, 70(1), 43-64. Sampson, R. J., & Groves, W. B. (1989). Co mmunity structure and crime: Testing socialdisorganization 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, 918-924. Sayrs, L. W. (1989). Pooled time series analysis: Quantita tive applications in the social sciences. Newbury Park: Sage Publications. Shaw, C. R., & McKay, H. (1942). Juvenile delinquency and urban areas. Chicago: University of Chicago Press. Shihadeh, E. S., & Ousey, G. C. (1998). Industr ial restructuring and violence: the link between entry-level jobs, economic deprivation, and black and white homicide. Social Forces, 77, 185-206. Skees, J. R., & Swanson, L. E. (1988). Farm structur e and rural well-being in the south. In L. E. Swanson (Ed.), Agriculture and community change in the U.S.: The congressional research reports (pp. 238-321). Boulder, CO: Westview Press. Snyder, A. R., & McLaughlin, D. K. (2004). Female-headed families and poverty in rural America. Rural Sociology, 69(1), 127-149. STATA. (2003). Cross-sectional time-series reference manual. STATA Press. 141

PAGE 142

Stofferahn, C. W. (2006). Industrialized farmi ng and its relationship to community well-being: An update of a 2000 report by Linda Lobao. Retrieved from http://www.und.nodak.edu/org/nd rural/Lobao%20&%20Stofferahn.pdf Swanson, L. E. (1982). Farm and trade center transition on an industrial society: Pennsylvania, 1930-1960. Ph.D. Dissertation, The Pennsylvania St ate University, University Park, PA. U.S. Department of Justice, Federal Bureau of Investigation. Uniform Crime Reporting program data [United States]: County level arrest and offense data, 1977-1983 [Computer file]. Washington, DC: U.S. Dept. of Justice, Federa l Bureau of Investigation [producer], 1984. Ann Arbor, MI: Inter-university Consortium for Po litical and Social Research [distributor], 1998. ICPSR #8703 U.S. Department of Justice, Federal Bureau of Investigation. Uniform Crime Reporting program data [United States]: County-level de tailed arrest and offense data, 1991 [Computer file]. ICPSR ed. Ann Arbor, MI: Inter-university Co nsortium for Political and Social Research [producer and distributor], 1994. ICPSR #6036 U.S. Department of Justice, Federal Bureau of Investigation. Uniform Crime Reporting program data [United States]: County-level de tailed arrest and offense data, 1992 [Computer file]. ICPSR ed. Ann Arbor, MI: Inter-university Co nsortium for Political and Social Research [producer and distributor], 1994. ICPSR #6316 U.S. Department of Justice, Federal Bureau of Investigation. Uniform Crime Reporting program data [United States]: County-level de tailed arrest and offense data, 1993 [Computer file]. ICPSR ed. Ann Arbor, MI: Inter-university Co nsortium for Political and Social Research [producer and distributor], 1995. ICPSR #6545 U.S. Department of Justice, Fede ral Bureau of Investigation. Uniform Crime Reporting program data [United States]: County-level de tailed arrest and offense data, 2001 [Computer file]. ICPSR3721v2. Ann Arbor, MI: Inter-univers ity Consortium for Political and Social Research [producer a nd distributor], 2006-01-16. U.S. Department of Justice, Federal Bureau of Investigation. Uniform Crime Reporting program data [United States]: County-level de tailed arrest and offense data, 2002 [Computer file]. ICPSR04009v2. Ann Arbor, MI: Inter-univers ity Consortium for Political and Social Research [producer and distributor], 2006-01-16. U.S. Department of Justice, Federal Bureau of Investigation. Uniform Crime Reporting program data [United States]: County-level de tailed arrest and offense data, 2003 [Computer file]. ICPSR04360-v2. Ann Arbor, MI: Inter-univers ity Consortium for Political and Social Research [producer and distributor], 2006-01-31. van Es, J. C., Chicoine, D. L., & Flotow, M.A. (1988). Agriculture techno logies, farm structure and rural communities in the corn belt: Polic ies and implications for 2000. In L. E. Swanson (Ed.), Agriculture and community change in the U.S.: The congressional research reports (pp. 130-180). Boulder, CO: Westview Press. 142

PAGE 143

143 Websdale, N. (1998). Rural woman battering and the ju stice system An ethnography. Thousand Oaks, CA: Sage Publications. Websdale, N., & Johnson, B. (1998). An ethnostatist ical comparison of the forms and levels of woman battering in urban and rural Kentucky. Criminal Justice Review, 23(2), 161-196. Weisheit, R. A., & Donnermeyer, J. F. (2000). The nature of crime: Continuity and change Change and continuity in crime in rural America. Criminal Justice 2000, 1, 309-357. Weisheit, R. A., & Fuller, J. (2004). From the field: Methamphetamine in the heartland: A review and initial exploration. Journal of Crime and Justice, 27(1), 131-151. Weisheit, R. A., & Wells, L. E. (1996). Rural crime and justice: Implications for theory and research. Crime and Delinquency, 42(3), 379-397. Weisheit, R. A., & Wells, L. E. (1999). Th e future of rural crime in America. Journal of Crime and Justice, 22(1), 1-27. Weisheit, R. A., Wells, L. E., & Falcone, D. N. (1994). Community policing in small town and rural America. Crime and Delinquency, 40(4), 549-567. Welsh, R., & Lyson, T. A. (2001). Anti-corpor ate farming laws, the Goldschmidt hypothesis and rural community welfare. Retrieved from http://www.askfarmerbrown.org/welshlyson.pdf Wheelock, G. C. (1979, August). Farm size, community structure and growth: Specification of a structural equation model. Paper presented at the Rural Sociological Society annual meeting, Burlington, VT. Wilson, W. J. (1987). The truly disadvantages: The inner city, the underclass, and public policy. Chicago, IL: University of Chicago Press. Wilson, W. J. (1996). When work disappears: The world of the new urban poor. New York: Alfred A. Knopf. Wilson, S. M., Howell, F., Wing, S., & Sobsey M. (2002). Environmental injustice and the Mississippi hog industry. Environmental Health Perspectives, 110(2), 195-201. Wimberley, R. C. (1987). Dimensions of US agristructure: 1969-1982. Rural Sociology, 52(4), 445-461.

PAGE 144

BIOGRAPHICAL SKETCH Stephanie Hays is originally from rural Iowa. After graduating high school in 1999, Stephanie left Iowa to pursue a college educatio n. She received a B.A. with distinction in sociology-criminology in 2001 and a Master of Human Relations degree in 2003 from the University of Oklahoma. She completed her M.A. degree in criminology, law and society with a minor in statistics from the University of Florida in 2005 and her PhD in criminology, law and society with a minor in family, youth and commu nity sciences in 2007. After completing her PhD, Dr. Hays returned to Iowa, where she accepted a position as an assistant professor of criminology and criminal justice at Buena Vista University in Storm Lake, Iowa. 144