Judicial decision-making has been a long-standing subject of criminological inquiry. It has been the explicit focus of theory (e.g. Steffensmeier, Ulmer, & Kramer, 1998; Albonetti, 1991; Farrell & Holmes, 1991) and is implicit in discussions of unwarranted sentencing disparity, determinate sentencing, and extra-legal offender attributes such as race. Central to each of these topics is the judge's sentencing decision and the differences in sentences that flow from the use of discretion. However, few studies have actually directly examined variation in judges' sentencing behavior and how this variation corresponds to judge and offender attributes. This dissertation fills that void by using data from Florida Circuit Courts to examine how judges' indelible attributes and experiences influence their use of imprisonment. Results demonstrate that judges are far from homogenous in their sentencing behavior and, in spite of sentencing guidelines, extra-legal offender attributes continue to matter but in nuanced ways. Findings include statistically significant effects for the judges' political party affiliation, age, time on bench, sex, race, and Hispanic ethnicity; several of these effects are conditioned by offender attributes. These effects are modest in magnitude but when considered cumulatively, they result in consequential differences in the probability that an offender is imprisoned. Judges also show considerable variation in their propensity to imprison even after controlling for differences in their traits. While judge attributes like race and sex matter, they do not adequately capture the bulk of inter-judge variation in the use of imprisonment. In short, criminal sentencing remains a highly individualized activity. / A Dissertation submitted to the College of Criminology and Criminal Justice in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester, 2012. / October 17, 2012. / decision-making, extralegal attributes, judicial attributes, logistic regression, multi-level models, sentencing / Includes bibliographical references. / Theodore Chiricos, Professor Directing Dissertation; Marc Gertz, Committee Member; William Bales, Committee Member.

The empirical relationship between inmates with and without tattoos upon post-release recidivism has been virtually ignored in modern criminological research. Only one published study has directly examined the relationship between inmate tattoos and recidivism (Putnins, 2002). This study tracked 898 released Australian juvenile offenders for a brief period and found support for a link between tattoos and violent recidivism. The purpose of the current study is to provide a rigorous empirical assessment of the consequence of inmate tattoos on the likelihood of recidivism among a large cohort of offenders released from prison. The study examines a cohort of 79,749 released inmates from Florida prisons from 1995 through 2001 and tracks them over a three year follow-up to assess the impact of several tattoo variables on recidivism. Findings reveal that released inmates with tattoos, particularly numerous and highly visible ones, are more likely to be reconvicted during the follow-up period. Further, the findings indicate that there are two distinct inmate profiles namely the younger novice to the corrections system and the older, longer-term prisoner that are distinguishable by the numbers of tattoo possessed that increase the odds of recidivism. The implications of the findings are discussed in terms of policy, theory, and future research. / A Dissertation submitted to the College of Criminology and Criminal Justice in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester, 2012. / April 2, 2012. / inmate, prisoner, recidivism, tattoo / Includes bibliographical references. / Thomas G. Blomberg, Professor Directing Dissertation; Joyce L. Carbonell, University Representative; William D. Bales, Committee Member.

Age is one of the most robust correlates of prison misconduct, with younger inmates more likely to commit disciplinary infractions. There has been, however, little attention to the potential nonlinear effect of age on prison misconduct. Most research to date has assumed that the relationship is linear and modeled it as such. In so doing, prior work has been unable to identify the potential nonlinear relationship between age and prison misconduct. This dissertation thus attempts to examine the precise nature of the age-misconduct relationship. Furthermore, it addresses the call for more research that investigates how prison experiences may influence prisoner reentry outcomes. Therefore, the purpose of this dissertation is to contribute to efforts to advance scholarship and better understand the relationship between age, prison behavior, and recidivism. First, it takes a nuanced look at the potential nonlinear relationship between age and prison misconduct. The approach taken here expands on prior research by examining granular age categories to predict several types of disciplinary infractions. Next, it investigates possible interaction effects of gender and, separately, race/ethnicity on the age-misconduct relationship. Finally, it explores the relationship between age, prison misconduct, and recidivism. Specifically, it examines the relationship between age and types of recidivism and whether various types of misconduct mediate this relationship. This dissertation draws on a Florida Department of Corrections admission cohort from 1995 to 2000 and a release cohort from 1995-2002. Negative binomial regression models were used to examine the relationship between age and prison misconduct and binary logistic regression models were used to examine the relationship between age, prison misconduct, and recidivism. The findings of this study shed light on the significance of identifying and understanding the nonlinear relationship between age and prison misconduct, and, further, the mediating effect of misconduct on the age-recidivism relationship. Theory, research, and policy implications of the findings are discussed and suggestions are made regarding future research. / A Dissertation submitted to the College of Criminology and Criminal Justice in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester, 2012. / June 19, 2012. / Age, Corrections, Prison Misconduct, Recidivism / Includes bibliographical references. / Daniel Mears, Professor Directing Dissertation; Melissa Radey, University Representative; William Bales, Committee Member; Sonja Siennick, Committee Member.

The effects of delinquency and criminal behaviors during early adolescence on events over the life course have been well-established in the criminological research. A segment of this research has revealed that the apparent causal relationship between delinquency and later life course events may be due to a third exogenous confounding variable, namely: genetics. While biosocial research has illuminated the need to include recognition of the proportional influence of genetic factors and environmental factors the research lacks an overarching theoretical framework that allows precision in research and guidance for future research. An evolutionary approach may represent such a framework. Employing data from a large national sample of sibling pairs, this project seeks to assess this assertion by analyzing the influence of antisocial behaviors during adolescence on sexual and reproductive strategies over the life course. Three key findings emerged from the analyses. First, the majority of items tapping antisocial behavior and delinquency in adolescence, as well as measures of sexual, reproductive, and relationship behaviors in adulthood were shown to be influenced primarily by genetic and nonshared environmental factors. Second, multivariate analyses revealed that antisocial conduct during adolescence has an effect on sexual/reproductive outcomes in adulthood. However, when genetically sensitive methodologies are employed the association is significantly diminished. Third, even after controlling for the influence of shared genetic factors and shared environmental factors some forms of antisocial behavior in adolescence had an effect on sexual/reproductive behaviors across the life course. The findings are discussed within the context of life history theory and evolutionary psychology. / A Dissertation submitted to the College of Criminology and Criminal Justice in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester, 2013. / June 14, 2013. / EVOLUTIONARY CRIMINOLOGY, EVOLUTIONARY PSYCHOLOGY, LIFE COURSE CRIMINOLOGY, LIFE HISTORY THEORY, REPRODUCTIVE BEHAVIOR, SEXUAL BEHAVIOR / Includes bibliographical references. / Kevin M. Beaver, Professor Directing Dissertation; Lisa A. Eckel, University Representative; Eric P. Baumer, Committee Member; Eric A. Stewart, Committee Member.

Dr. Matthew Robinson (2004) proposed a new inclusive theory of human criminal behavior, entitled Integrated Systems Theory. This theory put forth by Robinson (2004) and later Robinson and Beaver (2009), hypothesizes that human behavior is far too complex for any one theory or any one behavioral discipline to accurately predict and examine; rather the study of human behavior should include factors from multiple theories and disciplines. Integrated Systems Theory hypothesizes that human behavior is influenced by multiple factors across the life course, both biological and environmental. These proposed factors are exhaustive and include six different levels of influence, the cellular level, the organ level, the organism level, the group level, the community level, and the societal level. These factors include, but are not limited to, deviant peers, brain development, toxin exposure, genetic functioning, neighborhood characteristics, family structure, routine activities, and gender and racial inequality across the societal level. Integrated System Theory further expounds upon its hypotheses by positing that while these various factors all influence behavioral development, one key relationship also plays an important role, the interplay between genes and environment. It hypothesizes that all human behaviors are the result of interactions between genes and the environments that an individual is exposed to across the life course (Robinson, 2004; Robinson & Beaver, 2009). Currently, due to the exhaustive nature of this theory no empirical study has yet sought to test its various tenants. Therefore, this dissertation seeks to examine facets of this theory by focusing on the role played by both biological and social factors in the prediction of antisocial phenotypes. Focusing on the risks of prefrontal cortex development, cumulative genetic risk, deviant peers, maternal risk, socioeconomic status, school attachment, and neighborhood disadvantage, the influence of these contributors upon human behavior will be examined. Further, gene-environment interaction terms will be developed between cumulative genetic risk and the chosen environmental risks, to determine if these five different gene-environment interaction terms are strong contributors to antisocial phenotypes. To determine the relationship between these environmental and genetic factors in influencing human behavior, a range of antisocial outcomes will be predicted within negative binomial models. Outcomes include drug abuse, violent crime, property crime, and arrest rates. This dissertation seeks to determine if tenants of the Integrated Systems Theory are supported or refuted by the data and analyses. Results generated within the analyses indicated partial support for Integrated Systems Theory. Within the models brain development played a significant role in behavioral development. The importance of direct environmental effects were also generally supported within the data; deviant peer and school attachment emerged as some of the more robust environment contributors to antisocial phenotypes. Direct genetic effect failed to display consistent predictive power across the various negative outcomes. The importance of gene-environment interplay was generally supported within the models as well, especially when interacting with deviant peers and in the prediction of drug abuse and arrests rates. The meaning of the results, potential limitations, and contribution to the literature are discussed. / A Dissertation submitted to the College of Criminology and Criminal Justice in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester, 2013. / June 17, 2013. / Biosocial, Criminology, Gene-Environment Interplay / Includes bibliographical references. / Kevin M. Beaver, Professor Directing Dissertation; Anne Barrett, University Representative; Eric Stewart, Committee Member; Patricia Warren Hightower, Committee Member.

Social threat theory argues that the dominant group can be threatened in a variety of ways by minorities. Illegal immigrants are an example of one such group that has been described as posing a threat. Illegal immigrants have been identified by some as criminally, culturally, and economically problematic for native-born Americans. However, prior research on social threat has rarely examined perceptions regarding illegal immigrants. This research uses data from a telephone survey of a random sample of adults in the United States (N=1534) to test several of the propositions of social threat theory as they relate to illegal immigrants. Specifically, this study tests how contextual threat influences both perceptions of threat and punitive attitudes as well as how perceived threat affects support for punitive controls. Not only is this study one of the first to focus specifically on illegal immigrants, but it also examines multiple links in the theory's causal chain, making it perhaps the most thorough test to date. Overall, findings are supportive of social threat theory. Individuals who live in more threatening contexts, as indicated by a dynamic measure of the exposure of non-Latinos to Latinos, perceive more criminal, cultural, and economic threat from illegal immigrants. These individuals are also more likely to express support for both border and internal controls. Additionally, all types of perceived threat considered here are significant predictors of support for the punitive controls. Additional analyses suggest the effect of perceived threat may vary by subsample. Most notably, there is evidence of a ceiling effect for conservative respondents. / A Dissertation submitted to the College of Criminology and Criminal Justice in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester, 2013. / June 5, 2013. / Immigration, Public opinion, Social control, Social threat / Includes bibliographical references. / Ted Chiricos, Professor Directing Dissertation; Kathryn Tillman, University Representative; Gary Kleck, Committee Member; Brian Stults, Committee Member.

Prior neighborhood crime research has sought to explain differences in crime between black and white communities using such criminological theories as social disorganization and concentrated disadvantage. While supportive in their findings, these perspectives did not explain the entirety of neighborhood crime differences. The current study attempts to add to this literature by integrating concepts from the larger urban sociological tradition from which criminology grew out of, such as spatial assimilation and place stratification. Research in this tradition has found that African Americans have not been able to translate higher levels of income into more desirable communities, including those with low crime rates, due to a dual housing market steering them away from the best residential areas in a city. By including city level measures that could either eliminate or dampen the efforts of these channeling processes, the hierarchical linear models used found support for variations in place stratification across urban and suburban areas. However, convergence in crime levels for white and black communities was dependent on the social class of the community and crime type. / A Dissertation submitted to the College of Criminology and Criminal Justice in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester, 2014. / April 4, 2014. / Crime, Neighborhoods, Place Stratification, Race, Spatial Assimilation / Includes bibliographical references. / Brian J. Stults, Professor Directing Dissertation; Rebecca Miles, University Representative; Eric P. Baumer, Committee Member; Eric A. Stewart, Committee Member.

There is evidence in the literature that regulation may be a more efficient means of harm reduction in prostitution than criminalization. However, prostitution is illegal everywhere in the United States except for eight counties in Nevada where regulated brothels are licensed. Studies indicate these brothels are effective at controlling the violence, sexually transmitted diseases, and community disorder typically associated with prostitution. What remains unknown is whether they deter demand for illegal prostitution, which remains plagued by these harms. Comparisons of prostitution arrest rates among counties with and without brothels in Nevada, prostitution arrest rates among all U.S. states, and drug arrest rates among counties in Nevada were completed. While existing data are largely insufficient for evaluating the viability of Nevada's legal brothels as model alternatives to criminalization, it is clear that the brothels, which are located in Nevada's more rural areas, do not adequately address illegal prostitution in urban areas. Debate about the appropriateness of legalization will continue. Nevertheless, some consensus options, such as increasing access to resources for women trying to escape prostitution and emphasizing demand-side controls over supply-side controls, have emerged. / A Thesis Submitted to the College of Criminology and Criminal Justice in Partial Fulfillment of the Requirements for the Degree of Master of Science. / Fall Semester, 2010. / October 20, 2010. / Nevada, brothels, prostitutes, prostitution, legalization, regulation, harm reduction, violence, women, sex, disease, community disorder / Includes bibliographical references. / Daniel Mears, Professor Directing Thesis; Sonja Siennick, Committee Member; William Bales, Committee Member.

The impact of victim SP on the outcomes in a given criminal events has rarely been examined or fully understood. This study develops a new theory of victimization, which I will refer to as the Power Advantage Theory (PAT), that holds crime victimization is completed only when there are motivated offenders who have contact to victims and possess physical or psychological power advantage over the victims. Based on PAT, this study assesses the impact of 16 types of victim self protection (SP) actions on three types of outcomes of criminal incidents: whether the incident resulted in property loss, whether it resulted in injury to the victim, and whether it resulted in serious injury. Data on 27,595 personal contact crime incidents recorded in the National Crime Victimization Survey for 1992-2001 are used to estimate multivariate models of crime outcomes with logistic regression. Results indicate that self-protection in general reduces the likelihood of property loss and injury, compared to nonresistance. A variety of mostly forceful tactics, including resistance with a gun, appear to have the strongest effects reducing the risk of injury, though some of the findings were unstable due to the small numbers of sample cases of such resistance. The appearance, in past research, of resistance contributing to injury is found to be largely attributable to confusion concerning the sequence of SP actions and injury. In crimes where both occurred, injury followed SP in only 10 percent of the incidents. Combined with the fact that injuries following resistance are almost always relatively minor, victim resistance appears to be generally a wise course of action. In two auxiliary test, it was found that victims used forceful self-protection, especially weapon use and defensive gun use, in the most adverse circumstances and that victims' perceptions of the efficacy of SP were much more favorable than those implied by rates of actual post-SP injury although two measures were highly significantly correlated. These findings imply that actual effects of SP may be stronger than they appear to be in the previous study. Taken together, the results of three empirical tests generally support the hypotheses of PAT. / A Dissertation Submitted to the College of Criminology and Criminal Justice in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Summer Semester, 2005. / June 24, 2005. / Power Advantage Theory, Victimology, Self-Defense, Self-Protection, Victim Resistance / Includes bibliographical references. / Gary Kleck, Professor Directing Dissertation; James Orcutt, Outside Committee Member; Spencer Li, Committee Member.

The deskilling of labor and the loss of social networking has accounted for part of the crime problem in recent decades. Vocational training and other correctional programming is meant to reverse the effects of these problems by providing inmates with usable, legal, and employable skills. Over the last five to ten years, the role of rehabilitative services has become highly questionable. The lack of confidence in such services has led to a decline in the number and type of prison programming available to inmates. In addition, fiscal concerns on the part of privately owned correctional industries have sharpened this decline. This paper reports results of an evaluation performed on PRIDE Enterprises, one of the largest providers of privately operated inmate vocational training in Florida, in regards to its effect on inmates' post release employment and recidivism. This study found that working in the vocational program, PRIDE, significantly increases the likelihood of the participant being employed after release. This study also found that PRIDE has no direct effect on recidivism. / A Thesis Submitted to the College of Criminology and Criminal Justice in Partial Fulfillment of the Requirements for the Degree of Master of Science. / Summer Semester, 2005. / July 6, 2005. / Vocational Training, Life Course, Social Capital, Recidivism / Includes bibliographical references. / Gordon Waldo, Professor Directing Thesis; William Bales, Committee Member; Marc Gertz, Committee Member.