For the period 1981 to 1993, women represented between 21 percent and 26 percent of homicide victims in the United States annually. Although a considerable amount of research has been devoted to determining structural correlates of aggregate homicide rates across places in the United States, recent research shows that general correlates may not hold when homicide rates are disaggregated for specific subpopulations, for example, women. Using multiple regression techniques, this macro-level study analyzes 1990 data from 233 U.S. central cities to discern the utility of factors derived from traditional theories of homicide for explaining female rates of homicide. A model of female homicide victimization that includes factors derived from feminist theoretical arguments related to gender inequality and its supposed relationship to violence against women is also estimated A third model estimates the effects of gender inequality upon female homicide victimization rates when controlling for traditional correlates of homicide. The three models are then reestimated with the residuals of female homicide victimization rates against male rates to determine the utility of measures of social disorganization and gender equality when controlling for the influence of male rates. Results support traditional theoretical arguments for the influence of general social-structural factors of communities upon female homicide risk. The results do not, however, show support for the hypothesized relationship between measures of gender equality and female homicide victimization. Future research implications from this study are suggested / acase@tulane.edu

The research findings concerning the lifestyles/routine activities theory are mixed with contradictory results. The current study uses two National Crime Surveys conducted in 1975 and 1983 in the United States to examine the empirical adequacy of the modification role by lifestyle/routine activities between the demographic differences and the likelihood of criminal victimization. Both personal and household victimization are considered A modified path analysis model with latent variables is employed to improve the measurement of lifestyle/routine activities and test modification role of lifestyle/routine activities. A latent variable termed risk is found to account for the association among the lifestyle/routine activities measures and is strongly related to the likelihood of being victimized Previous studies of household property victimization have generally failed to take into consideration the characteristics of members of a household other than its head. Also, by relying only on household heads to report crime incidents, criminal victimization have been substantially underestimated. Therefore, this study makes use of the characteristics of other household members and includes their reported victimizations as well Another problem with previous research addressed here is the problematic method used to test the mediation hypothesis. Here, the hypothesis tested is broadened to consider other possible modification roles of lifestyle/routine activities. Under the modification hypothesis, the original association between demographic and structural variables with victimization may be modified or altered with the introduction of lifestyle/routine activities measures The findings support the modification role played by lifestyle/routine activities in both violent and property victimization for the two surveys. With the introduction of lifestyle/routine activities, the association between demographic effects and victimization is modified in ways that are largely consistent with the lifestyle/routine activities theory / acase@tulane.edu

Despite recent reports that juvenile crime has leveled off, the rates and types of offenses committed by juveniles continue to be of concern In particular, rates of aggressive offenses perpetrated by juveniles continue to be astounding. Researchers have a long history of examining juvenile delinquency in an attempt to determine which factors might put children at risk. Typical factors examined include ethnicity, family variables, history of substance abuse, intellectual abilities, and academic achievement. Delinquency has been measured in numerous ways including self-reports to determine the number and type of delinquent acts committed and official records of delinquency, such as legal history. Unfortunately, most studies have analyzed their data using delinquents as a homogeneous group, without examining possible differences in risk factors based on the level or seriousness of the acts involved. The current study examined samples of aggressive and non-aggressive juvenile offenders in an effort to develop a prediction model of offending. Developmental level and gender were among the variables considered. No single variable was found to predict aggression reliably within the total sample of offenders. A combination of four individual factors (substance-related diagnosis, age at first offense, drugs sold, and impulse control disorders) explained a significant amount of variance, however, they were found to have poor predictive value. A total risk variable was compiled by simply adding together the total number of known risk factors for each participant. This procedure resulted in correct prediction of aggression for one-third of the available sample. Subsamples of offenders were also examined. Implications of the findings and suggestions for future research are discussed / acase@tulane.edu

The research presented here provides an empirical test of six models of rape. These models are: (1) the social disorganization model, (2) a model which hypothesizes a southern subculture of violence, (3) a model which hypothesizes a racial subculture of violence, (4) a sexual inequality model, (5) a cultural model of rape, and (6) a model which hypothesizes a link between pornography and rape Results support the social disorganization model with over half of the variance in rape rates among Standard Metropolitan Statistical Areas (SMSAs) explained. The models suggesting a southern subculture of violence and a racial subculture of violence both failed to achieve empirical support. The sexual-inequality model received only minimal support; occupational sex-segregation emerged as the only significant predictor of rape rates. The cultural model also received minimal support. Finally, the hypothesis which suggests a relationship between pornography and violence against women was not supported. The variables retained in the final model explained over sixty-two percent of the variance in rape rates / acase@tulane.edu

In recent decades, there has been a resurgence of interest in ecological explanations of crime, especially as they relate to Shaw and McKay's (1942) research regarding community structure, crime and delinquency. Social disorganization, relative deprivation and subcultural deviance theories identify several variables measuring formal and informal social controls, inequalities, and learning opportunities that mediate the effects of the community structure and crime relationship. Robert Agnew (1999) poses a macro-level strain theory (MST), which suggests that strain is conditioned by social control and learning variables that influence systemic levels of negative affect and community crime rates. Preliminary tests of MST (Brezina, Piquero, and Mazerolle 2001; Hoffman and Ireland 2004; Pratt and Godsey 2003) provide partial support for the theory; yet, studies have been limited in abilities to operationalize variables and to model indirect effects. This dissertation tests MST using data collected from the Project on Human Development in Chicago Neighborhoods: Community Survey, 1994-1995; 1990 Census Data; and Homicide Incident Data from the Chicago Police Department for the years 1996-1999. Structural equation models measuring the mediating effects of strain on the relationship between community characteristics and homicide rates fit better than models of social disorganization and subcultural deviance. While variables measuring social control do condition the effects of strain on negative affect and crime, these models have poor fit. Thus, while the mediating effects of MST are supported, more research is needed on the moderating effects of social control and learning variables / acase@tulane.edu

State level data from fifty states for 1983, a period of severe economic recession and 1987, a period of relative economic recovery, are incorporated into separate multivariate analyses to determine if variations in the economy and other social structural factors, controlling for crime, exert significant influence on the application of social controls in 1983 and 1987. Particular attention is given to the relevance of labor market marginality, race, alternate controls and fiscal constraints for the various forms of punishment. Previous empirical research on formal social control has primarily focused on dependent variables which represent only a fraction of those under control--prison population and/or prison admissions. This study expands the study of social control beyond prison population and admissions by incorporating separate analyses of jail, probation, parole, and juvenile imprisonment as dependent variables. This research provides evidence to support the thesis that social structure exerts direct influence on various forms of punishment, controlling for crime effects. / Source: Dissertation Abstracts International, Volume: 53-03, Section: A, page: 0953. / Major Professor: Theodore Chiricos. / Thesis (Ph.D.)--The Florida State University, 1992.

A content analysis of crime news, appearing in Time magazine in selected years during the post-World War II period, explores the ideological nature of reports about crime within a random sample of articles. Examination of the content of articles about crime, criminals and criminal justice, appearing in Time at different junctures during the post-war period, makes possible an analysis of popular ideologies of crime within the context of changes and developments in the political economy of the United States. / The types of crime, characteristics of offenders, causal factors, and commands of what to do about crime within the articles are described. In addition, comparisons are made, within each of these areas, between articles appearing in years with low levels of unemployment and years with high unemployment rates. Comparisons are also made between crime news appearing in the immediate post-war period of economic expansion and reports about crime appearing during the more recent period of stagnation and decline, beginning at the onset of the 1970s. / Defining ideology as discourse which gives a distorted picture of contradictions, misrepresenting the basic contradictions of capitalism in ways which support the interests of the dominant economic class, this research provides evidence that the portrayal of crime within Time magazine during the post-World War II period is ideological. / Source: Dissertation Abstracts International, Volume: 52-06, Section: A, page: 2278. / Major Professor: Theodore G. Chiricos. / Thesis (Ph.D.)--The Florida State University, 1991.

The relationship between perceived control beliefs and treatment outcome with chronic adolescent offenders, who exhibit internalizing or externalizing behavior problems was explored. Perceived control was addressed by assessing perceived contingency and perceived competence separately. / Data from the records of 307 students committed to a state training school for hardcore delinquents were examined. Students were males between the ages of 14 to 18. / The results of this study support the importance of considering domains of perceived control. The efficacy of using two types of measures of perceived control for predicting treatment outcome is supported. The use of two measures of perceived control for different types of disorders is not strongly supported, at least with chronic adolescent offenders. / This study has implications for treatment and research. Two important points for treatment include: (1) working through denial of behavior problems is essential to treatment gain; (2) focusing on the view that one is responsible for successes and for resolving their problems appears important for treatment gain. Additional research is indicated to further clarify the efficacy of making the distinction between perceived contingency and perceived competence beliefs for internalizing and externalizing behavior problems. Research is also indicated to determine which therapies are effective in changing perceived control beliefs with chronic adolescent offenders. / Source: Dissertation Abstracts International, Volume: 52-08, Section: A, page: 2868. / Major Professor: Wallace A. Kennedy. / Thesis (Ph.D.)--The Florida State University, 1991.

Criminal justice expenditures in forty-eight American states for the 1948 to 1985 period are examined through the lens of three alternative models: public choice, insurgency and economic structuralism. Chapter one discusses the framework through which self-titled "government growth" studies address government expenditure increases and provides alternative perspectives for understanding government growth and contraction. Chapter two provides the theoretical groundwork for the three alternative models of criminal justice expenditures. Chapters three and four describe, justify and operationalize respective model determinants of criminal justice expenditures. Chapters five through seven test the relative strength of each determinant for three component series of criminal justice expenditures: police, corrections and judicial expenditures, respectively. / Finally, chapter eight reviews the most interesting findings of earlier chapters and discusses the significance of the dissertation for future government growth research and its relevance to teaching and research in public administration. / Source: Dissertation Abstracts International, Volume: 52-08, Section: A, page: 3067. / Major Professor: Barton Wechsler. / Thesis (Ph.D.)--The Florida State University, 1991.

Georg Rusche observed in the 1930s that the political economy contributed significantly to the development of penal systems in various countries. This research project explores Rusche's contention that the political economy significantly shapes the formation of criminal justice institutions and their operation. Specifically, this study examines federal criminal justice legislation between 1948 and 1987 in relation to the long cycles of world capitalist development, the economic conditions of capital, the economic conditions of labor, and the index crime rates. The findings support the hypothesis that a significant relationship exists between the political economy and criminal justice policy. A multivariate time-series analysis reveals that the economic conditions of capital and the long cycles remain significantly correlated with federal criminal justice legislation when controlling for the effects of the crime rate. The findings do not support the theory that the criminal justice system develops a blatantly more coercive or "hard" approach to the crime problem during the period of economic contraction. The implications of these findings for criminal justice theory development are discussed. / Source: Dissertation Abstracts International, Volume: 52-05, Section: A, page: 1903. / Major Professor: Theodore G. Chiricos. / Thesis (Ph.D.)--The Florida State University, 1991.