Inmate misbehavior is a relatively common phenomenon in correctional facilities throughout the United States. While many steps have been taken by correctional authorities to curb such behavior, their level of success is hampered by the lack of information on how social context affects inmate behavior. This research proposes to look at how context interacts with individual characteristics to affect rates of prison infractions in a variety of prison settings. Two major theories of inmate misbehavior, importation and prisonization, are tested and extended using hierarchical linear models on 1997 North Carolina prison population data.

This dissertation reports on interviews with members of five exclusive country clubs in the Northeastern United States. At these clubs, membership is extended only by selective invitation after a subjective screening process. The clubs have long histories of racial-ethnic homogeneity, but they now display some demographic diversity while preserving the economic and cultural homogeneity with which members are comfortable, and which they consider an important appeal of the private club. I focus on club members' explanations around three topics: their clubs' exclusivity, their racial-ethnic composition, and the status of women members. Subjects minimize the significance of the exclusion they perform by rhetorically pointing to forces beyond their control, and by promoting the American Dream of colorblind, meritocratic equal opportunity. While they use the dominant racial ideology of colorblindness, subjects also show a departure from colorblindness in their active development of and rhetorical emphasis on racial-ethnic diversity in their ranks. Concerning women's status, club members mostly accept the subordination of women in clubs. To justify it, they rhetorically rely on both the dominant gender ideology and the inequalities in men's and women's wealth and domestic responsibilities which originate elsewhere. Club members are called to account for their exclusivity by the American value of egalitarian equal access. But at the same time, other cultural values provide them with the tools needed to successfully explain themselves, even as their talk and actions contribute to the reproduction of class, race, and gender inequalities. This research describes the perspective of wealthy white people, and critiques it as inadequate to a full understanding of the consequences of their actions. It shows how country club members talk and act in ways that help preserve their privileges, and the reasons why they do so.

There are a number of reasons to expect that incarceration will have long-term, negative consequences for economic/labor market success, and that the consequences may be especially acute for minority ex-offenders. This study replicates and extends Bruce Western?s research on the impact of incarceration for wage mobility. I integrate Western?s life course approach to examining the impact of incarceration with a discussion of stratification processes that produce inequality in employment and earnings outcomes. I hypothesize that incarceration results in career earnings penalties over and above those associated with foregone human capital accumulation. I suspect that incarceration contributes to a decline in earnings for minority ex-offenders. At the individual level, I replicate Western?s research by estimating fixed-effects models to examine wages across the career trajectories of white, Latino and African American men from the National Longitudinal Survey of Youth for 1979-1998. When estimating these models, I test whether human capital accumulation that occurs inside or outside the labor market mediates the incarceration-earnings relationship. Furthermore, I examine how local labor market characteristics influence ex-offender career trajectories. I propose that prison records, race/ethnicity and spatial characteristics such as, violent crime rates, unemployment rates, minority concentration, and residential segregation influence the job prospects of workers within metropolitan areas. At the spatial level, I estimate random effects models to examine how local labor market characteristics shape the earnings trajectories of white, Latino and African American male ex-offenders. The individual level results supported the hypotheses that incarceration has a negative effect on earnings and that ex-offenders have lower earnings trajectories than non-offenders. This study did not replicate Western?s finding that the earnings penalty experienced by those who had been incarcerated varies by race/ethnicity. The spatial analysis results suggest that the prison effect on wages is not influenced by the spatial characteristics associated with the local labor market. However, the results indicate that the spatial characteristics of the labor market influence race/ethnicity wage disparities across the career. This study makes a contribution to the existing literature on the consequences of incarceration by linking attributes of ex-offenders, emergent career dynamics and local labor market prospects within a stratification framework.

The purpose of the current research is to examine infraction behavior in North Carolina Prisons by creating a typology of infraction of behavior and modeling the occurrence of the specific types of infractions, seeking evidence for and against possible infraction specialization. There are three specific phases of analysis. First, exploratory factor analysis is conducted to create a typology of infractions. Second, a Markov model is created to examine the extent of infraction specialization. Finally, hierarchical linear modeling logistic regression is used to both discover how three theories ? prisonization, importation and social control ? predict the occurrence of the different infractions types identified in the first phase of the analysis and as well as to help partially validate the infraction typology presented. Hierarchical linear modeling logistic regression results indicate that there are different predictors for the different infraction types, supporting the need to distinguish between them. Further, the Markov Model used to examine change of behavior over time supports limited specialization in infraction careers and shows that infraction behavior is not random. Overall, there is moderate support for the proposed typology and for the presence of specialization in infraction careers.

Data from 1983-1986 and 1988 waves of the National Longitudinal Study of Youth (NLSY) are drawn upon to explore how type of childcare utilized affects the extent of mother?s employment. Fixed-effects analysis suggests using relative care decreases mother?s hours of employment per week. Once age of the youngest child is introduced, nuclear care decreases mother?s hours of employment per week. Surprisingly, mothers of infants and toddlers are predicted to work slightly more hours per week than mothers of preschoolers. Husband?s hours of employment per week interacts with non-family care to decrease mother?s hours of employment. The effect of the number of children in the household on mother?s hours of employment is found to depend on the childcare utilized. Interestingly, mother?s years of education and husband?s annual income do not have statistically significant effects on mother?s hours of employment in this analysis. Some possible reasons and implications of these findings are discussed.

Data from both waves of the National Survey of Families and Households were used to examine differences between three relationship types in the division of household labor. Respondents were assigned to three groups: continuously married to the same partner (n = 5,035); continuously cohabiting with the same partner (n = 125); and transitional, from cohabiting to married, with the same partner (n = 216; total sample size = 5,376). The proportional contributions to hours spent on feminine (cooking, laundry, cleaning), masculine (outdoor tasks, auto repair), and neutral (running errands, paying bills) household tasks by male and female partners were examined for all relationship types. T-tests and OLS regression were used to determine differences between groups, and factors which impact proportional contributions at each wave of the survey. In addition, changes in contributions to each task type by male and female partners over time were examined in the context of group membership and other factors. Women in all groups contribute more to the feminine task hours than their male partners. When no controls are included, transitional males contribute significantly more than do continuously married men to the feminine task hours at both waves, and transitional women contribute significantly less than do continuously married women at the first wave. Transitional males decrease their contributions to feminine task hours between waves, and continuously married males increase their contributions to feminine task hours between waves. Transitional women increase their contributions to feminine task hours and continuously married women decrease theirs between waves. When controls for age, gender, gender ideology, number of children in the household, income, and education are included, there is no longer a significant difference between transitional and continuously married males? contributions to feminine task hours at the second wave. Changes in proportional contributions to feminine task hours differ significantly between transitional and continuously married men, and between transitional and continuously married women. Also significantly impacting these changes, although the effects are small in magnitude, are gender, change in gender ideology between waves, and education. These results suggest that the shift from cohabitation to marriage does carry with it normative, gendered expectations about the allocation of housework. However, continuously married men increase, and continuously married women decrease, their contributions to feminine household tasks?a finding which is unexpected and suggests the possibility of separate mechanisms which operate in the context and over the course of marriage. A lack of consistent effects for comparisons involving the continuously cohabiting group indicates heterogeneity within this group, and a need for future research to more clearly delineate types of long-term cohabitors.

Between 1996 and 2005, Uniform Crime Reports show an increase in girlsâ arrests for assaults. Previous research has suggested two possible reasons for this increase. Some suggest that girls are becoming more violent, while others argue that increases in girlsâ arrests for assaults may be an artifact of changes in parentsâ and police response to girlsâ conflicts in the home. To test these possibilities, this study uses data for 445 agencies (N=128,787) which continuously reported to the National Incident Based Reporting System (NIBRS) during the years 1996-2005 and information on state domestic violence arrest laws. Bivariate analyses, which assessed the patterns of female violence across years, failed to find evidence of increasing female violence. Through logistic regression analyses, support is found for increasing social control practices despite the fact that after controlling for injury, weapon use, and other characteristics of the incident, state arrest policies appear to equally increase the odds of arrest for boys and girls. Further, an unexpected finding is that African American girls were more likely to be arrested than white girls until arrest laws were introduced into the model indicating that the effect of race was mediated by domestic violence laws. Directions for future research are discussed.

The purpose of this research has been to improve the understanding of the social factors influencing educational attainment. Specifically, this research looks at the combined affects of race, region, and type of residence on educational attainment. Past research has shown that African Americans have consistently had very low levels of educational attainment. Other bodies of educational research have shown that residents of the South have had persistently lower levels of educational attainment than any other major region of the country. Furthermore, similar research has revealed that residents of rural areas also tend to have lower educational attainment than other residential areas. As it turns out, the highest concentration of African Americans and the highest concentration of rural Americans, reside in the South. It is this intersection, as it relates to educational attainment, that is the focus of this research. Data was obtained from the 2000 General Social Survey (GSS) and is analyzed using ordinary-least-squares regression.

The purpose of this research is to examine how dissatisfaction with skin color, as a component of perceived physical attractiveness, is associated with adolescent depression and self-esteem, controlling for skin color, beliefs toward skin color, gender, socioeconomic status, perceived discrimination, and peer and teacher stressors. Using a diverse sample of 388 middle-school students, I employ ordinary least squares regression to test hypotheses about the gender-specific and skin color specific effects of dissatisfaction with skin color on adolescent depression and self-esteem. The results show that dissatisfaction with skin color is positively related to depression and negatively related to self-esteem among adolescents. However, black or brown-skinned African American youth who are dissatisfied with their skin color experience higher levels of depression in comparison to white youth who are dissatisfied with their skin color. Implications of the findings are discussed.

This paper examines when jobs are autonomous in the labor process and under what kinds of circumstances workers achieve more control of their work process. The degree of skill, technology, and bureaucratization are addressed in previous literatures and are analyzed in this paper as sources of variation in work autonomy. The data employed for this study is the 2002 Australian National Organizations Survey, and the research target is core jobs, defined as jobs directly related to the primary product or service of the organizations. Ordinal Logistic regression is employed for this study and result shows that information based technology increase job autonomy for jobs that require higher education. Also, formalized jobs are likely to have less job autonomy. More generally, findings suggest that job autonomy is contingent on relative power in the labor process and that formalization is primarily a control device at least relative to the labor process.