Search Results

The key findings in this dissertation include: (1) selection adjustments partially or fully confounded the effects of gang joining; despite this (2) gang joiners had 70 percent the odds of earning a high school diploma and 42 percent the odds of earning a 4-year college degree than matched individuals who avoided gangs; (3) at the 11-year mark, the effect of gang joining on educational attainment exceeded one-half year; (4) gang joiners made up for proximate deficits in high school graduation and college matriculation, but gaps in 4-year college degree and overall educational attainment gained throughout the study; (5) gang joiners were less likely to be employed and more likely to not participate in the labor force, and these differences accelerated toward the end of the study; (6) gang joiners spent an additional one-third of a year jobless relative to their matched counterparts; and (7) the cumulative effect of gang joining on annual income exceeded $14,000, which was explained by the patterning of joblessness rather than the quality of jobs. The theoretical and policy implications of these findings, as well as directions for future research, are addressed in the concluding chapter of this dissertation.

Second, I investigate how educational assortative mating shapes husbands and wives income trajectories over the course of marriage. I use multilevel dyad models to analyze data from the National Longitudinal Survey of Youth 1979 (NLSY79). Educational assortative mating is captured by three types of educational pairings of spouses: educational hypergamy in which the wife is less educated than the husband, educational homogamy in which both spouses have same levels of education, and educational hypogamy in which the wife is more educated than the husband. I find that change in husbands income with marital duration was similar regardless of educational pairings of spouses, whereas change in wives income varied by educational pairings of spouses such that wives in educational hypogamy exhibited more positive change in income over the marital life course. The findings suggest that it remains important for husbands to bring income into the family no matter what educational levels they have relative to their wives, whereas the rise in women's education and in prevalence of educational hypogamy likely protects women from earning less after marriage.

Lastly, I examine how educational assortative mating shapes patterns of female breadwinning status over the course of marriage. I use group-based trajectory models to analyze data from the NLSY79. I find substantial movement in and out of the primary breadwinner role by wives across marital years and great heterogeneity in trajectories of female breadwinning status across couples. In addition, educational assortative mating plays a role in shaping patterns of female breadwinning status: educationally hypogamous couples are less likely than educationally homogamous or hypergamous couples to follow the traditional trajectory characterized by virtually no chance of achieving a female breadwinning arrangement over the first twenty years of marriage.

The final chapter includes the effect of parental conflict into skill development during childhood. Parental conflict is a non-tangible input related with psychological well-being of the parents. There is evidence that conflict adversely affects cognitive and non-cognitive skill development, but this is the first study that jointly analyzes the impact on both skills. Estimates suggest that reductions in conflict benefits skills and adult outcomes. Cognitive development is more affected during early childhood and non-cognitive development for later ages. The effect of reducing parental conflict on years of education completed is similar to the effect of increasing parental time but lower than the effect of increasing material investments.

In the first two studies, we explored the policy dimension of an individual's environment, specifically how the state-level policy environment shapes frequency of alcohol use. We used data from the National Longitudinal Survey of Youth 1997 (NLSY97) cohort, a nationally representative, longitudinal sample of emerging adults.

The first essay examines whether one's wage is based on information about the performance of one's personal contacts. I study wage determination under two assumptions about belief formation: individual learning, under which employers observe only one's own characteristics, and social learning, under which employers also observe those of one's personal contacts. Using data on siblings in the NLSY79, I test whether a sibling's characteristics are priced into one's wage. If learning is social, then an older sibling's test score should typically have a larger adjusted impact on a younger sibling's log wage than vice versa. The empirical findings support this prediction. Furthermore, I perform several exercises to rule out other potential factors, such as asymmetric skill formation, human capital transfers, and role model effects.

The second essay analyzes the influence of macroeconomic conditions during childhood on the labor market performance of adults. Based on Census data, I document the relationship of unemployment rates in childhood to schooling, employment, and income as an adult. In addition, a sample from the PSID is used to study how the background attributes of parents raising children vary over the business cycle. Finally, information from the NLSY79-CH is examined in order to characterize the impact of economic fluctuations on parental caregiving. Overall, the evidence is consistent with a negative effect of the average unemployment rate in childhood on parental investments in children and the stock of human capital in adulthood.

The third essay studies the bilateral trade of divisible goods in the presence of stochastic transaction costs. The first-best solution requires each agent to transfer all of her good to the other agent when the transaction cost reaches a certain threshold value. However, in the absence of court-enforceable contracts, such a poli cy is not in centive compatible. We solve for the unique maximal symmetric subgame-perfect equilibrium, in which agents can realize some gains from trade by transferring their goods sequentially. Several comparative statics are derived. In some cases, the first-best outcome can be approximated as the agents become infinitely patient.

The fourth chapter uses longitudinal survey data to measure the effect of student loan interest rates on college matriculation. Relying on exogenous variation in real student loan interest rates across high school graduation cohorts to identify this effect, I find no statistically significant effect of student loan interest rates on the decision to matriculate immediately to a four-year institution.

The second paper uses propensity score matching to estimate the impact that living in a multigenerational household (including the child’s mother and at least one maternal grandparent) has on test scores for children of the NLSY79. If the addition of a grandparent adds more resources (for example: income and care) to the household than s/he uses, child test scores will increase. If grandparents use more resources than they add, household resources will be diluted and child test scores will decrease. After using propensity score matching and child-level fixed effects, I find evidence that children living in these households score lower on tests than those who do not.

I control for the nonrandom decision of household type using a propensity score matching. I use kernel weighting to match children from multigenerational households with similar non-multigenerational households and fixed effects to control for unobserved heterogeneity. For robustness, I use cross sectional data to estimate the importance of matching method and time invariant independent variables that cannot be included in a fixed effects model. Even after propensity score matching, I find evidence that children living in these households score lower on tests than those who do not.

Using a representative sample of U.S. couples, we found that the correlation in parental wealth between spouses is 0.42. This parental wealth assortative mating accounts for about one-quarter of the conventional intergeneration wealth correlation, and at least seven percent of the intergenerational correlation in income. These results confirm Becker's (1973, 1974) notion that assortative mating might magnify measured parent/child correlations. experience more unemployment spells, marital disruptions, and negative health shocks than white children.

Lastly, for some PSID individuals aged 27 to 35 in 1996, experiencing family head unemployment at the young age (6-12) directly lowered a child's risk tolerance measures, while experiencing family unemployment around age 13 to 18 had little effect. A child is 10% less likely to be in the highest or second highest risk tolerance category, out of four categories given in the PSID, if experienced family head unemployment when young. However, experiencing family une mployment while young does not seem to directly affect a person's choice of occupations. The result suggests that family unemployment at younger age may serve as an instrument for risk tolerance under some circumstances since risk preference is an important, but usually under-reported variable for research on household decisions.

In my second essay, "Does Fetal Exposure to Tobacco Smoke Affect a Child's Adult Outcomes?," I examine whether fetal exposure to tobacco smoke affects long-run outcomes such as schooling attainment, employment, and labor market earnings.

In my third essay, "A New Look at the Effect of Credit Constraints on College Attendance," I examine whether credit constraints affect college attendance.

In the second study using data from the Maryland Adolescent Development in Context Study (MADICS), results show that initial ethnic differences in parental control and their trajectories over time are accounted for by variations in neighbor problems and their subsequent effect on parents' prevention-focus. Although levels of prevention-focus are related to parents' neighborhood context, African Americans are significantly more concerned with preventing negative outcomes for their adolescents than European American parents.

The third study assesses how parents' socialization goals mediate ethnic difference in parental control, independent of socioeconomic factors. Parents' filial-cultural piety and success goals completely mediate ethnic differences in parental strictness but not the amount of autonomy they grant their adolescents. The final study tests a psycho-social conceptual model of how parenting behaviors influence adolescents' school performance and if these processes differ by a function of ethnicity. The results show that a warm and stimulating home environment is related to mastery goal orientations and study habits that foster higher academic achievement, whereas adolescents who perceive greater restrictiveness and demandingness have performance goals and study habits related to worse school performance. The se patterns are consistent for European American and African American adolescents. Overall, these studies find that cultural factors, more so than social factors, explain variations in parenting behaviors, which have important implications for youth outcomes.

Net worth was divided into two variables, a dichotomous variable indicating those who had a negative net worth (defined as a net worth less than or equal to zero) and a continuous variable of the actual dollars of net worth of those who had a positive net worth (defined as a net worth greater than zero). Developing a separate negative net worth variable allowed this study to expand on prior research that focused solely on that aspect of net worth (Chen & Finke, 1996; Mountain & Hanna, 2012). Data was taken from the National Longitudinal Survey of Youth 1979 (NLSY79), using results primarily from the 2014 survey. A logistic regression was used to evaluate the negative net worth variable (Model 1) while an ordinary least squares (OLS) regression was used to analyze the influence on positive net worth (Model 2). This study found that present-fatalism increased the odds that an individual would have a negative net worth, while a future-orientation would decrease those odds. It found that present hedonism and future-orientation contributed to having a positive net worth.

In the second essay, I consider how U.S. families choose to invest in response to the onset of a health condition in a child. Family investments can reinforce, or compensate for the occurrence of a health-limiting condition. The results from this paper shed light on the importance of incorporating the family unit as part of public policies that involve children with serious health conditions.

With respect to their employment ambitions and perhaps prospects, the average military enlistee is likely to differ from the average American. In Chapter 2, we estimate the impact military service has on civilian wages across the wage distribution. For early 21st century veterans, we find that former military service grants civilian wage premiums at and below the median wage level but perhaps penalties at the high end of the wage distribution. For late 20th century veterans, who were mostly peace-time volunteers, we find evidence that veteran wage premiums were more constant across the wage distribution.

Military service adds additional challenges for married couples. In Chapter 3, I perform a trajectory analysis of the effect of military service on the likelihood of divorce. I find that these individuals were most likely to get a divorce in the first year following active duty service, with an increased probability of three to six percentage points. A within-racial group analysis shows that these effects are stronger for whites than minorities. I find that veterans who served during an earlier period (1980-1992) were unaffected, implying differing effects for wartime versus peacetime service.

The second chapter, The Effect of Housing Wealth on Labor Market Outcomes and Behavior , expands on the findings of Chapter 1 and previous literature that found a link between housing wealth and education by examining the impact of home wealth on adult labor market outcomes and behavior. Using the NLSY97, I measure the effect of a change in home price while in high school on income at age 26, job industry, criminal behavior, and community participation. I find evidence that home wealth shocks increase the probability of holding white collar jobs, increase wages, and positively impact behavior outcomes. These effects are likely manifesting through the increased education found in previous work.

In the second chapter, these pathways are used to predict young adult income and examine how experiences in this time period serve as mechanisms in the intergenerational transfer of status. The transition-to-adulthood pathways explain about one-third of the intergenerational transmission of status through young adulthood. For men, the pathway characterized by extended education mediates the relationship between parental wealth and young adult income, indicating a potential mechanism for resource hoarding in the upper-middle class. For women, on the other hand, this life stage may contribute to the poverty trap, as pathways related to less advantaged populations link family background to young adult income.

The third chapter contextualizes the transition to adulthood and examines how the impact of class on transition-to-adulthood pathways varies across institutional contexts in 20th century Europe. Class operates fairly consistently in Northern, Western, and Southern Europe, where young women with more educated parents tend to delay family formation and avoid rapid transitions to adulthood. The role of class in the transition to adulthood is distinct in Eastern Europe, however. Results are explained according to the welfare states and family systems operating in each region.

Together, the chapters illustrate that an individual's location within society influences how they are likely to experience the transition to adulthood, which has meaningful consequences for long-term outcomes. The influence of class on these pathways, however, may vary by context.

The findings unveil a nuanced picture of the connections between marriage and motherhood postponement, the development of childlessness intentions, and permanent childlessness. Preference for marriage delay and remaining childless emerge early in the life course, at times as early as during adolescence. Thus, the processes of selection into marriage delay and into preferences for childlessness start very early and set young women on trajectories of motherhood postponement which ultimately increase their chances of permanent childlessness. In addition, the prevalence of intentions to remain childless is on the rise, potentially leading to a spread of a childfree lifestyle among future generations. By addressing the associations between motherhood delay and childlessness intentions, this dissertation contributes to scholarly conversations about parenthood delay, childlessness, and family change.

The second chapter investigates the Impact of physical appearance on the transition from high school to full-time employment Due to changes in the structure of the economy since the 1980s, the average time it takes a job-seeker with only a high school diploma to gain full-time employment has been increasing. Several reasons have been proffered for the low transition from high school graduation to full-time employment, but these reasons have been shifting over time. Against this background, I propose an additional element, physical attribute, that may explain the low transition from high school graduation to full-time employment. Using both parametric and semi-parametric hazard models, I show that physical appearance affects the odds of transitioning from high school to full-time employment. I find that job-seekers who have only a high school diploma and who are well below average in physical height, spend on average, five more months unemployed compared to others. There is no significant reduction in the odds of exiting unemployment for taller job-seekers, suggesting a shortness penalty, but not a height premium. Additionally, I find the likelihood of obese job-seekers transitioning from high school to full-time employment is about 28% lower than that of non-obese job-seekers. With the economy predominantly service oriented, these results imply job-seekers with less than desirable physical attributes will face challenging labor market conditions for the foreseeable future.

The third and the final chapter examines the impact of parental education on the educational attainment of second generation immigrants (SGI). Leveraging the rich parental characteristics available in the NLSY, I contrast the effect of parental years of schooling on the years of schooling and degree attainment of natives and second generation immigrants. I find a positive correlation between parental schooling and the educational attainment for both natives and SGI. The impact is more pronounced for natives than for the average SGI. The results seem to indicate that some immigrant parents invest more in the education of their children to help them achieve socio-economic mobility. These also robust to the inclusion of several parental socio-economic characteristics and controls for the ability of the children.

The first essay examined the association between information search behaviors and retirement planning actions upon two dimensions of self-esteem, consisting of efficacy and worth. Both information search behaviors and retirement planning actions were found to be associated with both dimensions. Attained levels of education and the masculine gender were also found to be significantly associated with each self-esteem dimension.

Essay two explored creditworthiness as part of the identity self-verification (Stryker, 1980) and self-esteem buffer mechanism, and its association with pre-retirement planning behaviors. Higher levels of self-esteem, attained level of education, net worth, and net income were all found to be associated with individuals who were likely to engage in one or more pre-retirement planning behavior. Creditworthy practices, however, were not found to be associated with pre-retirement planning behaviors in this study.

Essay three postulated that respondents who possessed a composite psychosocial profile consisting of Rosenberg's self-esteem scale (Rosenberg, 1965), Pearlin's mastery scale (Pearlin & Schooler, 1978), and Rotter's locus of control scale (Rotter, 1966) would be associated with engaging in one or more retirement planning behavior. Results showed that a composite psychosocial profile is associated with individuals likely to engage in one or more retirement planning behaviors. Attained levels of education, net worth, net income, and age were found to be associated with individuals likely to engage in one or more retirement planning behaviors.

Using both duration analysis and panel data methods, the results indicate that drinking is positively related to improved socioeconomic outcomes as compared to total abstention, when endogeneity has not been taken into account. In contrast, under the duration analysis, estimation via instrumental variables approach indicates that alcohol consumption shortens employment duration. Panel data estimation indicates that the relationship between alcohol consumption and socioeconomic outcomes is rather an inverted U-shaped for some specifications, when endogen eity has been taken into account. Additionally, the effects of drinking on retirement wealth and earnings tend to diminish with the instrumental variables approach. The findings were unchanged even with abstainers partitioned into lifetime abstainers and infrequent or light drinkers (less than one drinking day per week). The results also confirm the positive association between human capital measures such as the level of education and economic outcomes and also the negative relationship between alcohol consumption and taxes on alcoholic beverages.

This dissertation contributes to the literature on alcohol-socioeconomic outcomes nexus and has implications for policies related to health, social capital and alcohol since a more inclusive alcohol and/or health policy could improve civic responsibility and narrow the health capital and social capital gap, both of which are critical to individual level socioeconomic success.

In families that end up divorcing or separating, the spouses invest less in their children during marriage. In addition, investment in these children declines starting about two years prior to the divorce itself, primarily due to lower investments by fathers. This new finding suggests that the anticipation of divorce may be detrimental to children through its impact on parental investment. In order to understand these results, in the first chapter of this dissertation I develop a model of a married couple learning about their match quality and engaging in match-specific investment in their children. The model shows that it is match quality, and not merely divorce, that affects children. I use two implications of the model to establish that the prospect of divorce impacts home investment in children during marriage. First, by using variation over time in the probability of divorce within a given marriage, I show that the probability of divorce is negatively related to parental investment in children during marriage. Second, I use information on marital conflict to show that the home investment levels of families near the margin of divorce respond to differences in the costs and benefits of divorce whereas in families far from the margin of divorce they do not. This finding implies that it is the possibility of divorce that is driving some of the differences in investment level s between families. My results establish a new pathway by which divorce may harm children: the anticipation of divorce decreases investment in children during marriage. This effect can explain a small but significant portion of the gap in investment levels between families that divorce and those that do not; roughly the same amount as is explained by the mother's cognitive ability or father's education. A standard collective model of the household with a household public good shows how a change in bargaining power within the household can result in a change in the demand for child quality. The marriage market sex ratio is an important driver of bargaining power in the household.

In the second chapter of this dissertation I establish that the marriage market sex ratio, by affecting the balance of power in the household, affects the allocation of household resources to children. The problem with credibly estimating this relationship is the potential endogeneity of the sex ratio. To account for this endogeneity, I use variation induced by two sources: (i) the "Baby Boom" in the United States, and (ii) immigration flows to the United States in the early part of the 20th century. Using data from the National Longitudinal Survey of Youth 1979 - Children Sample (NLSY79-CS) I find that mothers who faced higher sex ratios (i.e. a high relative supply of men) due to the state and year of their birth have higher household investment in children. In these families, women also have lower labor supply. In addition, using data from the National Longitudinal Survey - Mature Women cohort, the General Social Survey, and the 1980 U.S. Census, I show that the daughters of second generation immigrants to the U.S. who faced higher marriage market sex ratios have more years of completed education.

In the first part of the essay, data on young men from the National Education Longitudinal Survey is utilized to determine the relation of education and labor market outcomes to childhood behavior. The main finding is that eighth grade behavior is of the same order of importance to earnings as eighth grade test scores. Moreover, childhood behavior is associated with earnings at all educational levels. By contrast, achievement test scores are only related to earnings for young men with postsecondary degrees. Furthermore, behavior is persistent. Although school and family characteristics do relate to behavior, these factors explain only a small part of behavior.

In the second part, the relationships of motivation to test scores are investigated. Plenty of inferences regarding test scores and economic outcomes are drawn from tests administered to surveys participants, in which no performance-based incentives are supplied. Thus, the issue of effort, or motivation, might be crucial to the interpretation of the empirical findings. To illustrate the problem the black-white test score gap is examined. Given that blacks score lower than whites on standardized tests, a common conclusion drawn is that blacks are less capable than whites. If test scores are produced by two inputs, human capital and effort, where effort is costly, it is possible that the root of the test score gap is lack of effort, and not lack of knowledge. This lack of effort on the part of minorities might be the result of either actual or perceived lack of future opportunities. Using the National Longitudinal Survey of Youth supporting indirect evidence are provided. To directly investigate the effect of incentives on the black-white test score gap, an experiment was conducted. The experimental results, however, are inconclusive regarding the roots of the black-white test score gap.

I focus on two aspects of remarriage formation to understand how remarriage patterns compare to first marriage formation. First, I analyze the individual characteristics associated with the likelihood of remarriage for men and women. I pay particular attention to socioeconomic status (income, labor force status, and educational attainment), first marriage ties (co-residential children) and time since divorce as important factors in the remarriage market. Second, I examine educational and age assortative mating patterns in second marriages to identify whether remarriage follows a more contemporary homogamous pattern or a more traditional form where men marry less-educated and younger women. I use data from the National Longitudinal Survey of Youth, 1979 cohort (NLSY79) to answer these questions.

The results show remarriage formation patterns consistent with traditional first marriages where current and long-term economic status has a positive effect on remarriage for men, but not for women. For women, family background, race/ethnicity, age and parental status are associated with remarriage. The analysis of assortative mating shows that high-status men tend to marry less-educated, young women and both educational and age homogamy is unlikely for both men and women, regardless of socioeconomic status. These patterns are different from those present in contemporary first marriages where both men and women emphasize on spousal economic status as valuable traits. These findings suggest that remarriage differs from first marriage in its likelihood and assortative mating patterns and provide a foundation for future work to incorporate gender dynamics in research on remarriage.

Chapter two uses an audit study to determine the existence and extent of caste-based discrimination in the Indian private sector. The study also has policy implications for recent debates regarding introduction of caste-based quotas in Indian private sector jobs. Resumes with caste-specific names are sent to employers for entry-level jobs and callback rates measured. On average, low-caste applicants need to send 20% more resumes than high-caste applicants to get one callback. There is also heterogeneity in callback gaps by recruiter characteristics and firm size which indicates the presence of prejudice against low-caste workers and is consistent with commitments made by large firms to hire actively from among low-caste groups.

In chapter three I find partially identified treatment effect for arrest and other treatments by looking at recidivism for a sample of domestic assault offenders. The treatment effects are not fully identified due to non-compliance with assigned treatment and the possibility of a non-random treatment assignment. Partially identified treatment effects are estimated by making minimal assumptions on the counterfactual probabilities.

Chapter four (based on joint work with Wallace Mok) examines the difference in non-wage compensation between African Americans and whites in the US. Using data from the Current Population Survey (CPS) and National Longitudinal Survey of Youth (NLSY), we find that without controlling for the Armed Forces Qualification Test (AFQT) scores, white men are more likely to receive non-wage compensation and white women are not more likely to get non-wage compensation. With controls for AFQT scores we find that white men are not more likely to receive non-wage compensation but black women are more likely to get non-wage compensation. We also find that the percentage differences in total compensation and the percentage differences in wages across racial groups are essentially the same.

Individuals in my model make different career and family decisions due to observed characteristics that affect consumption and nonconsumption utility of choices. Choices differ also due to randomness in realizations of wages, translation of educational inputs into attainment, and choice-specific time shocks to utility. The human capital accrued through postsecondary education is modeled such that actual attainment is the measure of interest. The main education choice facing a woman in this model is how much effort to put towards completing additional schooling if she chooses to enroll. While a woman cannot directly choose to complete an additional year of school, her chances of doing so increase with the amount of effort she puts forth.

Using maximum likelihood estimation, I estimate a structural model of women's choices using data from the National Longitudinal Survey of Youth in years prior to welfare reform. The results from the structural model are qualitatively consistent with the relationships found in the non-structural portion of this research but suggest an important role for unobserved heterogeneity.

Specific plans exist and are under way to use maximum simulated likelihood estimation to estimate an empirical model that includes unobserved heterogeneity in women's tastes and ability. I will use estimated parameters from that model to simulate the effect of welfare reform on the collegiate attainment of low-income women.

Some literature suggests that using theory-driven, multilevel frameworks to address future expectations among young adults may be a promising approach. Future expectations, or the extent to which one expects an event to actually occur, have been shown to influence goal setting and planning and thus may guide behavior. Future expectations have been linked to several psychosocial outcomes, but the literature examining its associations with behavior is limited by small, homogenous samples and cross-sectional designs. Furthermore, its measurement tends to focus on single dimensions and may be missing important components of this construct. This dissertation, therefore, seeks to improve our understanding of future expectations and its relationship with adolescent risk behavior using data from the National Longitudinal Survey of Youth 1997 (NLSY97). Specifically, this research aims to 1) identify subclasses of future expectations defined b y clustering of unique combinations of expectations related to education, work, family, and delinquency, to test the associations between these subclasses and risk factors derived from an ecological model, and to determine how these classes relate to risk behaviors (delinquency, substance use, and sexual experience); 2) prospectively examine the relationship between future expectations and sexual risk behavior (number of sexual partners, inconsistent contraception use, and adolescent parenthood); and 3) determine how parental future expectations influence three risk behaviors (delinquency, substance use, and sexual risk) and school attainment, and to determine if youth future expectations mediate the proposed relationship between parental expectations and behavior. These analyses used interview data collected annually from 1997 through 2007 among approximately 3,000 youth ages 15 and older. Statistical techniques included latent class analysis, latent growth modeling, and various regression models.

Results of this dissertation support the empirical and practical importance of future expectations in understanding adolescent risk behavior. In my first chapter, latent class analysis supported the emergence of four distinct classes of future expectations. These classes were labeled the Student Expectations, Student/Delinquent Expectations, Victim Expectations, and Work/Delinquent Expectations classes according to their indicator profiles. Classes differed with respect to the sociodemographic characteristics associated with membership. Each class was also statistically associated with at least one adolescent risk behavior. In my second chapter, the prospective relationship between future expectations and sexual risk behavior was explored. Results indicated that these classes were uniquely associated with age at first biological child, number of sexual partners, and inconsistent contraception use. The Work/Delinquent Expectations class was consistently associate d with the greatest sexual risk among all outcomes when compared to the Student Expectations class. Membership in the Student/Delinquent Expectations class was associated with increased number of sexual partners and inconsistent contraception use, but not age at first biological child. The Victim Expectations class was not associated with any outcome when compared to the Student Expectations class. Lastly, the mechanism of parental influence was explored in my third chapter. Results suggest that parental expectations were strongly associated with adolescent behavior at baseline and over time; however, different parental expectations emerged as important for different behaviors and times. Furthermore, youth expectations partially mediated this relationship.

Analyses were carried out using survey data from the Luxemburg Income Study (LIS), the National Longitudinal Surveys (NLS, NLSY79, NLSY97), and the Multipurpose ISTAT (FSS 2003). A first main finding of this study is that the transition to economic independence has been delayed together with all the other events of the transition to adulthood. This process has occurred in all developed Western countries even if with some differences. A second finding is that parental social class can explain some of the variation in life courses, and that a higher social class is associated with a postponement in the transition. Also the role of family background, however, differentiates based on welfare state regimes, institutions, and the strength of family ties.

Essay 2 picks up where essay 1 leaves off by investigating college outcomes for individuals who enrolled during a recession. While many studies consider the enrollment decisions, little evidence exists on whether enrollment is successfully transformed into completed education for recession era enrollees. Employing an innovative competing risk model, I estimate the completion and drop-out probabilities for individuals who enrolled during a downturn. I find that individuals who enrolled in college at 18 and who experience a recession at enrollment, are less likely to complete a 4-year degree by age 24, are more likely to complete a 2-year degree, are more likely to drop out of college and are more likely to experience inactivity.

Essay 3 builds upon the negative effects of a recession on college-aged youths found in essay 2. In essay 3, I study educational attainment after individuals have exited their college-aged years. I investigate whether cohorts who experienced adverse economic conditions during young adulthood eventually caught up with their luckier counterparts who experienced more prosperous years. I find that individuals who experience adverse economic conditions during parts of the college-aged years (18-21) experience lower educational attainment than those who experience more prosperous college-aged years and these negative effects are still present up to ten years post college-age.

The goal of this dissertation is to explore how children influence their own environments by evoking caregiver or parent behavior. It examines whether child language and cognitive ability shape caregiver language stimulation and the home learning environment, and whether child academic, behavioral, and health characteristics influence parental use of housing vouchers.

For the first study, I use longitudinal data from the National Institute of Child Health and Development (NICHD) Study of Early Child Care to examine whether children with more advanced cognitive and language development evoke more stimulating language environments. I find evidence for an evocative response for toddlers (15 and 24 months), but not for preschoolers (54 months). The evocative response does not vary by child care context.

For the second study, I use longitudinal data from the Children of the National Longitudinal Study of Youth to examine whether children with more advanced language and cognitive skills evoke higher quality home learning environments, and whether these evocative effects are moderated by other child and family characteristics. Using multiple analytic techniques, I find evidence that more advanced skills do evoke higher quality home learning environments, and that these evocative effects function fairly similarly for children of different ages, gender, and socioeconomic background.

For the third study, I use data from the Moving to Opportunity (MTO) study to examine how child academic, behavioral, and health characteristics influence parental moving behavior. I find that, in parti cular, families with children who have multiple problems are much less likely to move than families whose children do not have problems.

By examining evocative processes across these three areas, I identify evocative processes that shape children's development contexts and discuss their implications for research, policy, and practice.

Chapter One develops a household allocation framework that analyzes how a child's health interacts with other important decisions, such as labor force participation, distribution of marital surplus, and divorce decisions.

Chapter Two develops an empirical model that accounts for possible interaction effects in determining the decision to divorce; specifically, how the decision of both parents deciding to work full-time in the market, type of divorce legislation they are subject to, and the health of any children interact with each other and affect the decision to divorce. The main finding of this paper shows that there are not just direct effects but significant interaction effects.

Chapter Three develops an empirical model that examines the role that birth order plays on the development of cognitive abilities in the presence of inputs provided by the mother and the father.

In terms of determinants of self-sufficiency, the results of logistics regression analyses suggest that having a higher paying job and working more hours will move low-income families into an economically better situation, and that an increase in the hourly wage and hours worked will improve self-sufficiency for low-income families. In addition, the respondent's intelligence level and educational attainment have positive relationships with self-sufficiency and changes in it; being single or with more children makes it more difficult economically compared to being married or having fewer children after controlling for other factors.

Based on Chow's F-test, there was a significant change in the behaviors of low-income families during welfare reform, from 1996 to 1998. The average hourly wage is a more important factor than hours worked for attaining higher self-sufficiency over time from 1996 to 2000. Educational attainment and the number of children have important different effects on self-sufficiency and changes in it between 1996 and 1998, while the importance of the respondent's intelligence level and job experience remain relatively the same.

The second essay investigates the effects of family size on child health. This essay is a joint study with Kabir Dasgupta. In this study, we use matched mother-child data from the National Longitudinal Surveys to study the effects of family size on child health. Focusing on excess body weight indicators as children's health outcome of interest, we examine the effects of exogenous variations in family size generated by twin births and parental preference for mixed sex composition of their children. We find no significant empirical support in favor of the quantity-quality trade-off theory in instrumental variable regression analysis. This result is further substantiated when we make use of the panel aspects of the data to study child health outcomes of arrival of younger siblings at later parities.

The third essay estimates the causal effect of being born out of wedlock on a child's health outcome and early academic achievements. Specifically, the study uses rich panel data from the National Longitudinal Survey of Youth 1979 (NLSY79) and the Children of the NLSY79 (NLSY79-child), coupled with a sibling fixed-effects model to address omitted variable bias attributable to unobserved family characteristics.

Chapter 2, the primary focus of this essay is to obtain new estimates of the price sensitivity of cigarette consumption and related outcomes using the 1979 cohort of the National Longitudinal Survey of Youth (NLSY79). This work expands on the existing literature on the topic by taking into account measures of time preference in the NLSY79 to examine interactions between these measures and price in the demand function for cigarettes. My specific hypothesis is that those who discount the future consequences of their current actions heavily are likely to be more sensitive to price than those who do not. I find that the people who discount the future heavily are more sensitive to price change.

Chapter 3, in this essay I explore the smoking behavior of pregnant women using the 1979 cohort of the National Longitudinal Survey of Youth (NLSY79). A key aspect of this research is the availability of smoking participation data before and during pregnancy. Thus, I consider the probabilities of quitting while pregnant as outcomes. I find that pregnant women who are cigarette consumers are less responsive to price changes because they are a future oriented group. Individuals who are more present-oriented are more likely to smoke and to consume more cigarettes given that they do smoke than those who are more future-oriented. Moreover, those who discount the future more heavily will be more sensitive to the money price of cigarettes than those who are more future-oriented. I find that a one percent change in the money price of cigarettes represents a larger percentage change in the full price for the former group. I focus on the role of time preference and the interaction between time preference and price in determining these outcomes.

Data from the National Longitudinal Survey of Youth, 1979 are used to estimate a series of multinomial logit approximations determining the odds of entering a specific relationship type. Results show that biological children have a positive effect for men and women on forming marital and cohabiting unions. However, this effect is greater for men than women, especially when biological children are coresident. Coresidential fathers are significantly more likely to form a marriage, especially if the union being formed is his first, and are less likely to enter into stepparenthood. Coresidential mothers have an increased likelihood of union formation, but there is no difference in union type formed or entering into stepparenthood. Having only nonresident children increases the likelihood of union formation for both men and women, but these results are weaker compared to those for coresident children.

There continues to be controversy in the literature regarding the importance of children in marriage and family formation. Some emphasize the potential economic returns of marriage, via specialization, as a leading determinant of union formation choices. Others argue demographic changes have altered the ways individuals select romantic partners. This dissertation finds support fo r the latter. Children no longer deter union formation for parents, and often enhance it, especially among fathers. Overall, men's and women's union formation strategies differ largely because of children, and if the cohort studied proves representative of subsequent cohorts, it is likely that such complex partnering and parenting situations will continue into the future.

The third chapter uses the weekly work history data from the NLSY's 1979 cohort to analyze the effect of leaving high school during a recession. These data allow me to precisely measure labor market outcomes and the school-to-work transition. I document severe but short-lived effects of leaving school in a recession on wages, job quality, and the transition time from school to work for men with 9 to 12 years of education. In contrast to published evidence on more educated workers, I find large effects on work hours on both the extensive and intensive margins. When workers leave high school in a recession, they work fewer total weeks and more part-time weeks in their first year in the labor market. They also take substantially longer to find a job, have less access to on-the-job training, and report lower promotion possibilities. Effects of the entry unemployment rate on wages are also large. A 4-point rise in the initial unemployment rate leads to a 21% decline in year-one average wage, a 32% fall in hours worked in the first year, and a 54% decline in first-year earnings. However, the effects of economic conditions are not persistent; by year four, there is no effect on wages, hours, or earnings.

The second essay explores how nonemployment spells and career expectations affect wages. Wages are affected by total nonemployment time, by recent work interruptions, and by some past interruptions. Interruptions affect women's wages further into the future compared to men, but the wage loss associated with any given interruption is less severe for women. One potential reason for the gender differences is that men are more likely to take time off from working for reasons that are negatively related to their productivity. Future career interruptions, which workers presumably anticipate in many cases, affect current investment in human capital to some degree for both sexes. A very small fraction of the gender wage gap is attributable solely to timing of experience.

The third essay examines the current viability of the basic predictions of the Mills-Muth monocentric model of city structure. One previous study uses a cross-section of cities to test the comparative statics predictions, namely that city area is increasing in population and income but decreasing in agricultural land value and commuting costs. While it finds support for the predictions, the data used are from 1970, and there has been a growing consensus that the monocentric model is no longer useful. Despite the increasing polycentricity of cities, there is evidence that the Mills-Muth comparative statics predictions hold for modern cities. Also, densely populated cities are more likely to have subcenters.

Longitudinal data shows that the likelihood of a child smoking more than triples if the child has an older sibling who also smoked. In my first essay, ``Endogenous Sibling Effects in Adolescent Substance Use'', I unpack this striking correlation, asking whether it is causal or a result of correlated unobservables such as parental investments that are endogenous to the child's behavior. In particular, I examine whether smoking or drinking by an older sibling influences the corresponding behavior of a younger sibling. To control for endogeneity in the older child's outcome, I use differences in smoking and drinking by gender and age among older siblings as instrumental variables. While previous studies have established gender differences in teen substance use, I find that these differences vary across age. For example, gender differences in drinking are small at younger ages, but males drink consistently more than females from mid to late adolescence. The instruments are plausibly exogenous of unobservables such as parental substance use, and will control for parental investments under the assumption that the older sibling's age and gender do not directly affect resources invested in the younger child. I empirically investigate this assumption using data on measures of parental investments, and find no evidence of a correlation between the instruments and younger sibling investments. The results point toward significant and positive sibling effects for smoking as well as drinking. These findings indicate the presence of opportunities for resource constrained parents to invest efficiently in favor of their firstborn, to reduce ``bad'' behavior. Positive sibling effects imply that curtailing the older child's behavior in this way can have spillover effects on the younger sibling, leading to greater payoffs to parents in overall child quality. Within-family social multipliers may also serve to amplify the effects of public policies aimed at curtailing smoking and drinking.

For almost two decades, a vast literature has concerned itself with the association between family structure and child outcomes. These studies have typically found that individuals who grow up in traditional families (with two biological parents) are better off across several indicators such as educational attainment, health and fertility as compared to their counterparts from other types of family structures like single mother or blended families (with a biological mother but one in which the father may be step for all, or step for some and biological for others). In my second essay, ``Family Structure, Parental Investments and Child Well-Being'', I adopt a parental investment perspective and propose that differences in parental investments across varied family structures may provide one explanation for differences in child outcomes by family structure. There are several interesting findings. First, children in single mother and blended families receive lower investments as compared to traditional families, with the gap in investments being larger between single mother and traditional families. Second, among single mothers, the group driving lower investments appears to be never married single mothers. Third, I find that joint biological children of both parents in blended families do not differ significantly in the amount of investments they receive relative to their counterparts in traditional families, but non-biological children of fathers are significantly disadvantaged in investment levels. This provides indirect evidence for a biological preference motive. Investigating this further solely within blended families, I find direct evidence in favor of a biological preference motive wherein biological children receive higher investments than their half siblings.

Parental investments in children are usually motivated by models of intra-household allocation which suggest that parents, in making investment decisions, have information about their children's endowments and respond to them. One factor is that parents motivated by efficiency concerns invest more heavily in better endowed children, presumably due to greater marginal returns from the investment for them. An alternative factor is that rather than reinforcing endowment differences, equity concerns may motivate parents to make compensatory investments in their children by investing relatively more in the less endowed children. There have been several empirical studies that have tried to test these implications, but they have almost entirely assumed that all families will reinforce (compensate) child endowments to the same extent. However, we might expect to see heterogeneity in the degree of unequal treatment. In my last essay, ``Family Structure and Intra-Household Resource Allocation'', I explore one such source of heterogeneity namely, family structure. For example, single mothers, experiencing greater resource constraints might have more incentives to make efficient investments in their children as compared to traditional families with two biological parents present. Studying how investment allocation varies across family structure can be viewed as an important link to understand why individuals from varied types of family structure have such different outcomes on a variety of indicators.

The main dataset for this study was constructed from the National Longitudinal Survey of Youth (NLSY97). Using the confidential NLSY97 Geocode File, the NLSY97 was appended with county- and state-specific data from various publically available sources indicating structural disadvantage, population composition, political conservatism, prosecutor's office characteristics, delinquency petition and crime rates, gender inequity, child health and well-being, and juvenile justice policy punitiveness. To take advantage of the longitudinal nature of the NLSY97 data, a combination of multilevel modeling techniques, event history analysis, and generalized linear modeling was employed to examine the effects of individual characteristics and contextual conditions on youths' risk of arrest and probabilities of charge, a court appearance, conviction, and placement. The findings suggest that the effects of gender and racial/ethnic group on youth penetration into the justice system are more pronounced at some decision-making levels and depend on contextual environment.

The results of the analyses by race, gender, and ethnicity suggest three major findings. First, racial disparities are present in youth risk of arrest, which are magnified in predominately non-Black communities. However, this study also found evidence of a compensatory effect whereby Black youth receive more favorable court dispositions than their non-Black counterparts. Second, the gender gap in youth justice system processing depends on state climates of women and children's health and wellbeing. Specifically, as women and children's health and wellbeing decrease, the gender gap in processing narrows and, in the case of court appearance, reverses. Third and finally, Hispanic youth are treated disproportionately more harshly in states with poor climates of children's health and wellbeing and in states with less punitive juvenile justice systems. Overall, the findings indicate that the reduction of gender and racial/ethnic disparities is unlikely without commitment to the structural reform of inequalities. Intervention efforts to reduce disparities should be multifaceted and include community-based youth-serving organizations and human services agencies, in addition to criminal and juvenile justice agencies.

In chapter two, we introduce gender ratios as a relatively under-exploited metric of fetal health and use it to estimate the causal impact of ambient prenatal pollution exposure on fetal deaths. Since a complete census of true fetal deaths is impossible to obtain, we exploit the differential in fetal susceptibility to environmental stressors across genders to estimate this effect. Males are more vulnerable to maternal stress in utero, and thus are more likely to suffer fetal death due to pollution exposure. We apply this metric in the context of the Clean Air Act Amendments of 1970 (CAAA) which provide a source of exogenous variation in county-level ambient total suspended particulate matter (TSPs). We find that a one standard deviation increase in TSPs decreases the percentage of live births that are male by 1.6 percentage points. We then use the observed differences in neonatal and one-year mortality rates across genders in response to pollution exposure to estimate total fetal losses in utero. Our preferred calculations suggest the pollution reductions from the CAAA prevented approximately 59,000 fetal deaths in 1972.

In chapter three, we use draft lottery number assignment during the Vietnam Era as a natural experiment to examine the effects of military service on crime. Using exact dates of birth for inmates in state and federal prisons in 1979, 1986, and 1991, we find robust evidence of effects on violent crimes among whites. In particular, we find that draft eligibility increases incarceration rates for violent crimes by 14 to 19 percent. Correspondingly, two-sample instrumental variable estimates imply that military service increases the probability of incarceration for a violent crime by 0.27 percentage points. Results for nonwhites are not robust. We conduct two falsification tests, one that applies each of the three binding lotteries to unaffected cohorts and another that considers the effects of lotteries that were not used to draft servicemen.

Data were drawn from the National Longitudinal Survey of Youth (1997), which consists of a nationally representative sample of 8,984 youth who were aged 12 to 16 years in 1997. Using hierarchical generalized linear models, ordinary least square regressions, and binary logistic regressions, this dissertation comprises three empirical papers investigating racial/ethnic differences in the relationship between wealth and health. The first paper examines the racial/ethnic differences in the time-varying wealth-health associations during young adulthood from age 20 to 30. The second paper examines racial/ethnic heterogeneity in the relationship between parental wealth and substance use behaviors, including cigarette smoking, heavy episodic drinking, and use of cannabis. The third paper examines how US adults' wealth components (financial assets, nonfinancial assets, secured debt, and unsecured debt) at age 30 were associated with physical and mental health at ages 30-36, as well as if the associations differed by race/ethnicity.

Overall, positive relationships were found between wealth and health. Adding wealth significantly attenuates the association between income, education, and racial/ethnic differences in health. Significant racial/ethnic heterogeneities were found across three studies. Together, these three papers offer a more nuanced understanding of the complex relationship among race/ethnicity, wealth, and health. The findings could inform inclusive, progressive asset-building policies, racial/ethnic tailored public health interventions, as well as wealth and health integrated services.

Chapter 2: How homeschooling affects homeschooled youth is critical to the public acceptance of homeschooling as a viable education choice. I focus on homeschoolers' college admission tests, college enrollment and degree, and labor income in their early career. I found that generally homeschoolers fall behind regular school students. I also use OLS and probit models to test for a relationship between homeschooling and college degree and labor income. I found negative and statistically significant effects of homeschooling on both completion of college degree and labor income. To control for selection issue I use propensity score matching and a methodology developed by Altonji et al in 2005 to reexamine the relationship between college degree and homeschooling. The results confirm the negative effect from homeschooling.

Chapter 3: Previous empirical studies have not considered whether schooling type such as public, private or homeschooling influences age at first marriage. Homeschooling could be an important factor in this decision, as it could allow parents to mold their attitudes toward marriage more closely. I use Cox Hazard Model to explore the relationship between schooling style and the timing of marriage along with other factors widely used in other literatures. As I find comparable results for many other factors that could have impact on the age of marriage, it shows little evidence that different schooling modes affect the age at first marriage. I argue the reason behind this phenomenon might be due to homeschooling efficiency and homeschoolers' sociability.

For the first question, "What effects do the parental status, marital status, and gender of a student have on educational attainment?" data collected for the NLSY97, n = 8,984, was utilized to compare academic attainment amongst participants in regards to gender, marital status, and parental status. Through a series of non-parametric tests, it was found that single, childless, female students had higher educational attainment than any other group, followed by married parent-students of both genders and single-mother students.

For the second question, "What effects do Tinto's pre-entry attributes of family background, skills and abilities, and prior schooling, have on educational attainment of the single-father student?" the impact of seven independent variables, representing Tinto's pre-entry attributes, on educational attainment for single-father students, n = 44 after removal of incomplete records, was studied. Non-parametric tests were utilized to study the relationship between the seven independent variables and educational attainment; an ordered logistic regression was conducted to study the relationship between the independent variables as a group and educational attainment of the single-father student. Results were largely non-significant; however, positive relationships were found to exist between educational attainment and occupation, parents in the childhood home, and average hours worked per week. While non-significant, these results do provide insight into potential future areas of research regarding the single-father student.

It has been suggested that dropout is the end result of a process of disengagement from school. In order to assess the effect of the event of dropout, it is necessary to separate dropout from the processes that lead to it. The extent to which this is accomplished by matching is assessed by comparing dropouts to matched non-dropouts on numerous background characteristics. As such, it is desirable to use a wide range of measures to compare the two groups.

I use the National Longitudinal Survey of Youth 1997 to address this question. Delinquency is measured in two ways: a six-item variety scale and a scale based on a graded-response model. Dropout is based on self-reports of educational attainment supplemented with official transcripts provided by high schools. Because of the breadth of topics covered in this survey, it is very well-suited to matching methods. The richness of these data allows comparisons on over 300 characteristics to assess whether the assumptions of matching methods are plausible.

I find that matching based on trajectory models is unable to achieve balance in pre-dropout characteristics between dropouts and non-dropouts. Propensity score matching successfully achieves balance, but dropout effects are indistinguishable from zero. I conclude that first-time dropout betw een the ages of 16 and 18 is not a turning point in offending trajectories. Implications for life course criminology and dropout research are discussed.

This study used sixteen years of National Longitudinal Survey of Youth 1979 data. Both pooled OLS and pooled logit regression in combination with individual FE methods were used to measure gaps in labor outcomes. In addition, to address bias arising from endogenous relationship between obesity and labor outcomes I estimate the weight effect using instrumental variables approach with county obesity prevalence as instrument for obesity. Finally, I apply Oaxaca-Blinder decomposition method to examine the factors that contribute to the observed gap in hourly earnings.

The data from wave one of the National Longitudinal Study of Youth 1979-Children (NLSY-C) and the data from wave one of the Panel Study of Income Dynamics-Child Development Supplement (PSID-CDS) were analyzed in parallel and the results were compared.

Analysis of the two datasets proceeded in two stages. At the first stage, the association between birth weight and cognitive ability was examined first using loess regression models in which both birth weight and gestational age were the predictors, and then by simple loess regression models in which birth weight was the sole predictor while gestational age was restricted to the normal range, i.e., between 37 complete weeks (259 days) and 42 complete weeks (294 days) to help control its confounding effects. The association was examined separately for subsamples defined by race and sex. At the second stage, the distribution of birth weight was broken into low, medium, and high, and the challenge of controlling multiple covariates was addressed by stratifying the three groups on the balancing score estimated from generalized propensity score models that can accommodate more than two groups.

The shape of the association between birth weight and cognitive ability as depicted by loess regression was not consistent across the two datasets. Among the children in the NLSY-C who were born near term and with normal birth weights, the association had an upside-down U-shape. Among the children in the PSID-CDS who were born near term and with normal birth weights, the association was roughly a continuously increasing one. The shape of the association persisted, and was slightly strengthened when multiple covariates that account for prenatal di fferences were controlled by balancing score-based stratification.

Results from the two data sets demonstrated consistent racial and gender differences. The association was stronger for whites than for blacks. White males had slightly higher cognitive scores than white females, while black females had slightly higher scores than black males.

I use PSID, NLSY, National Postsecondary Student Aid Study (NPSAS), and Census micro data to estimate the parameters of the model. I find that differences in higher education account for a higher percentage of the intergenerational correlation in earnings than do differences in early education. Liquidity constraints do not seem to be important for early or higher education. I also show that there exist complementarities between the two periods of investment in education. Finally, I find that early education is more important for the upward mobility of low-income families.

The second essay is motivated by a puzzle in the research literature: quantitative analyses tend to show that marriage is usually a financially-beneficial institution for low-income single mothers and their children, while, in the relevant qualitative literature, such women often identify men's limited financial resources as a key reason for their remaining unmarried. I attempt to identify the functional form of the relationship between men's earnings and their marital status, and I find that it is best described by using a transformed variable that expresses earnings as a percentile ranking relative to one's peers. I conclude that many low-skilled men may, as a condition of their marriageability, be required to demonstrate that they are at least as capable as their peers of improving their partners' financial prospects.

The third essay uses couple-level panel data on unmarried parents to examine the association between marital attitudes and expectations and relationship status. I find that unmarried couples hold overwhelmingly positive views of marriage but are more likely to break up than to marry over time. Multinomial logit results show that couples who are more optimistic about marriage are more likely to marry than to cohabit and are more likely to cohabit than to break up or to be romantically involved without living together. Measures of respondents' attitudes about gender roles and the trustworthiness of members of the opposite sex are also occasionally significant in the analyses of black and Hispanic couples.

The second chapter of this dissertation is titled "Fertility Response to the Tax Treatment of Children." This chapter uses variation in the child tax subsidy implicit in US personal income taxation over time and across states to estimate the effect of a decrease in the cost of raising a child on fertility. In a sample of 18,592 women age 20 to 43 from the Panel Study of Income Dynamics and the National Longitudinal Survey of Youth (NLSY79) surveyed between 1968 and 2013, we estimate that the tax subsidy for having a child does not seem to cause a significant fertility response, but some subgroups of the US population do have a positive and economically significant fertility response to the child tax subsidy. There are larger, statistically significant fertility effects for low-income women, single women, and women in the earlier half of our sample. The evidence suggests that not all child tax subsidy changes are equally salient for these subgroups as the fertility response is driven by increases to the Earned Income Tax Credit and not the value of the personal exemption or by increases to the Child Tax Credit.

In the final empirical chapter, I take an instrumental variables approach in estimating the causal returns to attending a selective institution. These estimates build upon prior findings in a few key manners. First, the breadth of data available in the NLSY79 permits the analysis of total family income and total family wealth, in addition to hourly wages or annual earnings. Also, the most recent waves of the NLSY79 report data on respondents through the age of 45, spanning an age range in which the measurement of SES is least susceptible to error. Findings from this chapter show that, in comparison to attending a less selective college, accessing a selective four-year institution leads to a greater likelihood of completing a bachelor's degree and attaining an advanced degree, higher hourly wages, and greater total family income and family net worth.

Results. The patterning of risk exposures from birth to age 8 was found to predict the initial level of average academic achievement and antisocial behavior at age 8, and the rate of change in average academic achievement and antisocial behavior over time between ages 8 to 14. Gender and race/ethnicity were found to influence both the patterning of childhood risks and the growth parameters of the development trajectories of academic and behavioral competence in early adolescence. Results for the influence of latent risk class membership and trajectories of competence on adolescent psychological well-being were inconclusive.

The first chapter provides the first evidence of the effect of a U.S. paid maternity leave policy on the long-run outcomes of children. I exploit variation in access to paid leave that was created by long-standing state differences in short-term disability insurance coverage and the staggered enactment of laws that banned discrimination against pregnant workers in the 1960s and 1970s. While the availability of these benefits sparked a substantial expansion of leave-taking by new mothers, it also came with a cost. I find the enactment of paid leave led to shifts in labor supply and demand that decreased wages and family income among women of child-bearing age. In addition, the first generation of children born to mothers with access to maternity leave benefits were 1.9 percent less likely to attend college and 3.1 percent less likely to earn a four-year college degree.

Chapter 2 examines the labor-market consequences of a broad expansion of access to paid maternity benefits. The theoretical implications of maternity leave policies are ambiguous, with the potential for positive effects that stem from greater attachment to the labor force among mothers but also negative effects that could result from shifts in relative labor demand. I show that the enactment of maternity benefits through STDI slowed the convergence of the gender wage ratio by 31 percent between 1975 and 1985. I also provide evidence that this effect was driven in large part by apparent substitution of men for women into high-profile professional and management positions.

First, drawing from literature in economics and sociology, I propose a dynamic, life course model of indebtedness that proposes three mechanisms that drive differentiation in household indebtedness: institutional context (period), social heterogeneity, and patterned disadvantage, or structural risk. Second, I use multilevel logistic regressions to explore the association between the hypothesized mechanisms and the likelihood of holding non-collateralized debt. While experiencing negative life course risks increases the likelihood of holding debt, I find that occupying positions of structural disadvantage--being black, being in poverty--decreases the likelihood of holding debt, while having advantages--higher education, being married, holding assets--increases the likelihood of holding debt, pointing to distinct differences in who can access debt to buffer life course shocks and who cannot. Examining the interrelationships between debts and assets further underscores the tenuous economic well-being of the disadvantaged. I find that those most likely to experience negative life events are both less likely to have financial assets with which to buffer these events and more likely to experience constrained access to on-collateralized debt.

Third, I employ multilevel linear regressions to examine the association between the proposed mechanisms and three unique indicators of debt burden. I find that many of the standard coefficients included in models of net worth are not significant predictors of the level of non-collateralized, non-revolving debt, suggesting that we know much more about the correlates of income and wealth than we do household debt. Rather, variation in debt burden may be better understood by heterogeneity in non-economic variables. To better estimate this heterogeneity, I utilize latent class regression models to estimate the early life course trajectories of debt burden for the NLSY79 cohort. I find four distinct trajectories of indebtedness, with varying consequences for later life financial outcomes. Overall, I conclude that household debt is nuanced and contextually contingent and can add to our understanding of long-term stratification processes when studied as a unique indicator of inequality.

Chapter one: Smoking, birth weight, and child development: Evidence from the NLSY79. One important justification for higher cigarette taxes is that pregnant smokers harm their children through in utero exposure. This paper investigates the impact of smoking while pregnant on subsequent child development, using the NLSY79 Mother-Child data, which allows for between-sibling comparisons. In utero exposure is related to an increase in behavioral problems among children and adolescents, but has no effect on intelligence. Targeted smoking cessation interventions may be cost-effective on the basis of nicotine's effects on child and adolescent development, though the optimal cigarette tax based on this justification is very small.

Chapter two: An empirical analysis of "acting white" (with Roland Fryer). There is a debate among social scientists regarding the prevalence of an insidious peer effect commonly referred to as "acting white". Using the National Longitudinal Study of Adolescent Health which allows one to construct an objective measure of a student's popularity, we demonstrate that there are large racial differences in the relationship between popularity and academic achievement, which we label "acting white". The patterns in the data appear most consistent with a signaling model in which investments in education are thought to be indicative of an individual's opportunity costs of group loyalty.

Chapter three: The political response to recent changes in U.S. wage inequality. While there has been much work on the causes behind recent changes in the wage structure, it is unclear to what extent voters and politicians have agitated for political reforms in response to increasing inequality. This paper investigates House and Senate voting records and House Election results to examine the political response to wage inequality. The results suggest that inequality is related to turnover in the House, leading to a slightly greater number of Democrats on net. Voting records show that inequality leads to polarization by party, with Democrats voting more liberally and Republicans voting more conservatively. The results are consistent with models in which relative position in the income distribution matters, and where the "bite" of inequality hurts the bottom more than the top.

Almost 1 out of 3 white women in the U.S. are obese and data on white women born in 1960-1964 from the National Longitudinal Survey of Youth 1979 (NLSY79) show that, at age 30, obese women have completed almost 1 fewer school grade, are less likely to participate in the labor market, and if they work they have wages that are lower by 17%; they are more likely never to have been married and, if they are married, their spouses have incomes that are lower by 27%.

I interpret these facts taking into account that (a) body mass may affect labor- and marriage-market opportunities; (b) behavioral factors, potentially influenced by schooling attainment and family income, play a key role in the accumulation of body weight; and (c) women may be heterogeneous in terms of their propensity to gain weight on the one hand and labor- and marriage-market endowments on the other hand.

To this end, I specify and estimate a dynamic model in which (a) wage and spousal income offers, as well as the arrival probability of marriage offers, depend on body mass; and (b) from the time they leave school, women make decisions about their labor market participation, marital status, and body mass to maximize lifetime expected utility. Their utility function represents preferences over consumption, leisure, marital status, and body mass, where preferences over body mass capture the psychic costs of keeping the balance between energy intake and energy expenditure. Such preferences depend on recent fertility history, consumption, and schooling attainment; reflecting the role played by family background factors in the accumulation of body weight, they also depend on unobserved permanent characteristics, as do wage and spousal income offers. The model is estimated by the method of simulated maximum likelihood using a pa nel dataset on the body mass, labor market, marriag e market, and fertility histories of white women born in 1960-1964 from the NLSY79.

My results can be summarized as follows. In line with results on schooling attainment and labor market outcomes, I find that unobserved permanent characteristics play quite an important role also in the determination of body mass outcomes, explaining 16% of the cross-sectional variation in obesity at age 30. Furthermore, these characteristics are important in accounting for the observed relationship between body mass on the one hand and wages and spousal incomes on the other hand: the type of women who are most likely to become obese receive wage and spousal income offers that are lower by respectively 53% and 29%.

In itself, obesity has negative consequences in the marriage market, but not in the labor market: the odds ratio of receiving a marriage offer declines by 15% for obese women, who also receive spousal income offers that are lower by 7%. Without such marriage market incentives, the prevalence rate of obesity at age 30 would be higher by 21%.

More than 10% of women in my sample are already overweight when they leave school. In the concluding counterfactual experiment I address the following question: how effective in preventing adult obesity would be policies aimed at eliminating excess weight at the time of school leaving? I find that the prevalence rate of obesity would be persistently lower, even though initially overweight women are intrinsically much more inclined to gain weight than the average.

Using NLSY79 data from 1990-2000, this paper employs an endogenous switching regression model to analyze wages of day and night shift workers and shift choice. The model is estimated using both the Lee two-step method and maximum likelihood. Two measures of local economic conditions, the local unemployment rate and the state leading index, are used. The models provide evidence that shift differentials and local economic conditions significantly impact shift choice. Of the two local economic condition variables used in the analysis, the leading index is a stronger predictor of shift choice. This paper develops a new method of analyzing the impact of the interaction between the shift differential and local economic conditions on shift choice, providing limited evidence that compensating differentials for night shift work may be lower when local economies are weak. The calculated interaction effects are small. Estimated wage premiums for night shift work are negative, and are approximately half of day wages in the 1990 cross-section. Estimated wage differentials for night shift work are smaller in pooled cross-section analysis, ranging from roughly 2% to 11% below day wages. Analyzing cross-sections over time indicates that shift differentials were below day wages throughout most of the 1990's but in 2000, night wages were approximately 7-11% higher than day wages. Overall, the results provide evidence that individuals take both the size of the wage premium and local labor market conditions into account when selecting working hours.

The first chapter of my dissertation, "Does Affirmative Action Help or Hurt the Production of Black Lawyers?," considers whether affirmative action actually harms the production of black lawyers in the U.S. Affirmative action has been a controversial policy since its inception in the 1960's as it relates to both the labor market and the education system. The central question focuses on how affirmative action affects minorities, specifically in law schools and in the production of black lawyers. Using a detailed survey of law students conducted by the Law School Admissions Council (LSAC) from 1991 to 1996, I examine whether affirmative action caused black law students to be mismatched to the wrong school and whether mismatch reduced the number of black lawyers in the U.S. According to the mismatch hypothesis, students would have obtained better outcomes if they were not mismatched and had attended a lower-level school that more closely matched their ability level. Opponents of affirmative action believe that mismatch is one of the direct consequences of the policy (Wilkens, 2005). I present evidence that mismatch caused at least a 5.5 percent reduction in the number of black lawyers, but as Rothstein and Yoon (2006) conclude, this effect may be more negative.

The second chapter of my dissertation, "Decoding the GED Signal: The Role of Non-Cognitive Ability and Measurement Error," considers the role of non-cognitive ability in explaining the low and negative returns to a GED credential. Bowles, Gintis, and Osborne (2001) have shown that non-cognitive ability actually explains much more of the wage variation than cognitive ability. It is clear that abilities besides intelligence are rewarded in the labor market, but identifying clear measures of non-cognitive ability and considering its effect on education and labor market outcomes are relatively new questions in economics (Heckman and Rubinstein, 2001). I find that GED recipients do have lower non-cognitive abilities and that it explains their relatively low economic returns. However, the observable measures of non-cognitive ability are at best proxies for unobserved ability, so I utilize instrumental variables and a latent variable model to consider the effect of measurement error.

The third chapter of my dissertation, "The Effect of Unilateral Divorce Laws on Incentives to Invest in Children and the Behavioral Outcomes of Children," utilizes state-variation in unilateral divorce laws to consider the law's effect on family investment on an important type of marriage-specific capital; children. I also consider how a reduction in family investment in "marriage-specific capital" impacts children by examining their behavioral and labor market outcomes using the National Longitudinal Survey of Youth 1979 (NLSY79) and the NELS data sets. Unilateral divorce laws were passed in many states primarily in the 1960's and 1970's. The policy provides an exogenous shock that may affect divorce rates, household bargaining, and children's future outcomes. Previous research has primarily focused on two specific issues: the impact of these laws on divorce rates using Census data and the implications of divorce laws on bargaining power within the household. This chapter focuses on the latter question.

Previously documented correlates of general victimization and offending serve as a conceptual framework for the models presented. These elements were tested to determine if common covariates of general offending (vandalism, stealing, hurting someone) influence bullying behaviors as well.

Data for this dissertation are derived from the National Longitudinal Survey of Youth (NLSY) and National Longitudinal Survey of Youth Ch ild Sample (NLSY-C). The analysis focuses on children in elementary and middle schools from 1994 to 1996 (N = 959). The age range in 1994 (Time 1) was from 10 to 12 years old and two years older in 1996 (Time 2). When investigating the duration of roles over time, results revealed that bullying behaviors were more persistent over time than general offending behaviors. A significant and strong predictor of bullying at Time 2 is having been a bullied victim at Time 1. Few of the criminological measures or other covariates have any significant influence on either bully victimization or bully offending.

Findings also indicated that bullying and general offending were qualitatively distinct phenomena. Neither prior general offending nor current general offending is a significant predictor of bully victimization or bully offending. Furthermore, the common correlates of general offending are not correlated with bully behaviors. The current study found that measures associated with criminological theories (school, parents, and peers) were not associated with bullying measures.

The second chapter is the first attempt to study the long-term effects of Medicaid on children's health outcomes by looking at the effects of the same Medicaid expansions that took place in the later 1980s and early 1990s. These expansions significantly increased the percentage of pregnant women and children eligible for Medicaid but did so at very differential rates across the states. The substantial variation in Medicaid eligibility thresholds by state, and year, and the age of the child provide the identifying variation for the analysis. By using restricted access data, containing state of birth and state of residence of children, I match children to the Medicaid eligibility rules in their year of birth and currently. The results suggest that the expansions were effective in improving the health of children from low-income families in the long run. Increased Medicaid eligibility at birth is associated with better health outcomes in the future. But interestingly eligibility at older ages (conditional on eligibility at birth) is not.

The third chapter examines the effects of using friends, relatives and acquaintances in job search on current and future wages and job tenure of individuals using the National Longitudinal Survey of Youth 1979 (NLSY79) data. Individuals who use contacts may differ from those that don't. For example, both females and African-Americans are less likely to use contacts in job search in the US. This paper uses switching regression models to deal simultaneously with an endogenous selection issue in contact's choice and the existence of two different regimes of wage and job tenure determination. Econometric estimates provide evidence for the existence of a selection effect on the choice of informal contacts and, after correcting the selection bias, using contacts has a positive effect on both wages and job tenure. The paper also explores whether some types of contacts result in greater wages and longer tenures. Gains from using informal contacts are largest for those that use male contacts of an older generation, rather than female or younger contacts.

The second chapter draws on data from the National Longitudinal Survey of Youth, 1997 cohort to investigate how men's and women's economic trajectories influence the transition to marriage among cohabitors. I find that growth in earnings and increased schooling hastens the transition to marriage, whereas a loss in earnings or employment encourages separation. The relationship between economic status and marriage varies little across gender and parental status, though the associations are more consistent among men and especially, fathers. The results of this study provide further evidence that having limited economic resources presents a significant barrier to marriage.

The first chapter examines how married couples' migration decisions differentially impact men's and women's earnings and the role that policy can play in improving post-move outcomes for trailing spouses. I use a difference-in-differences methodology to show that access to unemployment insurance for trailing spouses increases the likelihood that households move by 2.3 p.p. and improves the post-move labor market outcomes of women. I then build and estimate a structural model of dual-earner couples' migration decisions to evaluate the effects of a series of counterfactual policies. I show that increasing the likelihood of joint distant offers substantively increases migration rates, increases women's post-move employment rates, and improves both men and women's earnings growth at the time of a move.

The second chapter explores the role that joint geographic constraints play in dual-earner household migration decisions. I develop a measure of joint geographic constraints adapted from a pairwise occupational co-agglomeration index and demonstrate that being well-matched to one's spouse in terms of occupation clustering is positively associated with earnings for women and secondary earners. I show that higher values on the co-agglomeration index is associated with higher mobility rates for dual-earner households as well, consistent with the theory that occupational sorting impacts married couple's ability to overcome dual-earner migration frictions.

The first chapter analyzes the effects on high school graduation and other academic outcomes of academic requirements for participation in high school athletics. I use a simple conceptual framework to illustrate the possible effects of the requirement and derive testable predictions. Then, I combine data from the National Longitudinal Survey of Youth 1979 (NLSY79) with data on the rules to test the model's predictions. I find that requiring athletes to pass one additional course increased the overall likelihood of graduation among boys by two percentage points but did not affect female students, who at the time had limited access to interscholastic competition.

The second chapter (with Josefina Posadas) examines the role of grandparents' child-care provision on mothers' labor market participation. Using the NLSY79 and data from eleven European countries (SHARE), we find significant differences in characteristics of families who rely on this form of child care. Both ordinary least squares and instrumental variables estimates show that the availability of grandparents' care is linked to an increase in the probability of female labor force participation.

The third chapter (with Xavier Giné and Mónica Martínez-Bravo) studies the labor supply of Indian boat-owners. It uses daily data on labor force participation and the value of catches to test whether the response of labor supply to increases in wages and income is better explained by the conventional framework of inter-temporal substitution or by reference-dependent preferences. This chapter shows that boat-owners' labor participation depends not only on their expected earnings, but also on their recent earnings, supporting income-reference-dependent preferences models. However, the response to changes in recent income is small relative to the response to changes in expected earn ings. Furthermore, the results imply that short-term labor supply models should include recent earnings conditional and recent effort as control variables. Since recent earnings are positively correlated with expected earnings and negatively related to the probability of participation, omitting this variable yields downward-biased elasticity estimates.

A method is provided here that overcomes the lack of data for body composition. The clinical information available in the Third National Health and Nutrition Examination Survey 1988-94 (NHANES III) is used to estimate body composition for the survey participants in the National Longitudinal Survey of Youth (NLSY 1979). The inclusion of estimated body composition in the estimated wage equation shows that the effect of lean body mass on the wage rate is positive while the effect of body fat is negative.

Estimated body composition is then used to examine the role of physical differences in the gender wage gap. The decomposition of the gender wage gap shows that most of the previously unexplained differences in wages between men and women can be attributed to the gender differences in body composition. The explanatory power of estimated body composition rises significantly with occupational physical strength requirements. This result suggests that estimated body composition is capturing occupational requirements previously omitted from the past studies.

The findings presented in this dissertation indicate that body composition plays an important, though previously unidentified, role on wage determination. It is clear that capital investments in body composition yield economic dividends by impacting hourly wages of workers. Empirical studies that do not address differences in body composition risk obtaining biased results. Future public health policies should take into consideration the combined but opposite effects of body fat and lean body mass. It is not body size alone, but the compositional makeup of the human body, that public health policies may need to address.

The second essay examines the link between breastfeeding and childhood obesity. Heath economics researchers view breastfeeding as a determining factor as to whether a child becomes obese. There are many theories, involving both biological and psychological factors, as to why breastfeeding is negatively linked to childhood obesity. This essay argues that the breastfeeding decision is not an exogenous one, so estimation technique such as ordinary least squares is not the correct way to estimate the relationship between breastfeeding and childhood obesity. Instruments are used to generate exogenous variations in the breastfeeding variable. After correcting for any estimation bias due to the breastfeeding variable being endogenous, this study documents the benefits of breastfeeding.

The third essay analyzes 19 semesters of student evaluations at Kansas State University. Faculty fixed effects are sizable and indicate that, as assessed by students, the best principles teachers also tend to be the best non-principles teachers. OLS estimates are biased because principles teachers are drawn from the top of the distribution and because unmeasured faculty characteristics are correlated with such variables as the response rate and student effort. Student ratings are lowest for new faculty but stabilize quickly. Expected GPA of the class is not an important determinant of student ratings, but equitable grading is; and the rewards for equitable grading appear larger for principles classes. The lower ratings in principles classes are fully accounted for by greater class size.

To address this gap in the literature, I develop an index of advantage that quantifies the number of advantages individuals accumulate throughout their education, including such factors as school resources, educational aspirations, and coursework taken. A total of 13 items covering three domains (individual, family, and school) are included in the index, with each item weighted by its independent effect on college attainment. I use data from the National Longitudinal Survey of Youth (NLSY79), a nationally representative sample of young men and women who were 14-21 years old in 1979, restricting the sample to civilian respondents self-reporting as black, Hispanic, or white, and for whom data was collected for work limitations and/or depressive symptomology on at least one time point, to test whether or not (1) increasing number of educational advantages as well as educational attainment is related to physical and mental health over time, and (2) educational advantages and educational attainment result in diverging health trajectories between respondents with high versus low educational advantages, and between respondents with high versus low educational attainment.

The results suggests that the index is associated with a widening disparity over time in predicted probabilities of work limitations and the level of depressive symptomology between respondents in the 10 th and 90 th percentiles on the index of advantage. Similar results were found between respondents with a college education versus those with less than a high school education. Our results point to a need for more extensive examination of the educational system as a potential mechanism of existing health disparities as well as a viable area for future research and policy intervention.

The study in the second chapter focuses on how Conscientiousness, a personality trait, helps workers mitigate the adverse effects of graduating during a recession on early career outcomes. By analyzing college graduates who graduated in the 1980s, I find that Conscientiousness reduces the income losses of workers who graduate during a recession. More specifically, those whose Conscientiousness scores are in the upper quartile are sheltered from the losses. The mitigation effect primarily results from workers' adjustments in their labor supply. Workers high in Conscientiousness tend to work more weeks, try harder to find full-time jobs, and work more hours in these full-time jobs in response to the adverse labor market entry conditions. However, this study does not find any mitigation effects for cognitive ability.

The simulation results showed that the proposed model can recover the true trait levels more accurately than the selected alternative methods and the estimation of the person trait levels is more accurate with more items and lower proportions of zeros. Regarding the accuracy of the item parameter estimates, middle proportions of zeros, larger sample size, and more items provide more accurate estimates under the tested conditions. When sample size was larger than 2000, the item parameters were estimated accurately in most conditions. The simulation results also showed that both marginal maximum likelihood estimation method (MMLE) and Bayesian estimation (BE) methods can provide accurate item parameter estimates with large enough sample sizes. Each estimation method had its own advantages and disadvantages in computation ti me and convergence rate.

The empirical results included many outcomes that were not obtained using previous methods, especially in investigating the psychometric properties of the multiple substance use items from both propensity and level perspectives. Limitations and future directions of this study are discussed.

Results indicate that "ladder jobs" demonstrated a significant crime-decreasing effect, while employment exhibited a crime-increasing effect. In addition, the magnitude rate of "ladder jobs" versus employment increased as youth aged; that is, the advantages of "ladder jobs" gradually outweigh the disadvantages of employment in the sense of crime prevention. Furthermore, job income partially mediates the crime-increasing effect of employment on delinquency, and job stability partially mediates the crime-decreasing effect of "ladder jobs" on delinquency. However, parental control, which is measured as direct supervision, does not play a mediating role between employment and delinquency. In sum, from a crime-prevention standpoint, a job that pays little now, but improves the chances of a long-term career appears to better than a dead-end job that pays comparatively well in the short-term. The findings also imply that the discussions of employment and of internships among youth should address the importance of future-oriented feature of occupations, and not just the immediate monetary gains from the employment.

Then the relationship between maternal schooling and children's high school outcomes is investigated in the second chapter. Three outcomes are high school completion, high school diploma receipt and high school graduation grades. Using changes of compulsory attendance laws as instruments for mother's high school completion, the results suggest that mother's education is exogenous in the estimation of high school completion. Having a mother with at least a high school education will increase her child's probability of completing high school by 8 percentage points. While mother's schooling and family income have significant effects for all three outcomes, mother's cognitive ability is only meaningful to the outcome of grades. The effect of mother's high school completion is similar for sons and daughters, wealthy and poor families, and mothers with different cognitive abili ties.

Besides intergenerational transmission of education, I also work with Prof. Ozturk on occupational promotion. We found that, in models with no controls for individual unobserved factors, females are less likely to be promoted in highly female jobs. Males on the other hand are more likely to be promoted in these jobs compared to their male counterparts in jobs with lower percentage of females. However, the role of occupational feminization is no longer significant once unobserved heterogeneity is controlled for along with the skills/ task measures of an occupation. We do find overall wages to be lower for everyone in jobs with a female majority. However, there are no significant differences in the return to promotion by gender once occupational measures are controlled for, even though the occupational wage gap due to feminization persists.

In the second study, I focus on the effect of sibling structure on educational attainment and the role of siblings’ education in this relationship. Data from the National Longitudinal Survey of Youth 1979 (NLSY79) are used for analyses. I find that the effects of sibship size and sibling sex composition on educational attainment are mediated through siblings’ educational achievements. These effects are divergent for men and women. For women, sibship size and sex composition do not impact their educational attainment after accounting for siblings’ educations. For men, only the number of brothers (but not sisters) has a negative effect on their educational attainment after controlling for siblings’ educational achievements.

In the third study, I investigate the influence of birth order on the prestige and sex type of adolescents’ occupational aspirations using the first wave of the NLSY79. The results indicate that for both females and males, firstborn and lastborn adolescents on average expect higher prestige occupations compared to middleborns, and lastborns are more likely to have nontraditional occupational aspirations than firstborns and middleborns. Taken together, the results suggest that the gender gap in important child and adult behavioral outcomes is smaller among individuals with fewer siblings, fewer brothers, and among lastborn young adults.

Among findings, I observed significant associations between parental divorce when emerging adults were ages 18/19 with higher overall depressive symptom scores at ages 18/19 and 20/21. Emerging adult sex did not moderate the association between parental marital status and depressive symptoms. However, higher parental marital conflict scores during adolescence moderated the association between marital status at age 18/19 and depressive symptoms at ages 22/23 and 24/25, such that when parents divorced when the emerging adult was 18/19, higher parental conflict was associated with higher depressive symptomology at ages 22/23 and 24/25 compared to those whose parents remained married at 18/19, who did not differ on depressive symptoms regardless of marital conflict.

For quantitative discipline, the model matches data on the wage premium to occupational experience and the co-movement of occupations' productivity. The distribution of duration is then endogenous. For comparison's sake, if a standard model with homogeneous job seekers matches the job finding rate, then it also determines expected duration and understates it. That standard model implies just over half of the long-term unemployment in 1976-2007 and almost no rise in the recent recession. But, with heterogeneity by occupation, this chapter nearly matches long-term unemployment in the period 1976-2007 and 70% of its rise during the Great Recession.

The second chapter studies the link between wage growth and the match of a worker's occupation and skills. The notion here is that if human capital accumulation depends on match quality, poor matches can have long-lasting effects on lifetime earnings. I build a model that incorporates such a mechanism, in which human capital accumulation is affected by imperfect information about one's self. This informational friction leads to matches in which a worker accumulates human capital more slowly and has weaker earnings growth.

Second, this dissertation is also one of the first studies to examine whether the relationship between residential mobility and desistance from crime depends on the context of the move. Although the average effect of moving may be protective, the effect likely depends on a number of factors. Two factors may be especially salient to residential moves during the transition to adulthood: (1) whether the move occurs in the presence of other adult social roles and (2) whether the move results in improvements in community context. There are reasons to expect residential mobility to have stronger or weaker effects depending on these features.

This dissertation uses public and restricted geocode data from the National Longitudinal Survey of Youth 1997 (NLSY97). These data contain a wealth of information about the transition to young adulthood, including respondents' residential mobility, crime and substance use, adult social roles, and community context. In addition, restricted geocode data allows me to construct residential mobility patterns of respondents from 1997-2013 and determine the county-level characteristics of every residential location respondents reported living at during the survey.

In particular this study answered the following questions: (1) What is the relationship between the amount of training employees received from the workplace and wage? (2) What is the relationship between the amount of training employees received from the workplace and career aspirations achievement? (3) What is the relationship between employees' self-concept and wage? (4) What is the relationship between employees' self-concept and career aspirations achievement?

The results of the correlation analysis supported three of the four research hypothesis. Training was shown to have a positive relationship to the work outcomes of wage and career aspirations achievement. Self-concept was only found to have a positive relationship to career aspirations achievement. Even though training and self-concept were shown to have positive correlations with wage and career aspirations achievement, when the regression analyses were conducted they were shown to have very little direct impact on the amount of variance associated with each of these variables.

It appears that it is not the individual variables that impact the work outcomes of wage and career aspirations achievement, but that it is the way these variables interact with the individual and workplace variables that impact the work outcomes. In addition, variables outside of the scope of this study may account for a large portion of the variance in wage and career aspirations achievement. The interaction between the variables in this study, and other variables outside of the scope of this study, are worthy of further investigation. Finally, further study in regards to the interaction of the variables used in this study is required to gain a fuller understanding of how much influence they have on each other and the overall impact on wage and career aspirations achievement.

Methods : The estimates in this report were derived from the National Longitudinal Survey of Youth 97...Cross-sectional estimates were calculated using the SAS 9.2 statistical package to account for the complex survey design. Descriptive and inferential statistical analysis was conducted on the weighted data.

Results : Within the sample, the estimated prevalence of psychological distress (PD) among young adults within the U.S. population in 2008 was an estimated 11.34%. Those who reported life satisfaction scores of 1 - 5, were more likely to self report PD (COR 14.83; 95% CI, 10.66 - 20.62). The association remained significant in the adjusted analysis. Persons reporting being uninsured (COR 2.04; 95% CI, 1.66 - 2.49) or having partial-year unknown source (COR 2.37; 95% CI, 1.80 - 3.11), or having partial-year government coverage (COR 2.93; 95% CI, 1.84 - 4.67), or classified as full-year government coverage (COR 3.51; 95% CI, 2.74 - 4.50) were more likely to experience PD than full year private coverage. However, the associations were not significant when controlling for covariates. Significant associations were observed between young adults who had military experience, incarceration experience, or were past victims of violent crimes when compared to those who were not in the crude analysis only. Health insurance coverage status generally did not significantly influence young adult's psychological well-being in the study.

Conclusions : The associations between mental health status and health insurance coverage status may have been strongly influenced by socio-demographic characteristics. Young adults, who reported poorer social health and poorer overall health when examining health insurance coverage status, were more likely to report poorer mental health status. It is important to understand ones mental, social, and physical health, in order to truly detail one's comprehensive health. Integration of mental health screening, screening for social health and the implementation of health care reform may help to improve mental health outcomes among young adults, especially the disadvantaged.

The National Longitudinal Survey of Youth, 1997 (NLSY97) is used in this study to test hypotheses derived from Hirschi's (1969) social bond theory and Baumrind's (1966) parenting typology. These hypotheses examine the effects of family process variables and parenting styles on adolescent delinquency moderated by the effect of the child's race/ethnicity and gender, and class of the family. Based on OLS Regression results of the study revealed there was a negative relationship between most, but not all of the family process variables and delinquency. As hypothesized, as parent-youth relationship, communication, monitoring and limit setting increased, delinquency decreased. In other analyses authoritative parenting compared to authoritarian and neglectful parenting resulted in less delinquency. When separate equations were estimated this pattern of findings held for Whites and Blacks but not Latinos. White and Black adolescents with a neglectful or authoritarian mom were more likely to be delinquent than White and Black adolescents with an authoritative mom. Dad parenting was only significant for Whites indicating that adolescents with authoritarian dads were more likely to be delinquent than Whites with authoritative dads. There was no effect of parenting on Latino respondents.

Similar results were revealed when separate equations were estimated for males and females. That is, males and females with a neglectful or authoritarian mom were more likely to be delinquent than males and females with an authoritative mom. Dad parenti ng was only significant for males indicating those with authoritarian dads were more likely to be delinquent than males with authoritative dads. T-statistics indicated there were no significant differences between males and females.

Class of the family did not have an effect and there was no interaction between the parenting styles and class. However, this could be attributed to its poor measurement.

To do this several statistical procedures are utilized, and the results of selected descriptive and logistic analyses are presented. The descriptive statistics control for race/ethnicity and examine patterns within the groups; these findings display important relationships to onward and return migration. The inferential statistical method employed is logistic regression for the sample as a whole, which examines the effects across the groups, and the direction of migration.

Where past research has not investigated the complexities of repeat migration in combination with race/ethnicity, there are several notable results from this study. Specifically, this research finds that in terms of onward migration, whites are significantly more likely to move onward than are blacks or Hispanics even after controlling for key socioeconomic factors. Changes in marital status are significantly related to migration, and to the direction of repeat migration; individuals who change from "single to married" are likely to be onward migrants, whereas those who change from "married to single" are likely to be return migrants. This study finds there are differences in rates of return migration by level of education for racial/ethnic groups. Moreover, the relationship between onward migration and employment status is different for Hispanics than blacks and whites.

Chapter two shifts in focus from exploring the mother's outcomes to evaluating children's mental health outcomes. A large body of theoretical and empirical research explores the causal effect of the number of siblings on various dimensions of children's outcomes. I estimate the impact of increases in sibship size on children's mental disorders using matched mother-child data from the National Longitudinal Surveys of Youth. Using an instrumental variables technique, my findings provide no substantial evidence in support of the traditional quantity-quality fertility trade-off. By exploiting the panel data and using fixed effects to account for omitted factors, my findings show that an additional younger sibling is detrimental for a child's anxious/depressed index and likelihood of visiting a psychiatrist for a mental disorder. The estimated effects are greater for female and non-black children. This relationship is larger in magnitude in the long-term, as compared to a shorter time horizon.

The second essay asks whether government-provided child care increases the employment of mothers. I use NLSY Head Start enrollment data to calculate non-experimental estimators of the average treatment effect of participation on the mother's employment, including matching and weighting on the propensity score. I find no statistically significant effect of the treatment on employment using these methods, a finding confirmed by using several comparison groups. In addition, negative and significant effects are found for white mothers. However, using a regression discontinuity (RD) design resulted in small positive effects of Head Start participation for mothers' employment growth (5%) for some sample restrictions, but no effect using other samples. For those mothers participating in welfare, some sample restrictions using RD resulted in a larger positive effect of Head Start of 8%. The RD estimates differ substantially from those of the matching and weighting estimators, which could suggest that the latter do not fully remove the selection bias.

Given the longitudinal approach, this study starts with a comparison of patterns of school-to-work trajectories in four distinct welfare state regimes: the United States, the United Kingdom, Germany, and Sweden. By aligning the individual pathways of individuals between age 16 (enrolled in high school) and age 25, social sequence analysis enabled us to reveal sharp differences in school leaving pace, school enrollment, and instability of early work careers. The analyses suggest that the variation in selection and sorting within youth careers can be largely explained by indicators of the different welfare state regimes. Based on comparisons of younger and older birth cohorts there is evidence supporting convergence theory of welfare states – early careers liberal states are becoming less volatile, while those in social-democratic states are becoming more insecure.

Chapter 2, "Great Expectations? Women's Work and Fertility in the Face of Career Uncertainty", is motivated by the fact that women in different occupations exhibit markedly different patterns of fertility and labor supply over the life-cycle. Women in high-powered occupations tend to give birth later and exhibit a downward-sloping participation profile across the life cycle, while their low- powered counterparts not only give birth earlier but their participation profile is instead upward sloping. Using a life-cycle model in which married households make fertility and female labor supply decisions and learn about their income profile parameters, I find that uncertainty about one's wage growth is key in explaining the different fertility patterns of women, by occupation. High wage penalties due to work interruptions help explain why a significant proportion of professional women do not return to the labor market after exiting upon childbirth.

The results indicate that mothers are more likely to have lower levels of psychological well-being compared to childless women at earlier family life stages. However, this disadvantage decreases as mothers and their children age. The mother's age at the birth of her first child also plays a role in the trajectories of the level of psychological well-being. Although the negative association between psychological well-being and motherhood appears to decline over time, those who became a mother at earlier ages experience much slower declines compared to those who did not have a first child until their early thirties. It also appears that marital status is an important moderator between parental status and psychological well-being. Motherhood is associated with psychological benefits for the married, but the opposite pattern is found for the never married. Moreover, entering a first marriage is associated with greater improvements in psychological well-being for women with a child compared to childless women. The association between motherhood and psychological well-being also varies depending on the types of marital disruption. Compared to those who remain married, divorce is harmful for women with a child; however, being a widow is detrimental for childless women. Additionally, for both married and never married women, employment is not associated with increases in psychological well-being when it is also combined with motherhood.

This research suggests that the association between motherhood and psychological well-being is contingent upon the family life stages, the age at transition to motherhood, and other roles that women hold while being mothers.

Chapter 2 results indicate that injured men do not suffer significant initial earnings losses as a result of a workplace injury. A gradual decline in earnings in the years that follow the injury develops, from earnings about 8% less than the uninjured workers the year after injury to almost 16% less five years after injury, with no sign of recovery.

In Chapter 3, the annual earnings for the female injured workers in the year of injury are 9.7% less than the uninjured workers. Recovery is quick, as the earnings of the injured female workers return to levels similar to the uninjured workers in the years following injury. Restricting the sample to women who work fulltime results in substantial and persistent earnings losses following injury.

For married men in Chapter 4 there is some evidence that their annual family incomes are lower than the uninjured men in the years following injury. In contrast, there is no evidence of any significant relative losses in family income for injured married women. There is some evidence of an added worker effect for the injured men, as the wives experience increases in their annual earnings and annual hours worked. Likewise, for fulltime women there are large losses in total family income, tempered by an increase in the husbands' labor market efforts.

Finally, in Chapter 5 I find that while the volatility of total family income is relatively stable for all men, the volatility of family income without disability payments increases in the years following injury or first report of a disability. On the other hand, the volatility of total family income and income before disability payments are consistently stable for the female sample.

Utilizing employer-reported results of equal employment opportunity law compliance from the Multi-City Telephone Employer Survey, I predict the likelihood that an individual in the National Longitudinal Survey of Youth 1979 (NLSY) was employed by a firm that considers equal employment opportunity legislation in their hiring. Using observations from NLSY years 1986-2000 and an empirical test for employer learning and statistical discrimination developed by Altonji and Pierret (1997), I examine statistical discrimination in wages by education and race for male employees in firms that are not likely to actively comply with antidiscrimination legislation.

I find that statistical discrimination by education is evident for black males but not for white males, suggesting that information for black males in the labor market is inferior. Statistical discrimination by education is then shown to be more pronounced for black males employed by firms that are unlikely to actively comply with equal employment opportunity legislation. An investigation of statistical discrimination by race in these firms is inconclusive in both a public learning framework and an asymmetric learning model.

These results show that a complete analysis of statistical discrimination must consider differences in the incentive firms have to discriminate. While there is not clear evidence of statistical discrimination by race, the results do illustrate that productivity information for black males in firms that are unlikely to comply with antidiscrimination legislation is inferior to the information these firms have about white males. Equal employment opportunity laws may correct a market failure of underinvestment in information by firms, suggesting that policies to enhance the information available to firms not subject to these laws may be beneficial.

Chapter 3 examines the relationship between student loan debt and post-graduation living arrangements for college graduates in the United States, taking labor market factors into account. Using data from NLSY97, I explore both the parent-youth co-residence outcomes at a point in time and the dynamics of college graduates' movements back home. I estimate a random effects panel probit model and a discrete time proportional hazard model. The estimation results show that an increase in the amount of student loan debt owed at the time of graduation significantly increases the hazard of moving back home after graduation.

Chapter 4 uses NLSY97 data to investigate the factors that affect the amount of educational loans students borrow to attend college, with a particular interest in whether parental divorce would affect a college student's educational debt burden. The estimation results indicate that youths from divorced families are more likely to have a higher debt burden upon leaving school than youths from intact families. Among the youths from divorced families, those whose biological parents lived in states that permit courts to extend child support beyond the age of 18 for college expenses (post-majority states) do not seem to take fewer loans than those whose biological parents lived in non-post-majority states.

The expansions in Medicaid eligibility had intended and unintended consequences on several outcomes. First of these was on private health insurance. There was a possibility that near-poor individuals who initially had private coverage switched to Medicaid. Since this indicates only a shift in financing of care from private to public insurance, the consequence would be little or no change in health care utilization and health. In the first essay, I found that Medicaid expansions had no effect on privately covered individuals, but a large impact on those who were initially uninsured.

The second potential consequence of Medicaid expansions was on infant and child health outcomes. Even though expanded eligibility gave rise to a substantial increase in enrollment rates, the effectiveness of the public program was conditional on improving the health of infants and children. In the second essay, the effect of Medicaid on infant and child health has been extensively examined and only limited evidence has been found on Medicaid coverage improving health outcomes.

A third possible effect of the Medicaid expansions was on labor supply and welfare participation of female heads. Elimination of the link between the cash assistance program and Medicaid created new incentives in the labor markets for welfare recipients. The expected consequences were an increase in labor supply and reduction in welfare rolls. I analyzed this issue in the third essay and found very little evidence on the effect of Medicaid on labor supply and welfare participation for female heads.

This study uses data from the National Longitudinal Survey of Youth 97 (NLSY 97), a non-experimental database, to evaluate the effects of U.S. youths' school-based learning (SBL) program participation on early labor market outcomes. Estimates from ordinary least squares regression, the linear regression with probabilistic matching or instrumental proxy, are compared to those obtained from bootstrapping regression analysis. Four outcome variables, including employment in college, employment, total worked hours and hourly wage rate, are used to gauge the early labor market outcomes of youth from the NLSY97.

Findings, at an alpha level of .05 when the first type of instrumental variable (IV) correction method is adopted, reveal that SBL program participants are significantly less likely than non-participants to enroll in college, and that SBL program participants have a lower probability of enrolling in college than do non-participants.

In comparison with college enrollment, findings from the analysis of employment and total worked hours outcomes on SBL program participation are not statistically significant with selection bias corrections. Rather, only the estimates derived from the Tobit regression for the correction of the censored data show that SBL program participants have a lower number of total worked hours than do non- participants. Due to evidence which shows that youths' wage differential is small in the early labor market, the findings from Heckman's two-step correction with an alpha level of .10 show that SBL program participants are more likely to have higher hourly wages than non-participants.

In addition, in looking at the seven specific types of SBL program participation, the significant likelihood of enrolling in college or being employed, all of the estimates derived from Heckman's two-step correction show no significance.

For the total worked hours outcome, the correction of the censored data using the Tobit regression shows that internship/apprenticeship program participants have lower total worked hours than non-participants, with an alpha level of .001.

Due to the small wage rate differential for youth in the early labor market, a significance level of .10 is used for this outcome. Heckman's two-step correction reveals that SBL program participants are more likely than non-participants to have higher hourly wage rates. In view of the seven types of SBL program participation, the findings from Heckman's correction reveal that internship or apprenticeship program participants are more likely than non-participants for all SBL programs to have a higher hourly wage rate.

In summary, this study shares corrected estimates from alternative approaches. Based on criteria from the 200-time and 500-time bootstrapping, the best selection bias estimation among these four corrections uses Heckman's two-step correction.

The dissertation consists of three papers. The first paper examines how initial legal status affects lawful permanent residents' (LPRs) asset building by investigating three types of financial assets--bank account ownership, investment account ownership, and retirement account ownership. The second paper tests the impact of being banked at an earlier stage of immigration on immigrants’ subsequent asset holding, with self-selection bias addressed by using a nationally representative data set. The third paper examines wealth trajectories of children from immigrant and native-born families from their mid-20s to their mid-30s, with a focus on the role parental financial assets play in shaping these trajectories.

Overall, the results show that how immigrants fare financially in the United States largely depends on what resources they can access in the United States. Institutional-level support in asset building and quality social networks may help them to achieve better financial outcomes. In addition, the findings reveal that children of immigrants were disadvantaged in wealth growth during their young to mid-adulthood compared to their peers from the native-born families. Together, these three papers turn new ground in extending asset-building research and policy to immigrant population, an important segment of the U.S. economy and society. The findings may inform inclusive asset-building policies, immigrant econo mic integration programs, and immigrant tailored financial services.

The third paper explores the consequences of union dissolution, specifically cohabitation dissolution, on mental health outcomes, investigating the moderating effects of gender and parenthood. The findings suggest that gender differences in the association of cohabitation dissolution with psychological distress are contingent upon the types of psychological distress under consideration and also reveal that cohabitation dissolution intertwined with non-marital parenthood is harmful to mental health, especially for young women.

My first essay “Homeownership Achievement of Second Generation Immigrants Accounting for Social Network Effects” is the first one to study the homeownership of second generation immigrants. Using the National Longitudinal Survey of Youth 1979, my paper identifies a lower propensity to own a house for native born individuals with immigrant parents compared with native-born individuals with native-born parents. I find that a metric to account for ethnic social networks (using the geographic clustering of Hispanics) can help explain the homeownership rate difference between these two groups. My estimation results show that networks of Hispanic homeowners increase the likelihood of homeownership for a Hispanic household, and the network effect is even larger for second generation Hispanics. This result suggests that the effect of knowledge spillover on homeownership decision results not only from the concentration of Spanish speakers, but also through other channels such as social activities which are rarely discussed in the literature.

The interaction of a household’s level of wealth and the down payment constraint has also been found to be among the most important influences on households’ homeownership decisions. The second essay “The Choice of Mortgage Down Payment Assistance and its Effect on Mortgage Outcomes” focuses on how households select into different mortgage down payment assistance programs (DPAs), and how this selection affects future loan performances. My study uses monthly panel data of the Ohio Housing Finance Agency’s (OHFA) Mortgage Revenue Bond (MRB) first- time home buyer program. Two forms of DPAs are available in this program. I explain how borrowers select into DPAs by using a two-step algorithm for dynamic games first proposed in Bajari, Benkard and Levin (2007). I first use a polychotomous choice model of DPA, and a survival model of the probability of default to generate hypothetical borrower choices. I then use the simulated data to recover parameters of a dynamic model of loan default, so that I can model borrowers’ choice of DPAs based on their expectations about future interest rates, house prices and income, and I then predict subsequent loan performances. Households that expect future income to increase should prefer assistance in order to smooth consumption, but the resulting higher monthly payment could lead to a higher risk of future delinquency and default. I am also able to examine whether borrowers selecting into DPAs are less risk averse, which leads to a higher probability of default. The policy question is how to improve loan performance, specifically, whether a new credit score requirement imposed in the end of 2009 will reduce future delinquency and default.

The third essay “The Neighborhood Effect of Loan Applications and Down Payment Assistance Usage” focuses on how neighbors’ decisions to obtain an OHFA MRB loan affect both households’ home mortgage loan applications and the usage of down payment assistance. Existing literature finds that borrowers with limited information about the mortgage market are more likely to select a subprime mortgage loan. By combining MRB loan data with Home Mortgage Disclosure Act data on mortgage applications, I test whether neighbors usage of MRB loans increase a household’s probability of receiving a MRB loan, and whether neighbors usage of DPA increases a household’s likelihood of using a DPA. Additionally, I control for lender influences on MRB loan and DPA usage, and compare it to neighbors’ influences. This study sheds light on two information channels of governmental loan programs: neighborhood knowledge spillovers and lender provided information. It also compares the relative importance of each channel, and thus could be used to explain why the government should provide easily accessible information to target areas.

So, it is natural to ask whether the relationship between family structure and students' educational outcomes has changed from Cohort 79 to Cohort 97? If yes, what role does the change in family structure play in explaining the change in college completion rates? The first chapter of my dissertation answers these two questions. Chapter one of my dissertation documents the changing family circumstances of U.S. college students, and their relationship to students’ patterns of college graduation using the 1979 and 1997 cohorts of the National Longitudinal Survey of Youth (NLSY). My empirical models of graduation probability show that students who come from those non-traditional families in the 1979 cohort are just 3 percentage points less likely to graduate from college than those who lived with both of their biological parents up to age 18. However, this gap increased to 9 percentage points for the 1997 cohort. A decomposition exercise based on the estimates demonstrates the share of the change in completion rates that the model attributes to the various observed explanatory variables, particularly to the change in family structure. Results indicate that students' characteristics and family background account for 45 percent of the change in graduation rate, with family structure accounting for 12 percent of the change in graduation rate. The central contribution of this analysis shows the importance of the family structure in explaining changes in college completion over the past 30 years in the United States.

Next, using the NLSY97, I observe that while studying in 4-year college, students from traditional families receive around 12000 dollars parental transfer in total, however, students from non-traditional families only receive 6000 dollars parental transfer in total. So, the goal of the second chapter is to understand whether parental transfer can explain some of the disadvantages students from nontraditional families have in college study. The results of probit model show that one thousand dollars increase in parental transfer is associated with a 2 percentage points increase in 4-year college enrollment. Moreover, one thousand dollars increase in the parental transfer is associated with a 0.5 percentage points increase in 4-year college graduation.

Chapter two develops a framework that nests both symmetric and asymmetric employer learning, and derives testable hypotheses on racial statistical discrimination under different processes of employer learning. Testing the model with data from the NLSY79, we find that employers statistically discriminate against black workers on the basis of both education and race in the high school market where learning appears to be mostly asymmetric. In the college market, employers directly observe most parts of the productivity of potential employees and learn very little over time.

In chapter three, we investigate how the process of employer learning and statistical discrimination varies over time and across employers. The comparison between the NLSY79 and the NLSY97 cohorts reveals that employer learning and statistical discrimination has became stronger over the past decades. Using the NLSY97 data, we identify three employer-specific characteristics that influencing employer learning and statistical discrimination, the supervisor-worker race match, supervisor's age, and firm size. Black high school graduates face weaker employer learning and statistical discrimination if they choose to work for a black supervisor, work for an old supervisor, or work in a firm of small size.

In the last chapter, we are interested in the associations between verbal and quantitative skills and individual earnings as well as the employer learning process of these two specific types of skills. There exist significant differences in both the labor market rewards and employer learning process of verbal and quantitative skills between high school and college graduates. Verbal skills are more important than quantitative skills for h igh school graduates, whereas college-educated workers benefit greatly from having high quantitative skills but little from having high verbal skills. In addition, employers directly learn verbal skills and continuously learn quantitative skills in the high school market, but almost perfectly observe quantitative skills in the college market.