Despite a substantial literature on the effects of education on women's fertility, little is known about possible variation in effects by selection into college. Women’s increasing educational attainment motivates further attention to the impact of education on fertility patterns, particularly among college‐educated women who have a low likelihood of attending and completing college. With data on U.S. women from the National Longitudinal Survey of Youth 1979, we examine effects of timely college attendance and completion by propensity score strata using a hierarchical linear model and stratum‐specific discrete‐time event‐history models. We find evidence for significant, systematic variation in effects. The fertility‐decreasing college effect is concentrated among disadvantaged women with a low propensity for college attendance and completion, approaches zero as the propensity for college increases, and then reverses to a fertility‐increasing effect among the most advantaged women.

Applications of classic fixed and random effects models for panel data are common in sociology and in ASR. A primary advantage of these models is the ability to control for time-invariant omitted variables that may bias observed relationships. This paper shows how to incorporate fixed and random effects models into structural equation models(SEMs)and how to extend the standard models to a wide variety of more flexible models. For instance, a researcher can test whether a covariate's impact on the repeated measure stays the same across all waves of data; test whether the error variances should be allowed to vary over time; include lagged covariates or lagged dependent variables; estimate the covariance of the latent time-invariant variables with the observed time varying covariates; and include observed time-invariant variables in a fixed effects model. The paper explains how to take advantage of the estimation, testing, and fit assessment capabilities that are readily available for SEMs and how to reveal flaws not visible with typical assessment techniques. The paper is oriented towards applied researchers with most technical details given in the appendix and footnotes.

Precarious employment—that is, jobs that entail a nonstandard contract, are short term, and/or do not provide fringe benefits like health insurance and retirement savings—has become a widely discussed topic in the media and a key research topic among scholars. Despite increasing scholarly and public interest in precarious employment, however, few studies have considered the effects of such employment beyond typical work and career outcomes. Using longitudinal panel data from the National Longitudinal Survey of Youth 1997, this study examines the effect of precarious employment on the social and political lives of contemporary young adults. The first chapter reviews the literature on the rise of precarious employment and the parallel phenomenon of delayed adulthood. The second chapter investigates the effects of precarious employment on parental co-residence and moving back home during young adulthood. I find evidence that nonstandard employment, short tenure, no employer-provided health insurance, and no employer-provided retirement benefits results in greater likelihood of living at home with parents. The third chapter analyzes the effects of precarious employment among young adults on having a first marriage. Here too, I find evidence that all four forms of precarious employment have negative effects on having a first marriage by the normative age. The fourth chapter assesses how precarious employment impacts political participation during young adulthood. Findings suggest that young adults that experience nonstandard employment are no less likely to be politically active than their peers with formal employment, but those that experience short tenure and no employer-provided benefits are less likely to vote, attend political meetings, and donate money to a cause. The fifth chapter discusses the theoretical and policy implications of these findings.

Abstract

This dissertation uses panel data to quantitatively assess the effects of college completion and elite college attendance on individual labor market outcomes during the Great Recession (2007-09). The effects of the Great Recession, the most protracted and severe economic downturn experienced in the U.S. since World War II, were felt unevenly across levels of educational attainment. After controlling for observable precollege variables such as cognitive ability, socioeconomic and demographic background, and high school experiences, substantial treatment effects of college completion remained during the Great Recession, though these were heterogeneous across the type of outcome and across individuals. Disadvantaged individuals benefitted the most from college completion on measures of employment, while more advantaged individuals benefitted greatest from college on measures of job quality. Furthermore, comparing effects of college among young workers who experienced expansionary economic contexts to those who experienced recessionary contexts showed that the patterns of effects described above were specific to recessionary contexts. Thus, experiencing a recessionary context led to an increase in the effect of college on employment among those least likely to complete college, and an increase in the effect of college on job quality for those most likely to complete college, conditional on employment. These results are consistent with the job competition model of the labor market, which utilizes a labor queue and predicts occupational downgrading at the top of the labor queue and crowding out of employment near the bottom of the labor queue. A similar hypothesis was not supported for the effects of elite college attendance during the Great Recession. The findings of this dissertation suggest that the economic context interacts with the effects of educational attainment on individual labor market outcomes in uneven ways, producing a unique constellation of education effects according to the economic context. Therefore, fluctuations in the business cycle can contribute to the stratification of individuals by affecting their labor market outcomes directly, but also by affecting the relationships between preexisting individual characteristics, educational attainment, and labor market outcomes.

This dissertation examines the ways that age shapes residential patterns in the United States. Sociologists have learned much about residential segregation since the founding of the discipline. However, few directly consider age as an element of segregation itself or as an important factor shaping racial segregation. I address these gaps by studying age as a dimension of residential segregation and by focusing on how the age structure in households, neighborhoods and the population is associated with segregation by age and by race. I use US Census data to study segregation in neighborhoods and metropolitan areas from 1940 to 2015. In the three chapters I explore segregation by age, longitudinal aging patterns in census tracts and the connections between age structure, neighborhood aging patterns and racial segregation. I find that age indeed structures residential patterns in the United States and is due greater attention in the sociological segregation literature and in public debates about generational change.

Sociology and social stratification researchers have long been concerned with the issue of educational inequality, as education is considered a key tool for achieving social mobility and has been shown to be a powerful determinant of an individual's opportunities and quality of life in modern society (Blau and Duncan 1967; Hout 1988; Fischer and Hout 2006). Given the significant role that education plays in determining social mobility, there has emerged a dynamic of competition between social classes for attaining higher education. While marginalized classes seek to overcome their disadvantaged status through education, thus allowing them greater social mobility, higher classes continuously enact social closure behavior through education to secure their advantaged status. Therefore, education is one of the most competitive endeavors in modern society. In this dissertation, I explore the class dynamics of acquiring educational advantages given educational expansion in the United States and South Korea (hereafter Korea). Of key interest are strategies of securing social mobility through education that entail a move away from public education and increased reliance on resources outside of formal schooling. In this dissertation, I examine the effect of three extra-curricular activities on education and labor market outcomes, with a particular focus on the transitional periods of entry into and exit from postsecondary education, along with patterns of student retention and attrition during the years of college enrollment.

Specifically, the first chapter of my dissertation uses data from the Educational Longitudinal Survey of 2002 to look at the average and heterogeneous effects of shadow education on SAT scores. The data include extensive information on not only students' socioeconomic background, but also high school information such as GPA, educational aspirations and motivation, which is helpful in estimating the propensity to shadow education and the relationship between shadow education and SAT scores. I use PSM to examine the average relationship between shadow education and SAT achievement, paying attention to the issue of pretreatment heterogeneity. Also, using the stratification-multilevel and smoothing-differencing methods, I examine how the effects of shadow education vary according to likelihood of receiving shadow education. I find that shadow education has significant positive effects on SAT scores and that the effects vary by individual propensity to use shadow education. The pattern of treatment effect heterogeneity indicates positive selection, which indicates that those who are more likely to use shadow education--those who are socioeconomically advantaged--benefit more from shadow education than those who are less likely to use it. Also, public resources such as school prep courses, books, and videos, neither alleviate the effects of private shadow education nor change the pattern of the treatment effect heterogeneity. My findings suggest that shadow education is an emerging mechanism that exacerbates educational inequality in the United States.

In the second chapter, using the National Longitudinal Survey of Youth from 1997, I examine the relationship between college student employment and dropouts. Since NLSY97 is surveyed annually and includes extensive information about students' educational backgrounds such as high school academic achievements, college financial aid, and the respondent's educational history, it is particularly useful to examine how student employment affects first year attrition and bachelor's degree completion. Using PSM, I estimate the average effects of treatment on the treated and I verify evidence of the treatment effect heterogeneity of student employment on college dropout by using the stratification-multilevel and smoothing-differencing methods. In this chapter, utilizing complex counterfactuals, (e.g., intense work [20 hours or more] vs. moderate work [less than 20 hours] vs. no work), I also examine variations in the effect of work intensity on dropout. In this study, I find that engaging in intense work has deleterious effects on first-year retention and on graduation within six years; however, the effects of intense work vary by likelihood of participation in intense work. The most advantaged students--who are least likely to engage in intense work--experience the most negative consequences from intense work, while such activity is less harmful to those from disadvantaged social backgrounds. I also find that this effect heterogeneity can be attributed to different financial situations and reasons for working between advantaged and disadvantaged students. This finding has two key implications. First, advantaged students should carefully consider engaging in intense work, as it can negatively affect bachelor's degree completion. Second, although the effect of intense work is less harmful for disadvantaged students, providing sufficient financial aid to them is still an important task, as this could help them to balance the intensity of work and school life.

The final chapter of my dissertation aims to examine the effects of English training abroad (hereafter ETA) on labor market outcomes in Korea. To examine how ETA affects employment and wages, I conduct survival analysis and quantile regression using data from the Korea Employment Information Service's 2007 Graduate Occupational Mobility Survey. The key finding of this study is that even though the average effects of ETA seem to be modest as most prior research has indicated, ETA does appear to have substantial positive effects on getting a good job and earning higher wages. ETA proved especially helpful for those who did not attend elite colleges. That is, ETA is a useful tool for non-elite students to supplement their weak formal education. Based on these findings, I conclude that ETA has a substantial impact on labor market outcomes in South Korea, and thus labor market opportunities are strongly determined by an socioeconomic background, as the cost of participation in ETA presents a barrier to entry for individuals from lower socioeconomic backgrounds.

Adolescents with high educational and occupational expectations are more likely to experience positive attainment outcomes in young adulthood. While most adolescents express high expectations, continued racial/ethnic disparities in attainment warrant renewed investigation of the expectation-attainment relationship. I examine racial/ethnic heterogeneity in the expectation-attainment relationship using nationally representative data from the Education Longitudinal Study of 2002 (ELS) and the High School Longitudinal Study of 2009 (HSLS). In Chapter 2 I examine whether the relationship between adolescent educational expectations and enrollment in a four-year college differs by race/ethnicity and whether this relationship changed over time. I find that the expectation-enrollment relationship is positive for all students, but it is smaller for black and Hispanic students in the ELS cohort. However, for the HSLS cohort, the gaps have largely closed. In Chapter 3 I use ELS to examine the relationship between adolescent educational expectations and college completion, whether this relationship varies by race/ethnicity, and how much of the relationship is direct versus mediated. I find a positive effect of high educational expectations on college completion for all groups. Thus, pre-college social background factors appear to account for much of the explanation as to why black and Hispanic students do not translate their expectations into completion. Yet, the mediating effects of on-time college enrollment are significantly larger for those who are black and Hispanic. Although black and Hispanic students are less likely to enroll in college, enrolling in college on-time helps link their expectations for college to college completion. In Chapter 4 I use ELS to examine the effects of having uncertain occupational expectations (relative to high) during adolescence on two educational outcomes. I find that occupational uncertainty during adolescence is negatively associated with degree attainment in young adulthood such that adolescents who report occupational uncertainty are less likely to attain bachelor degrees and less likely to attain advanced degrees or be enrolled in an advanced degree program by age 26 than their peers with high occupational expectations. Although white and Hispanic students were more likely to be uncertain, the effects of uncertainty on degree attainment were similar across racial/ethnic groups in the sample.