Boys,Girls,and the Second Shift: Paid and Unpaid Labor in High School and Adolescents’ Enrollment in College

Abstract

Sociological research examines the gender gap reversal in higher education and the gender division in paid and unpaid labor for adult women, especially “the second shift literature,” as two distinct topics. In this study, we extend the insights of the second shift literature to research on youth labor and adolescents’ enrollment in higher education. Using data from the Youth Development Study from 1988 to 1992, we find that the negative association of unpaid labor with adolescents’ college enrollment odds was at least as large as, if not greater than, that of paid labor. Although labor engagement had adverse impacts both for female and male adolescents during this time, the negative associations of youth labor with college enrollment were more pronounced for male students. We discuss the implications of these findings and explain their relevance to more contemporary cohorts of high school students in the conclusion.

Keywords

education sex and gender labor and labor movements

Over the past few decades, America witnessed increased gender egalitarian attitudes, rising college enrollment rates for women, and a staggering increase in women’s labor participation (Brooks and Bolzendahl 2004; Buchmann and DiPrete 2006; Cotter, Hermsen, and Vanneman 2011; Mason and Lu 1988). Despite these social changes, women have continued to shoulder the majority of household work, a phenomenon referred to as “the second shift” (Hochschild and Machung [1989] 2012). The elevated demand for adult women’s labor has direct implications for the labor performed by youth in the household: with mothers juggling work in and outside of the home, adolescent children—particularly female adolescents—have been under pressure to take on more household responsibilities (Goodnow 1988; Zelizer 1985). The labor division between adolescent males and females may affect their educational outcomes. In this study, we examine how male and female high school students’ engagement in paid and unpaid labor was associated with their enrollment in four-year colleges in the early 1990s.

Research extensively documents the effects of paid employment on a variety of educational and occupational achievements (Greenberger and Steinberg 1986; Marsh and Kleitman 2005; Staff and Mortimer 2007), but the few studies that directly examine gender-specific effects of youth labor yield inconsistent findings (Carr, Wright, and Brody 1996; Marsh and Kleitman 2005; Steel 1991). In this article, we contend that, just as unpaid labor performed by adult women at home has implications for the balance between work and home life, examinations of youth labor effects on male and female high school students’ college enrollment must integrate their household labor into the analysis.

To our knowledge, no study has explored gender differences in the effects of unpaid labor on adolescents’ educational outcomes, largely because most nationally representative surveys do not include measures of adolescents’ unpaid labor. We take advantage of data from the Youth Development Study (YDS), a survey that was administered to high school students after the gender gap reversal in higher education began to surface in the early 1980s. Using the YDS’s extensive measures of paid labor and homecare activities, we pose two central research questions:

Research Question 1: What were the associations of high school students’ paid and unpaid labor with college enrollment?

Research Question 2: How did youth labor associations with college enrollment differ between male and female students?

Our findings suggest that after taking students’ family background and school-related activities into consideration, unpaid labor performed during high school showed at least as large a negative association as paid labor with enrollment in a four-year college. While both forms of labor hindered college enrollment for males and females in 1992, these associations were larger for males.

In addition to the implications these findings have for understanding the gendered effects of labor, our study also contributes to research on the gender gap reversal in higher education. Although American male students have enjoyed considerable advantages in the educational and occupational arenas (Alon and Gelbgiser 2011; Blau and Kahn 1994; Bobbitt-Zeher 2007; Buchmann, DiPrete, and McDaniel 2008; Downey and Yuan 2005; Entwisle, Alexander, and Olson 1994; Feingold 1988), women reached parity with men in the number of bachelor’s degrees earned in 1982 (McDaniel 2010; Vincent-Lancrin 2008). Since then, females have continued to outpace their male counterparts in rates of college enrollment and completion (Buchmann and DiPrete 2006; Cho 2007; Cook 2006; DiPrete and Buchmann 2013; Freeman 2004; Goldin, Katz, and Kuziemko 2006). Our study considers the associations of youth labor with college enrollment in the early 1990s, a time when the gender reversal was becoming more prominent. Keeping in mind the historical contexts in which the second shift and gender gap reversal in higher education occurred and prevailed, our analysis aims to provide theoretical and empirical insights on youth labor effects on the gender disparities in higher education in contemporary society.

Background

Women comprised just 30 percent of college degree earners in 1950, but then exceeded men in college enrollment and degree completions in 1982. Today, women continue to make up the majority of the college student body. When explaining these gender differences in higher education, scholars generally agree that the spread of gender egalitarian views, women’s pursuit of careers traditionally held by men, degree benefits, and delayed family formation contributed to women’s increased educational attainment (Buchmann 2009; Buchmann and DiPrete 2006; Charles and Luoh 2003; Cho 2007; Goldin et al. 2006; Hubbard 2011). Missing from this discussion, however, is the effects of youth labor on male and female adolescents’ college enrollment.

Research on labor effects for educational outcomes focuses primarily on youths’ paid employment. Scholars indicate that employment during high school reduces time spent on homework, student effort and aspirations, the number of math and science courses taken in secondary school, and academic achievement (Greenberger and Steinberg 1986; Kablaoui and Pautler 1991; Lillydahl 1990; McNeil 1984; Mortimer and Finch 1986; Mortimer et al. 1996; Singh and Ozturk 2000; Steinberg and Dornbusch 1991). Some also suggest that labor performed during high school helps youths to build time management skills and, to some extent, their credentials for future educational and occupational development (Chaplin and Hannaway 1996; Stone and Mortimer 1998). Taken together, the reported evidence suggests that the negative effects of paid employment in high school on adolescents’ educational outcomes outweigh the positive (Marsh 1991; Marsh and Kleitman 2005).

Some studies that examine the influences of paid labor on college education also test whether the estimated effects differ across gender, yet the results of these studies lead to ostensibly inconsistent conclusions.¹ Marsh and Kleitman (2005) noted that the underlying arguments associated with the gender-specific effects are often underdeveloped. To more fully understand the gender-specific effects of youth labor on college enrollment, we suggest taking adolescents’ unpaid labor performed within the home into consideration. To the extent that time constraints or time management matters for school outcomes, it is important to note that the vast majority of American children also spend time on household work as part of their daily routine in addition to paid employment, schoolwork, and other extracurricular activities (Blair 1992). Equally important, adolescents’ household labor is highly gendered itself and closely tied to how male and female high schoolers allocate time to paid employment. Analyses that only include paid labor without unpaid labor thus may provide an incomplete picture of youth labor effects across gender. Several studies highlight the conceptual distinctions for youths’ paid and unpaid labor experiences (Marsh 1991; Mortimer 2003; Shanahan and Flaherty 2001; Zimmer-Gembeck and Mortimer 2006), but no empirical research directly investigates gender differences in the effects of unpaid labor on adolescents’ educational outcomes.

The second shift literature on working women provides a foundation to explore the gender-specific effects of youth paid and unpaid labor on college enrollment. Scholars estimate that working mothers average 21 hours per week completing household chores and 15 hours per week caring for their children, while working fathers only spend 15 hours per week on household chores and 9 hours caring for their children (Milkie, Raley, and Bianchi 2009:499). Cooking, cleaning, and other traditional “women’s work” continue to remain a part of women’s daily responsibilities, as opposed to yardwork, home improvement, and other traditionally masculine tasks (Milkie et al. 2009). Hochschild (1997) showed that, despite the disadvantages caused by their “second shift” duties and the gender-segregated labor market, a significant proportion of mothers dedicate their time and energy to paid work, rather than housework, because of the satisfaction and rewards that are reaped.

Like their mothers, female adolescents tend to engage in more household labor than male adolescents (Blair 1992; Gager, Cooney, and Call 1999; McHale et al. 1990; Raley and Bianchi 2006). Gager et al. (1999) found that on average, male and female youths perform about 15 and 17 hours of household labor per week in the ninth grade and nine and 13 hours per week in the 12th grade.² Blair (1992) suggested that these gender disparities begin as early as age five, and even when parents try to equalize home responsibilities among their children, sisters often spend more time on housework than their brothers (Raley and Bianchi 2006). Like their mothers, adolescent girls are also primarily responsible for feminine housekeeping and care work, like cooking or taking care of younger siblings, while adolescent boys engage in more masculine home improvement tasks, like lawn mowing (Gager et al. 1999). In addition to homecare, youths also engage in paid employment, a conventional practice for many American teens, especially for those saving up for college tuition. Research shows that adolescent girls’ participation rates in paid employment increased but lagged behind those of adolescent boys between 1970 and 1990, and have reached parity with those of adolescent boys since the 1990s (Fernandes-Alcantara 2012). However, adolescent females tend to earn lower wages than their male counterparts, primarily because girls are more likely to hold freelance jobs, such as babysitting, while boys tend to have more structured jobs with formal employers, regular hours, and set rates of pay (Besen-Cassino 2008; Herman 2000). These different experiences in paid and unpaid labor may facilitate distinct postsecondary pathways for adolescent males and females, motivate them to pursue occupational careers or college education after high school, and/or socialize them to balance school and home responsibilities in distinct ways.

Gender-specific Effects of Youth Labor

Our baseline hypothesis is that youth labor imposes similar constraints on the time that male and female students can allocate to school-related activities, thereby producing similar, undifferentiated negative effects on their likelihood of continuing on to higher education. Alternatively, the above comparison of adult women’s and adolescent females’ experiences in paid and unpaid labor directs our attention to at least three theoretical explanations for gender-specific labor effects on college enrollment. We summarize the predictions of these theories and their additional implications in Table 1. The predictions use causal language for ease of presentation. However, it should be noted that our survey data allow only for descriptive and associational conclusions—not the determination of causality.

Table 1.

Summary of Predicted Youth Labor Effects by Gender.

Main effectGendered pattern	Predicted effects on college enrollment				Additional implications
	Paid employment		Unpaid household labor
	Negative		Negative
	Male	Female	Male	Female
Gender segregation	Stronger negative effect	Weaker negative effect	Unchanged	Weaker negative effect	Boys may be more likely to hold paid formal jobs outside of home settings than girls leading to employment directly after high school. Girls may be more likely to do unpaid chores at home that can be multitasked with homework, allowing them to engage more in educational tasks.
Time bind	Unchanged	Weaker negative effect	Unchanged	Weaker negative effect
Gender socialization	Stronger negative effect	Weaker negative effect	Stronger negative effect	Weaker negative effect	Girls may be more able to manage time on different tasks, including paid labor, unpaid labor, and student activities.

First, gender segregation in both paid and unpaid labor may facilitate distinct postsecondary pathways for boys and girls. Regarding paid labor, “pull-out” theories argue that students’ decision to remain or leave school is based on a cost-benefit analysis. Because male students are more able to secure out-of-school employment, they are more likely to drop out of high school than female students (Bradley and Renzulli 2011; Jordan, Lara, and McPartland 1996). The same argument may apply to our analysis of college enrollment. Because male adolescents are more able than female adolescents to work in structured jobs with regular hours, their paid labor during high school is more likely to serve as a pathway to certifications, apprenticeships, or formal employment directly after high school, and not to college. This implies that the paid labor performed by male adolescents during high school may have a stronger negative effect on their college enrollment than the paid labor performed by female adolescents. Regarding unpaid labor, gender segregation in housework implies that girls may spend more time on household chores that are conducive to multitasking with academic work, such as watching younger siblings while doing their homework, whereas home improvement tasks typically performed by boys do not lend themselves as easily to balancing with schoolwork. If female adolescents’ household labor does not completely disengage them from schoolwork, unpaid labor may potentially have a less negative effect on female adolescents’ college enrollment than for males.

Second, in The Time Bind, Hochschild (1997) argued that working parents, despite their “family comes first” mantra, may actually prefer to work because of the rewards and sense of belonging conferred by engagement in paid employment. Housework, on the contrary, is often unrecognized and unrewarded. As such, the benefits of paid work become more appealing than housework. In a similar vein, we may suspect that female adolescents holding freelance jobs with low wages during high school may feel that those jobs provide little rewards and few occupational opportunities after high school. These dissatisfying experiences may drive adolescent girls away from low-wage jobs after high school and motivate them to pursue higher education credentials for better employment prospects. If so, paid labor experiences during high school may have a less negative effect on the college enrollment of female adolescents. Extending this argument, we may also reason that because females perform more unpaid labor on average, they may be more motivated to pursue work that is rewarding. This motivation may to some extent reduce the negative effects of unpaid labor on the college enrollment of female adolescents.

Finally, gender socialization may play a role in gender-specific youth labor effects on college enrollment. The gender socialization perspective argues that culturally inscribed norms linked with sex-based characteristics provide the foundation for expectations placed on males and females and how they act (Eccles, Jacobs, and Harold 1990). Scholars suggest that family is of large importance in explaining the gender roles boys and girls are socialized to perform. In many ways, girls learn to take on the behaviors displayed by their mothers or other adult women, and for boys, the behaviors of their fathers or other adult men. Children’s sex-typed interests and activities remain prevalent even among siblings from the same household (McHale, Crouter, and Tucker 1999; McHale, Crouter, and Whiteman 2003). As a result of gender socialization practices, girls spend more time on housework and also display greater engagement in schooling than boys.

Much of the second shift literature shows that as a result of taking on paid employment and more household labor at the same time, adult women spend more time multitasking, enabling them to complete more tasks in less time and to build skills for different roles that they assume (Bianchi, Robinson, and Milkie 2006; Sayer 2007). Extending this argument, if female children and adolescents are socialized to spend more time on both schoolwork and household labor, we may reason that these gendered socialization processes influence male and female students’ ability to balance schoolwork and youth labor throughout their educational careers. Learning to balance activities may potentially reduce the negative effects of youth labor on female students’ educational performance and allow them to pursue a college education despite higher demands for their labor. An additional implication of this explanation is that, to the extent that girls gain multitasking skills from gender socialization, they also may be more able to manage the time on other student activities.

For the purpose of our study, the explanations derived from the gender segregation, time bind, and gender socialization perspectives are not mutually exclusive; they complement each other to guide our exploration of the gender-specific effects of youth labor on adolescents’ college enrollment. All three perspectives suggest that paid and unpaid labor performed by adolescents during high school may have stronger negative effects on male students than on female students. Below, we examine the associations of paid and unpaid labor with college enrollment using the YDS data from the 1990s. We then discuss the implications of our findings to the continuing inequalities in youth labor and the gender gaps in higher education today.

Data and Method

We analyze data from the YDS, a longitudinal study that began in 1988 with a random sample of 1,139 ninth graders in the St. Paul School District in Minnesota. The YDS was designed to examine the working experiences of youths and how those experiences shape later work, health, and educational outcomes. Respondents were surveyed annually from the ninth to the 12th grade in high school and well into their adult years. In total, there are 19 waves of data. Our main analyses use the first five waves of the YDS to examine how paid and unpaid labor performed during all four years of high school (Waves 1–4) were linked to the respondents’ four-year college enrollment in the fall of 1992 (Wave 5) following their completion of high school.

The base-year YDS survey in 1988–1989 included an oversample of 129 Hmong immigrant youths. These respondents are excluded from our analyses because a different interview procedure was used to accommodate the adolescents’ language barriers. Of the remaining 1,010 respondents, we excluded 18 cases with outlier values for unpaid labor (average unpaid labor hours per day > 20), reducing our sample size to 992. After eliminating missing cases in the dependent variable and using multiple imputations (m = 20) for missing values in the explanatory variables to preserve cases (Little and Rubin 2002), our analytical sample consists of 914 respondents.

The YDS has two strengths for our analysis. First, the YDS captured the crucial period when the gender gap reversal in higher education began to surface and continued to expand in the 1980s. This allows us to investigate the dynamics between youth labor, entrance into college, and the differences among males and females who experienced the widening gap in college enrollment early stages in its development. Second, the YDS contains the most detailed measures for adolescents’ unpaid labor in addition to their paid labor. While several nationally representative datasets contain measures for youths’ paid employment, no surveys administered around the same time contain such detailed information about adolescents’ household labor. In the online supplement, we compare the YDS with three nationally representative datasets from 1988 to 2009. The results confirm that the findings from the YDS are reasonably generalizable to the 1988 high school cohort in the nation.

Measures

College enrollment

To examine the associations of youth labor in high school with college enrollment, we measure college enrollment during the academic year following high school graduation. Enrollment is the first step toward attaining a college degree, a good predictor of future earnings, and other social outcomes (Pew Research Center 2011; Settersten and Ray 2010). We use four-year college enrollment as the dependent variable because much of the literature on gender differences in higher education focuses on four-year institutions (DiPrete and Buchmann 2013).³ This variable is coded as 1 if a student attended a four-year college in the fall of 1992 and 0 otherwise.⁴

As noted, our initial sample consisted of 992 respondents. Of these, 396 have missing values in the college enrollment measure in the 1992 survey. We recoded 204 students who indicated that they were not enrolled in any school in 1992 as not attending college. We then used the reported highest education attainment in Waves 10 through 19 to further impute the missing values in the dependent variable, with those who reported an educational credential of associate degree or less in 1998–2011 coded as 0 (not attending a four-year college in 1992).⁵ In all, these procedures reduced the number of missing cases in the dependent variable from 396 to 78.⁶

Youth labor

Our key independent variables are paid and unpaid labor. Paid labor is measured as the average hours per day a student spent on paid employment. We compute this average across four years of high school to examine how paid labor performed over the span of students’ secondary education combined to impact college enrollment in the year following their high school completion. As noted, the YDS includes detailed information about adolescents’ unpaid labor. To measure the average hours per day a high school student spent on housework and other unpaid labor, we summed the time respondents spent on the available measures of unpaid labor in the YDS, including house cleaning, doing dishes, setting the table, cooking, taking care of younger children and/or elderly relatives, yardwork, laundry, taking out the trash, shopping for food, taking care of pets, and other family-related tasks.⁷ In supplementary analyses, we also test the separated effects of these tasks. The results are consistent in direction with the analysis using the combined measure, but none of the coefficients for any single task are significant. This suggests that time spent on all of these tasks together adds up and is consequently associated with college enrollment odds.

Control variables

We include three groups of control variables. First, to ensure that the estimated relationships between youth labor and college enrollment are not a function of background characteristics, we control gender, race (black, Asian, Hispanic, and other race; reference group = white), students’ age, highest parental education, annual family income, parental structure (single parent, other-two-parent, and other family arrangements; reference group = two biological parents), and total household size in multivariate analyses. Second, we control student activities in our analysis because they constrain the time that students may allocate to youth labor and schoolwork (Shanahan and Flaherty 2001). These are measured as the average hours per day students spent on extracurricular activities, volunteering, and homework. Finally, we control students’ grade point average (GPA) in the final year of high school, educational aspirations, occupational aspirations, marriage aspirations, and the number of children that the students expect to have in the future (see Shanahan and Flaherty 2001).

Analytic Strategy

We use logistic regression to analyze the associations of youth labor with college enrollment. Our analyses proceed in three stages. First, we examine the relationship between paid and unpaid labor and college enrollment without controlling for other factors. Next, we add three sets of control variables in sequence to examine how the relationships between youth labor and college enrollment are mediated by student background characteristics, engagement in other activities, academic ability, and future orientations. Finally, we run analyses including the interactions of gender and the two youth labor measures to examine the links of paid and unpaid labor to boys and girls. To better illustrate the relationships between youth labor and college enrollment, we calculate the predicted probabilities of college enrollment for male and female students and report the results in a graphical presentation.

Results

Table 2 displays the total, male, and female means for each variable included in the analysis. Consistent with the national trends in the early 1990s (American College Testing Program 2013), female high school students in St. Paul showed a slightly higher college enrollment rate than male students, but the difference was not substantial nor significant. While high school students in general spent more time on unpaid labor than paid labor, females’ time on both forms of labor were significantly higher than those of males. At 1.82 hours per day on average, females spent nearly one-third of an hour more on paid labor per day than males, and at 2.58 hours per day, female students spent over one half hour more on unpaid labor than males. These gender differences resemble patterns documented in previous studies (Blair 1992; Gager et al. 1999; Raley and Bianchi 2006). In additional analyses reported in the online supplement, we find that adolescent boys were more likely than adolescent girls to hold paid formal jobs outside of home setting (e.g., business, store, and farm) as opposed to home settings (e.g., babysitting). Compared with male adolescents, female adolescents also spent more time on household chores that allowed them to multitask and on tasks that may be considered “more feminine” (McHale et al. 1990). These patterns lend support to the gender segregation explanation.

Table 2.

Descriptive Statistics and Gender Differences in Means.

Variables	Total sample				Male	Female
Variables	Mean	Standard deviation	Minimum	Maximum	Mean	Mean
Dependent variables
College enrollment	0.31	0.46	0.00	1.00	0.29	0.32
Key independent variables
Paid labor hours per day	1.67	1.36	0.00	8.09	1.49**	1.82**
Unpaid labor hours per day	2.31	1.74	0.00	13.14	1.99**	2.58**
Student and family characteristics
Female	0.55	0.50	0.00	1.00	—	—
White^a	0.75	0.43	0.00	1.00	0.75	0.75
Black	0.09	0.29	0.00	1.00	0.08	0.10
Hispanic	0.05	0.21	0.00	1.00	0.05	0.05
Asian	0.03	0.18	0.00	1.00	0.04	0.03
Other race	0.08	0.27	0.00	1.00	0.08	0.07
Age in years	15.43	0.60	15.00	19.00	15.51**	15.36**
Parental education in years	13.14	2.18	10.00	21.00	13.15	13.13
Parental income in US$1,000	59.97	43.02	2.50	210.00	58.02	61.60
Two biological parents^a	0.56	0.50	0.00	1.00	0.58	0.54
Two other parents	0.13	0.34	0.00	1.00	0.12	0.14
Single parent	0.25	0.43	0.00	1.00	0.24	0.25
Other family types	0.06	0.25	0.00	1.00	0.06	0.06
Total household size	3.98	1.44	1.00	10.00	3.90	4.04
Student activities
Extracurricular time in hours per day	1.09	1.03	0.00	10.29	1.30**	0.91**
Volunteer time in hours per day	0.11	0.25	0.00	2.48	0.11	0.11
Homework time in hours per day	1.28	1.05	0.00	8.14	1.10**	1.44**
Academic ability and future orientation
GPA on a 4.0 scale	2.74	0.77	0.33	4.00	2.63**	2.83**
Educational aspiration in years	16.96	2.98	10.00	21.00	16.57**	17.29**
Occupational aspirations^b	3.59	0.60	1.00	4.00	3.58	3.60
Marriage aspirations^c	3.21	0.88	1.00	4.00	3.11**	3.30**
Number of children expect to have	2.05	1.54	0.00	20.00	1.91*	2.16*
N	914				415	499

Note. Means and standard deviations are from imputed datasets m = 20. Test significance indicates differences between male and female students. GPA = grade point average.

Whites and two biological parents are the reference group.

How important student thinks occupation/career for their future: 1 = not at all important, 4 = extremely important.

How important student thinks marriage for their future: 1 = not at all important, 4 = extremely important.

p < .05. **p < .01 (two-tailed).

On average, male students were slightly older than female students. This is likely explained by male students’ higher grade retention rates from elementary to high school (DiPrete and Buchmann 2013). Although girls spent less time on extracurricular activities, they spent significantly more time on homework, had higher GPAs, and had higher educational aspirations than their male counterparts. Girls also ranked marriage as more important to their future and expected to have more children. Coupled with the finding that female students spent more time on paid and unpaid labor, these results may suggest that in 1992, females were more likely to juggle a number of responsibilities than males overall.

Paid and Unpaid Labor Effects

Table 3 reports the logistic regression analysis for the relationships between youth labor and college enrollment. Model 1 shows that an additional hour of paid labor per day was associated with a decrease in the odds of enrolling in a four-year college by a ratio of .76 (e^−.269 = .76, p < .01), or 24 percent. Similarly, one hour of unpaid labor per day was associated with a 27 percent decrease in the odds of enrolling in a four-year college. These findings lend support to our argument that explorations of youth labor effects must integrate unpaid labor into the analysis. Upon the inclusion of student characteristics in Model 2, the coefficients decrease in size but remain statistically significant. This suggests that the negative associations of youth labor with college enrollment were partially mediated by students’ sociodemographic backgrounds. Females, Asian Americans, parental education, and family income had significantly positive associations with odds of college enrollment.

Table 3.

Logistic Regression Coefficients for Four-Year College Enrollment on Paid and Unpaid Labor (N = 914).

Variables	Model 1		Model 2		Model 3		Model 4
Paid labor hours	−0.269	(0.061)**	−0.237	(0.072)**	−0.225	(0.076)**	−0.183	(0.089)*
Unpaid labor hours	−0.306	(0.055)**	−0.249	(0.063)**	−0.320	(0.070)**	−0.278	(0.076)**
Student and family characteristics
Female			0.384	(0.176)*	0.361	(0.189)	−0.027	(0.223)
Black			−0.197	(0.317)	−0.348	(0.336)	0.170	(0.374)
Hispanic			−0.188	(0.415)	−0.493	(0.444)	−0.126	(0.536)
Asian			1.894	(0.479)**	1.169	(0.511)*	1.156	(0.590)
Other race			−0.726	(0.409)	−0.785	(0.413)	−0.282	(0.466)
Age			−0.219	(0.158)	−0.262	(0.163)	−0.195	(0.189)
Parental education			0.320	(0.044)**	0.293	(0.046)**	0.277	(0.052)**
Parental income			0.010	(0.003)**	0.011	(0.003)**	0.006	(0.003)
Two other parents			−0.379	(0.266)	−0.322	(0.274)	−0.231	(0.313)
Single parent			−0.067	(0.283)	0.036	(0.296)	−0.083	(0.342)
Other family types			0.276	(0.423)	0.290	(0.452)	0.135	(0.534)
Total household size			.0124	(0.073)	0.122	(0.077)	0.067	(0.088)
Student activities
Extracurricular time					0.266	(0.089)**	0.287	(0.110)**
Volunteer time					−0.370	(0.404)	−0.333	(0.445)
Homework time					0.588	(0.096)**	0.341	(0.104)**
Academic ability and future orientation
GPA							1.590	(0.188)**
Educational aspirations							0.178	(0.042)**
Occupational aspirations							0.120	(0.183)
Marriage aspirations							0.373	(0.128)**
Number of children expect to have							−0.021	(0.068)
Constant	0.253	(0.156)	−2.179	(2.622)	−2.060	(2.652)	−11.403	(3.295)**
Pseudo-R²	.057		.215		.264		.402
BIC	1,084.100		987.238		952.583		831.519

Note. Pseudo-R² and BIC are from imputed dataset m = 1. Omitted categories are white students and two biological parents. Standard errors are in parentheses. GPA = grade point average; BIC = Bayesian information criterion.

p < .05. **p < .01 (two-tailed).

Model 3 further controls the time that students allotted to extracurricular activities, volunteering, and homework. Research finds that high school students’ time on paid labor reduces the time they have for homework (Lillydahl 1990; Marsh and Kleitman 2005).The same logic also may apply to extracurricular activities and volunteering. As shown in Table 3, because the inclusion of these three student activities may explain some variation in students’ paid labor, the negative association of paid labor hours with college enrollment decreased in magnitude. For unpaid labor, however, we find that when students’ time on other school-related activities is controlled, unpaid labor became more negatively associated with college enrollment. This pattern may appear if high school students’ unpaid labor is largely due to family demands, regardless of the time they may have spent on other student activities (see Blair 1992). In other words, whereas students’ time on paid employment depended on school-related activities, their unpaid labor did not. As a result, unpaid labor diverted students’ attention and energy from schoolwork more than paid labor. Thus, after controlling students’ engagement in extracurricular activities and time on homework, the link between unpaid labor and college enrollment was larger than that of paid labor, though the difference is not statistically significant.

Finally, Model 4 adds students’ GPAs and future orientations. As expected, students’ academic performance in high school had a strong and large effect on entrance into college. Educational and marriage aspirations also shared a significantly positive relationship with college enrollment.⁸ These findings suggest that students who performed well in school, had high educational aspirations, and considered marital commitment to be an important part of their future were more likely to attend college.

Arguably, because youth labor may be linked to college enrollment odds by affecting students’ academic performance in high school, the inclusion of academic achievement in Model 4 may lead to an underestimate of paid and unpaid labor associations. Bearing this in mind, we find that, independent of students’ characteristics, time devoted to other activities, academic achievements, and future orientations, one hour of paid and unpaid labor, respectively, decreased the odds of going to a four-year college by 17 (e^−.183 = .83, p < .05) and 25 percent. This suggests that, in addition to the relationship with college enrollment through academic performance in high school, students’ time on paid and unpaid labor was also negatively associated with their odds of college enrollment through other mechanisms. Such mechanisms may include, but are not limited to, less time and attention preparing for standardized tests, distraction from educational endeavors, or limits on their ability to acquire information and skills for college applications. Taken together, although the coefficients of paid and unpaid labor are not statistically different, the consistent pattern of different effect sizes suggests that the often-overlooked unpaid labor may have had at least as large, if not greater, associations with high school students’ college pursuits than the well-documented paid labor effects.⁹

Gender-specific Analyses of Youth Labor

Table 4 reports the regression analysis of the gender-specific associations of youth labor with college enrollment. Model 1 shows that, without other control variables, one hour of paid labor reduced male students’ college enrollment odds by 35 percent (p < .01), whereas for females, paid labor reduced the odds of enrolling by 16 percent (p < .05). The gender difference in the associations of paid labor with higher education observed here (p < .10) is consistent with the findings of Steel (1991) and, to some extent, Carr et al. (1996).

Table 4.

Logistic Regression Coefficients for Four-Year College Enrollment on Labor and Gender Interaction (N = 914).

Variables	Model 1		Model 2		Model 3		Model 4
Paid Labor Hours × Male	−0.422	(0.099)**	−0.330	(0.113)**	−0.340	(0.120)**	−0.303	(0.142)*
Paid Labor Hours × Female	−0.198	(0.082)*	−0.166	(0.095)	−0.141	(0.099)	−0.094	(0.118)
Unpaid Labor Hours × Male	−0.362	(0.100)**	−0.303	(0.114)**	−0.424	(0.129)**	−0.502	(0.152)**
Unpaid Labor Hours × Female	−0.329	(0.070)**	−0.226	(0.075)**	−0.281	(0.081)**	−0.201	(0.087)*
Student and family characteristics
Female	0.107	(0.322)	0.014	(0.357)	−0.164	(0.378)	−0.873	(0.456)
Black			−0.189	(0.316)	−0.331	(0.335)	0.198	(0.374)
Hispanic			−0.184	(0.416)	−0.470	(0.443)	−0.059	(0.532)
Asian			1.881	(0.481)**	1.173	(0.514)*	1.238	(0.598)*
Other race			−0.713	(0.408)	−0.762	(0.412)	−0.243	(0.466)
Age			−0.211	(0.158)	−0.254	(0.163)	−0.180	(0.190)
Parental education			0.321	(0.044)**	0.295	(0.046)**	0.280	(0.052)**
Parental income			0.010	(0.003)**	0.010	(0.003)**	0.006	(0.003)*
Two other parents			−0.371	(0.266)	−0.309	(0.275)	−0.210	(0.315)
Single parent			−0.076	(0.283)	0.029	(0.295)	−0.112	(0.344)
Other family types			0.301	(0.424)	0.317	(0.454)	0.173	(0.543)
Total household size			0.123	(0.073)	0.121	(0.077)	0.057	(0.088)
Student activities
Extracurricular time					0.279	(0.090)**	0.308	(0.111)**
Volunteer time					−0.371	(0.404)	−0.351	(0.451)
Homework time					0.590	(0.096)**	0.342	(0.104)**
Academic ability and future orientation
GPA							1.599	(0.189)**
Educational aspirations							0.182	(0.042)**
Occupational aspirations							0.123	(0.183)
Marriage aspirations							0.373	(0.129)**
Number of children expect to have							−0.021	(0.068)
Constant	0.286	(0.231)	−2.095	(2.568)	−1.929	(2.663)	−11.237	(3.320)**
Pseudo-R²	.068		.217		.266		.406
BIC	1,091.880		999.515		963.818		840.877

p < .05. **p < .01 (two-tailed).

Comparing the respective associations of paid and unpaid labor with college enrollment further contextualizes the impact of youth labor on higher education for male and female high school students. Model 1 shows that unpaid labor had a negative relationship with both male and female students’ college enrollment. The consistent, negative coefficients of paid and unpaid labor suggest that juggling work and family obligations may have limited the time students could devote to school. Adding student characteristics in Model 2 reduces the size of the youth labor associations for both males and females, but the gender differences in the paid labor coefficients are noticeable: for males, one hour of paid labor reduced the odds of enrollment by 28 percent (p < .01); for females, the association was insignificant. Unpaid labor also reduced male students’ college enrollment odds by about 26 percent. For female students, one hour of unpaid labor reduced the odds of college enrollment by 20 percent.

Models 3 and 4 add student activities, academic ability, and future orientations. As in Model 2, the coefficients of paid labor are still large and significant for male students, but small and insignificant for female students. For male students, adding these controls increases the negative associations of unpaid labor with college enrollment. For female students, however, doing so reduces the magnitudes and statistical significance of unpaid labor associations in Model 4 (p < .10 for gender differences; see the online supplement for details). As noted earlier, because students’ academic performance in high school was also influenced by paid and unpaid labor, the estimated labor associations with college enrollment in Model 4 are likely conservative. Together, these analyses again suggest that, controlling for student characteristics, time allotted to student activities in high school, and future orientations, unpaid labor had at least as a strong negative relationship with teenagers’ college enrollment as paid labor. The gendered patterns in these relationships are consistent with those implied by gender segregation, time bind, and gender socialization.

As suggested earlier, multitasking by adolescent girls may be a result of gender segregation in unpaid labor and/or gender socialization. While these explanations are not mutually exclusive, we may test the additional implication by gender socialization that girls also may be more able to manage the time on other student activities. In supplementary analyses, we include the interactions of gender and the three measures of student activities in Models 3 and 4. The results do not alter our main findings about the gender-specific youth labor associations, and further suggest that before controlling for students’ GPAs and aspirations, teenagers’ time on student activities had either more positive associations with females (i.e., extracurricular activities) or more negative associations with males (i.e., volunteer activities), whereas the positive association of homework time was visibly greater with males’ than with females’ college enrollment. In the full model, time on extracurricular activities shows similar positive associations across gender, but time on volunteer activities was significantly more negatively related to males than to females. These results are consistent with the gender socialization explanation that female students were more able to manage the time on different tasks, whether these tasks are paid labor, unpaid labor, or student activities.

Overall, the findings in Table 4 suggest that engagement in youth labor may have differing consequences for boys and girls. Recall from Table 2 that female students spent more time on unpaid labor than male students. While the gender-specific patterns found in Model 4—that the paid labor association with females was statistically insignificant and the unpaid labor association was smaller with female than with male students—lend support to the argument that youth labor was in part imbued with gendered meanings, one must also consider the different amounts of time male and female adolescents allocated to paid and unpaid labor. To further explore the influences of males’ and females’ youth labor, in Figure 1, we calculate the predicted probabilities students enrolled in a four-year college. Panel A shows the predicted probabilities of college enrollment from Model 1 in Table 4, which includes no controls. Panel B shows the predicted probabilities from the full model. The markers indicate the male and female sample average labor hours. Gender differences in predicted probabilities are tested with 95 percent confidence intervals.

Figure 1.

Predicted probabilities of college enrollment by youth labor hours and gender, with 95 percent confidence intervals for gender differences.

Panel A shows that, when performing zero hours of paid labor (and with unpaid labor held at the sample mean), female students had a slightly higher probability of going to college (approximately .41) than male students (approximately .36), but the difference was not statistically significant (p > .05). The same pattern applies when unpaid labor is held at zero hours and paid labor held at the sample mean. However, these gender differences became statistically significant, and maintained or widened in magnitudes as the number of youth labor hours increased. In all, Panel A suggests that, without considering other factors, female students in 1992 had an advantage over male students in college enrollment, and the patterns of influence were consistent across paid and unpaid labor.

The observed pattern above changes in Panel B. Most notably, the female advantage in college enrollment becomes insignificant after accounting for family characteristics, student activities, academic performance, and future orientations. Instead, we see that when engaging in no youth labor, males had a slightly higher probability of going to college than females. Increasing hours of labor engagement reversed this male advantage. At about 0.70 hours of paid labor per day, the predicted probabilities were the same for males and females. Although not statistically significant, the pattern of gender differences suggests that as more hours of paid labor were accumulated beyond this point, male students may have begun to have lower probabilities in college enrollment than females. The same pattern applied to the relationship between unpaid labor hours and college enrollment. For males, the predicted probabilities for college enrollment decreased drastically from .40 at zero hours of unpaid labor to .08 at four hours of unpaid labor, a total loss of .32. For females, this decrease was much smaller—a total loss of .15. Thus, females were more adaptable to the potential negative effects of labor for higher education compared with their male counterparts. At 1.70 hours of unpaid labor, males and females experienced almost the same probability of enrolling in college, after which more hours of unpaid labor may have disadvantaged males.

Finally, Panel B shows that when male and female students matched their own group-average paid labor hours, that is, the average hours females engaged in paid labor and the average hours males engaged in paid labor (with unpaid labor hours held at the total sample mean), females had a slightly higher probability of college enrollment than males in the early 1990s. Yet, when looking at the graph for unpaid labor, because female students engaged in a larger amount of labor on average, their predicted probability for college enrollment at the female mean for unpaid labor hours was slightly lower than the male probability.¹⁰ These analyses suggest that although the negative associations of unpaid labor with college enrollment were smaller for females than for males, because females engaged in more unpaid labor on average, the accumulation of negative unpaid labor associations still may have disadvantaged females’ college enrollment in 1992.

Discussion and Conclusion

The emergence of the gender gap reversal in higher education since the 1980s has solicited a great deal of scholarly attention. In this study, we bring insights from the second shift literature to examine the impacts of youth labor in high schools on male and female students’ college enrollment. Our analyses of the 1988–1992 data from the YDS suggest that adolescents’ unpaid household labor may have had a similar or greater negative association with college enrollment odds as do their paid employment. The negative impacts of youth labor on college enrollment were more pronounced for male students than for female students.

Our finding that adolescents’ unpaid labor may be strongly linked to college enrollment—especially for adolescent males—underscores its significance for research investigating the influences of youth labor on educational outcomes. Building on the theoretical discussions of gender segregation, the time bind, and gender socialization, we demonstrate how socially inscribed norms about the type of youth labor girls and boys are expected to perform may have consequences for their educational transitions in the long term. That is, boys and girls, much like their parents, are subject to different expectations about the kinds of work they do and how much time they should dedicate to such work, both in the home and in the labor market. These differences can alter the time they devote to or balance with educational activities and other tasks, and may even influence their decisions regarding future educational and occupational goals. Together, these theories help inform our understanding of the gender-specific associations of paid and unpaid labor with college enrollment and the gender gap reversal in higher education, a trend that became more prominent during a time of changing gender divisions of labor inside and outside of the home.

To explain how youth labor may be differentially linked to girls’ and boys’ educational outcomes, both the gender segregation and gender socialization perspectives imply a multitasking effect. According to the gender segregation argument, the separation of male and female students into distinct types of paid and unpaid labor tasks could allow female adolescents to spend more time multitasking by mixing their labor with academic work. Using the gender socialization argument, female adolescents are socialized to better balance homecare responsibilities and school performance and thus may develop stronger multitasking skills than male adolescents over time. For the purpose of our study, both mechanisms may be at work. Indeed, often they are co-occurring and mutually reinforcing the gendered patterns of the youth labor effects. To the extent that adolescent females are more able to multitask, not simply because the gendered structure of youth labor affords them to do so, however, we should expect that female students are also more able to balance other student activities. As noted, our supplementary analyses suggest that female students were more able to manage the time spent on different tasks. This finding is consistent with a gender socialization explanation concerning the gender-specific effects of youth labor.

While our study is rooted in a historical context in which the second shift and gender gap reversal in higher education occurred and prevailed, an important question is how our analyses might relate to today’s prospective college students. In additional analyses reported in the online supplement, we compare the YDS with National Longitudinal Study of Adolescent Health (Add Health:1994/95) and High School Longitudinal Study of 2009 (HSLS:2009). The results suggest that the college enrollment rates since the early 1990s have slowly approached 50 percent while the gender gap continued to widen, and that adolescents’ time on paid employment has decreased over time. No nationally representative data include detailed measures of unpaid labor, but scholars note an increase in housework time among American youth from 1980 to the 2000s (Juster, Ono, and Stafford 2004). These trends suggest more variation in college enrollment as an outcome, greater disparities in enrollment by gender, and an increasing importance of unpaid labor relative to paid employment in explaining these variations. Additional regression analysis using the HSLS:2009 indicates significant associations of paid employment with college enrollment in the early 2010s. Taken together, we conclude that youth labor may continue to relate to male and female students’ college enrollment odds. We also note that increasing social inequalities in the United States are likely to place greater constraints on adolescents from disadvantaged families that are especially burdened for unpaid household labor and in need of paid employment. Subsequently, this may make youth labor a more negative factor on college enrollment for disadvantaged students than for their privileged counterparts, a topic scholars may examine in future research.

Our findings should not be interpreted to suggest that any engagement in paid employment and household chores by adolescents would reduce their college enrollment odds. Rather, our analyses simply suggest that high school students are squeezed between paid employment, household labor, student activities, and schoolwork. We acknowledge that some exposure to household work may help develop a sense of responsibility and thus benefit adolescents in their future development. Consistent with this view, in supplementary analyses, we find an initial positive effect from less than 20 hours of paid labor per week and a negative effect from more than 20 hours of paid labor for male students. However, the positive effect disappears after controlling for students’ characteristics, school activities, and academic achievements. This suggests that the exposure effect of youth labor is limited for boys and also mediated by other factors. In our study, the negative association between youth labor and college enrollment is the dominant pattern, which makes it difficult to discern the potential exposure effects of appropriate household work for these adolescents.

As in other studies using survey sampling designs, our analysis may be affected by potential selection bias. For example, students in the YDS sample may have self-selected no engagement in paid or unpaid labor. Some of these students may have benefited from some paid and unpaid labor by taking on responsibilities and learning time management skills, and thus increased their chances of enrolling in college. The inclusion of these students may attenuate the negative associations of paid and unpaid labor with college enrollment because the positive youth labor associations cancel out the negative. This may lead to conservative estimates of negative youth labor associations in our analysis. It is also possible that any depletion of students’ time on schoolwork might negatively affect the odds of their college enrollment. If so, the self-selection of those who perform no youth labor may lead to an overestimate of the paid and unpaid labor associations reported in this article. Finally, students who did not engage in any paid and unpaid labor may have come from well-off families, and were thus likely to go to college regardless of whether or not they had engaged in youth labor. Although our analysis includes controls of family characteristics, the possibility of potentially omitted variables governing this relationship cannot be completely ruled out. Without precise measures to estimate the potential selection bias, we urge caution in interpreting the statistical effects observed in our regression analyses restrictively as direct causal influences of youth labor on college enrollment. Generalization of the patterns from our analyses to individual cases with vastly different personal and family situations is also unwarranted.

In conclusion, our study highlights the complexity of youth labor effects while also expanding on the theoretical mechanisms that may reduce the negative effect of labor on college enrollment for females. We acknowledge that gender is only one of the sociodemographic factors that may interact with youth labor to produce inequalities in higher education. Other status factors, such as socioeconomic background, may also affect how engagement in labor influences high school students’ educational outcomes. While these questions are beyond the scope of this study, our results suggest that the relationship between labor and education is more nuanced than previously conceived. Our analyses provide a foundation for future explorations of how different forms of labor and their gendered connotations are linked with the widening gender gap in college enrollment.

Footnotes

Acknowledgements

We would like to thank Brian Powell, Michael Wallace, and Mary Fischer for their helpful feedback on this article.

Authors’ Note

All authors contributed equally. They are listed in reverse alphabetical order.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Supplemental Material

Online Supplement for this article is available online.

References

Alon

Sigal

Gelbgiser

Dafna

. 2011. “The Female Advantage in College Academic Achievements and Horizontal Sex Segregation.” Social Science Research 40(1):107–19.

American College Testing Program. 2013. “Recent High School Completers and Their Enrollment in 2-year and 4-year Colleges, by Sex: 1960 through 2012.” Data derived from the Census Bureau, Current Population Survey, 1970 through 2012. Unpublished Tabulations.

Besen-Cassino

Yasemin

. 2008. “The Cost of Being a Girl: Gender Earning Differentials in the Early Labor Markets.” NWSA Journal 20(1):146–60.

Bianchi

Suzanne M.

Robinson

John P.

Milkie

Melissa A.

2006. Changing Rhythms of American Family Life. New York: Russell Sage Foundation.

Blair

Sampson L.

1992. “Children’s Participation in Household Labor: Child Socialization versus the Need for Household Labor.” Journal of Youth and Adolescence 21(2):241–58.

Blau

Francine D.

Kahn

Lawrence M.

1994. “Rising Wage Inequality and the U.S. Gender Gap.” The American Economic Review 84(2):23–28.

Bobbitt-Zeher

Donna

. 2007. “The Gender Income Gap and the Role of Education.” Sociology of Education 80(1):1–22.

Bradley

Christen L.

Renzulli

Linda A.

2011. “The Complexity of Non-completion: Being Pushed or Pulled to Drop Out of High School.” Social Forces 90(2):521–45.

Brooks

Clem

Bolzendahl

Catherine

. 2004. “The Transformation of U.S. Gender Role Attitudes: Cohort Replacement, Social-structural Change, and Ideological Learning.” Social Science Research 33(1):106–33.

10.

Buchmann

Claudia

. 2009. “Gender Inequalities in the Transition to College.” Teachers College Record 111(10):2320–46.

11.

Buchmann

Claudia

DiPrete

Thomas A.

2006. “The Growing Female Advantage in College Completion: The Role of Family Background and Academic Achievement.” American Sociological Review 71(4):515–41.

12.

Buchmann

Claudia

DiPrete

Thomas A.

McDaniel

Anne

. 2008. “Gender Inequalities in Education.” Annual Review of Sociology 34:319–37.

13.

Carr

Rhonda V.

Wright

James D.

Brody

Charles J.

1996. “Effects of High School Work Experience a Decade Later: Evidence from the National Longitudinal Survey.” Sociology of Education 69(1):66–81.

14.

Chaplin

Duncan

Hannaway

Jane

. 1996. “High School Employment: Meaningful Connections for At-risk Youth.” Urban Institute. Unpublished Manuscript.

15.

Charles

Kerwin K.

Luoh

Ming-Ching

. 2003. “Gender Differences in Completed Schooling.” The Review of Economics and Statistics 85(3):559–77.

16.

Cho

Donghun

. 2007. “The Role of High School Performance in Explaining Women’s Rising College Enrollment.” Economics of Education Review 26(4):450–62.

17.

Cook

Genn

. 2006. “Boys at Risk: The Gender Achievement Gap.” The American School Board Journal 193(4):4–6.

18.

Cotter

David

Hermsen

Joan M.

Vanneman

Reeve

. 2011. “The End of the Gender Revolution? Gender Role Attitudes from 1977 to 2008.” American Journal of Sociology 117(1):259–89.

19.

DiPrete

Thomas A.

Buchmann

Claudia

. 2013. The Rise of Women: The Growing Gender Gap in Education and What It Means for American Schools. New York: Russell Sage Foundation.

20.

Downey

Douglas B.

Yuan

Anastasia S. Vogt

. 2005. “Sex Differences in School Performance during High School: Puzzling Patterns and Possible Explanations.” The Sociological Quarterly 46(2):299–321.

21.

Eccles

Jacquelynne S.

Jacobs

Janis E.

Harold

Rena D.

1990. “Gender Role Stereotypes, Expectancy Effects, and Parents’ Socialization of Gender Differences.” Journal of Social Issues 46(2):183–201.

22.

Entwisle

Doris R.

Alexander

Karl L.

Olson

Linda S.

1994. “The Gender Gap in Math: Its Possible Origins in Neighborhood Effects.” American Sociological Review 59(6):822–38.

23.

Feingold

Alan

. 1988. “Cognitive Gender Differences Are Disappearing.” American Psychologist 43(2):95–103.

24.

Fernandes-Alcantara

Adrienne L.

2012. Youth and the Labor Force: Background and Trends. Washington, DC: Congressional Research Service.

25.

Freeman

Catherine E.

2004. Trends in Educational Equity of Girls & Women: 2004. Technical Report No. NCES 2005-016. Washington, DC: U.S. Government Printing Office.

26.

Gager

Constance T.

Cooney

Teresa M.

Call

Kathleen T.

1999. “The Effects of Family Characteristics and Time Use on Teenagers’ Household Labor.” Journal of Marriage and Family 61(4):982–94.

27.

Goldin

Claudia

Katz

Lawrence F.

Kuziemko

Ilyana

. 2006. “The Homecoming of American College Women: The Reversal of the College Gender Gap.” Journal of Economic Perspectives 20(4):133–56.

28.

Goodnow

Jacqueline J.

1988. “Children’s Household Work: Its Nature and Functions.” Psychological Bulletin 103(1):5–26.

29.

Greenberger

Ellen

Steinberg

Laurence D.

1986. When Teenagers Work: The Psychological and Social Costs of Adolescent Employment. New York: Basic Books.

30.

Herman

Alexis

. 2000. Report on the Youth Labor Force. Washington, DC: U.S. Department of Labor.

31.

Hochschild

Arlie

. 1997. The Time Bind: When Work Becomes Home and Home Becomes Work. New York: Henry Holt.

32.

Hochschild

Arlie

Machung

Anne

. [1989] 2012. The Second Shift: Working Families and the Revolution at Home. New York: Penguin Group.

33.

Hubbard

William H. J.

2011. “The Phantom Gender Differences in the College Wage Premium.” The Journal of Human Resources 46(3):568–86.

34.

Jordan

Will J.

Lara

Julia

McPartland

James M.

1996. “Exploring the Causes of Early Dropout among Race-ethnic and Gender Groups.” Youth & Society 28(1):62–94.

35.

Juster

F. Thomas

Ono

Hiromi

Stafford

Frank P.

2004. “Changing Times of American Youth: 1981–2003.” Institute for Social Research, University of Michigan, Ann Arbor, MI.

36.

Kablaoui

Bassem N.

Pautler

Albert J.

1991. “The Effects of Part-time Work Experience on High School Students.” Journal of Career Development 17(3):195–211.

37.

Lillydahl

Jane H.

1990. “Academic Achievement and Part-time Employment of High School Students.” The Journal of Economic Education 21(3):307–16.

38.

Little

Roderick J. A.

Rubin

Donald B.

2002. Statistical Analysis with Missing Data. New York: John Wiley.

39.

Marsh

Herbert W.

1991. “Employment during High School: Character Building or a Subversion of Academic Goals?” Sociology of Education 64(3):172–89.

40.

Marsh

Herbert W.

Kleitman

Sabina

. 2005. “Consequences of Employment during High School: Character Building, Subversion of Academic Goals, or a Threshold?” American Educational Research Journal 42(2):331–69.

41.

Mason

Karen O.

Yu-Hsia

. 1988. “Attitudes toward Women’s Familial Roles: Changes in the United States, 1977-1985.” Gender & Society 2(1):39–57.

42.

McDaniel

Anne

. 2010. “Cross-national Gender Gaps in Educational Expectations: The Influence of National-level Gender Ideology and Educational Systems.” Comparative Education Review 54(1):27–50.

43.

McHale

Susan M.

Bartko

Todd

Crouter

Ann C.

Perry-Jenkins

Maureen

. 1990. “Children’s Housework and Psychosocial Functioning: The Mediating Effects of Parents’ Sex-role Behaviors and Attitudes.” Child Development 61(5):1413–26.

44.

McHale

Susan M.

Crouter

Ann C.

Tucker

Corinna J.

1999. “Family Context and Gender Role Socialization in Middle Childhood: Comparing Girls to Boys and Sisters to Brothers.” Child Development 70(4):990–1004.

45.

McHale

Susan M.

Crouter

Ann C.

Whiteman

Shawn D.

2003. “The Family Contexts of Gender Development in Childhood and Adolescence.” Social Development 12(1):125–48.

46.

McNeil

Linda M.

1984. Lowering Expectations: The Impact of Student Employment on Classroom Knowledge. Madison, WI: Wisconsin Center for Educational Research.

47.

Milkie

Melissa A.

Raley

Sara B.

Bianchi

Suzanne M.

2009. “Taking on the Second Shift: Time Allocations and Time Pressures of U.S. Parents with Preschoolers.” Social Forces 88(2):487–517.

48.

Mortimer

Jeylan T.

2003. Working and Growing Up in America. Cambridge, MA: Harvard University Press.

49.

Mortimer

Jeylan T.

Finch

Michael D.

1986. “The Effects of Part-time Work on Adolescent Self-concept and Achievement.” Pp. 66–89 in Becoming a Worker, edited by Borman

Kathryn M.

Reisman

Jane

. Norwood, NJ: Ablex.

50.

Mortimer

Jeylan T.

Finch

Michael D.

Ryu

Seongryeol

Shanahan

Michael J.

Call

Kathleen T.

1996. “The Effect of Work Intensity on Adolescent Mental Health, Achievement, and Behavioral Adjustment: New Evidence from a Prospective Study.” Child Development 67(3):1243–61.

51.

Pew Research Center. 2011. Is College Worth It? College Presidents, Public Access Value, Quality and Mission of Higher Education. Retrieved December 16, 2014 (http://www.pewsocialtrends.org/files/2011/05/higher-ed-report.pdf).

52.

Raley

Sara

Bianchi

Suzanne

. 2006. “Sons, Daughters, and Family Processes: Does Gender of Children Matter?” Annual Review of Sociology 32:401–21.

53.

Sayer

Liana C.

2007. “Gender Differences in the Relationship between Long Employee Hours and Multitasking.” Pp. 403–35 in Workplace Temporalities: Vol. 17. Research in the Sociology of Work, edited by Rubin

Beth A.

Oxford, England: Elsevier.

54.

Settersten

Richard

Ray

Barbra E.

2010. Not Quite Adults: Why 20-somethings Are Choosing a Slower Path to Adulthood, and Why It’s Good for Everyone. New York: Random House.

55.

Shanahan

Michael J.

Flaherty

Brian P.

2001. “Dynamic Patterns of Time Use in Adolescence.” Child Development 72(2):385–401.

56.

Singh

Kusum

Ozturk

Mehmet

. 2000. “Effect of Part-time Work on High School Mathematics and Science Course Taking.” The Journal of Educational Research 94(2):67–74.

57.

Staff

Jeremy

Mortimer

Jeylan T.

2007. “Educational and Work from Adolescence to Early Adulthood: Consequences for Educational Attainment.” Social Forces 85(3):1169–94.

58.

Stearns

Elizabeth

Moller

Stephanie

Blau

Judith

Potochnick

Stephanie

. 2007. “Staying Back and Dropping Out: The Relationship between Grade Retention and School Dropout.” Sociology of Education 80(3):210–40.

59.

Steel

Lauri

. 1991. “Early Work Experience among White and Non-white Youths: Implications for Subsequent Enrollment and Employment.” Youth & Society 22(4):419–47.

60.

Steinberg

Laurence

Dornbusch

Sanford M.

1991. “Negative Correlates of Part-time Employment during Adolescence: Replication and Elaboration.” Developmental Psychology 27(2):304–13.

61.

Stone

James R.

Mortimer

Jeylan T.

1998. “The Effect of Adolescent Employment on Vocational Development: Public and Educational Policy Implications.” Journal of Vocational Behavior 53(2):184–214.

62.

Van de Ven

Wynand P. M. M.

Van Praag

Bernard M. S.

1981. “The Demand for Deductibles in Private Health Insurance: A Probit Model with Simple Selection.” Journal of Econometrics 17(2):229–52.

63.

Vincent-Lancrin

Stéphan

. 2008. “The Reversal of Gender Inequalities in Higher Education: An On-going Trend.” Pp. 265–98 in Higher Education to 2030: Vol. 1. Demography. Paris, France: OECD Centre for Educational Research and Innovation.

64.

Zelizer

Viviana A. Rotman

. 1985. Pricing the Priceless Child: The Changing Social Value of Children. Princeton, NJ: Princeton University Press.

65.

Zimmer-Gembeck

Melanie J.

Mortimer

Jeylan T.

2006. “Adolescent Work, Vocational Development, and Education.” Review of Educational Research 76(4):537–66.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.23 MB