Family Socioeconomic Status and Academic Achievement Among Korean Adolescents

Abstract

This study examined pathways through which family socioeconomic status may influence adolescents’ academic achievement. We focused on parental monitoring and adolescents’ after-school time-use patterns as linking mechanisms. Participants were 441 twelve- to fourteen-year-old Korean adolescents who participated in the Korea Welfare Panel Study. Higher family economic pressure was linked with lower parental monitoring through elevated levels of parental depression. Parental monitoring was associated with more time spent in structured learning-oriented activities and less time spent in unstructured nonacademic activities. Both types of time-use activities and parental monitoring were associated with academic achievement. The results supported both the family stress model and the family investment model, indicating that families’ socioeconomic conditions were directly and indirectly linked to adolescents’ academic achievement. The findings are discussed in the context of Korean culture, which emphasizes high parental involvement and economic investment in children’s education. We also present broader implications beyond the local context.

Keywords

socioeconomic status academic achievement parental monitoring time use family stress model family investment model

Introduction

The educational achievement of young students has long been a major interest of researchers and the public, because it is a robust predictor of socioeconomic attainment in adulthood (Harper, Marcus, & Moore, 2003), subjective well-being (Masten et al., 2005), and physical health outcomes (Devaux, Sassi, Church, Cecchini, & Borgonovi, 2011). The recent “tiger mom” phenomenon in the United States (Kohler, Aldridge, Christensen, & Kilgo, 2012) reflects the public’s heightened interests in Asian educational models that have been considered to be closely associated with higher academic success. Students in East Asian countries, such as South Korea (hereafter, Korea), Singapore, China, and Japan, continue to earn top performances on various international tests measuring the academic achievement of adolescents, particularly in mathematics and science (Mullis, Martin, Foy, & Arora, 2012; OECD, 2010). Along with this trend, many cultural studies that focus on Asian parents’ increased involvement in children’s academic performance and its association with academic success have been conducted recently (Jeynes, 2011; H. Park, Byun, & Kim, 2011).

Korea has a different educational environment in comparison with the United States in terms of an extraordinarily high degree of parental aspiration, interest, and involvement in their children’s academic performance. This distinction is mainly rooted in the Confucian history of Korea. With their adoption about 2,000 years ago, the two Confucian virtues of “human-heartedness” and “knowledge” have especially influenced the conception of education in Korea. Since the adoption of Confucian virtues, success in Korean society has been defined in terms of acquiring academic knowledge and success (the virtue of knowledge) and the sacrifice, devotion, and support that come from parents (the virtue of human-heartedness), which are viewed as the most essential ingredients for the educational success of children (Y. S. Park & Kim, 2006). This so-called “educational zeal” of parents has been deeply rooted in this tradition and has become a unique characteristic of Korean education (Sorensen, 1994).

More importantly, however, the quality and amount of parental involvement and investment in children’s academic performance seem to be limited to the utilizable human and economic capital that parents have. Despite students’ exceptional performances in international assessment, concerns about weakened public schooling and the increased use of extracurricular private tutoring have grown in recent years in Korea (Kim & Lee, 2010). This is because extracurricular private tutoring or academic institutes for core subjects, such as mathematics and English, have been associated with increased test scores, and these resources have become important strategies for parents to enhance their children’s academic success (H. Park et al., 2011).While some parents may take advantage of such educational opportunities by investing substantial amounts of money and time into these strategies, children with parents who do not have sufficient human and economic capital to arrange opportunities for private education may be at risk of falling behind in a setting of intense academic competition. Because economic polarization has intensified in Korea since the economic crisis in the late 1990s (Kwack & Lee, 2007), this discrepancy among children in educational settings, based on parents’ economic and human resources, is a significant problem which could lead to the potential of educational inequality and, ultimately, to an intergenerational transmission of poverty.

Such unique characteristics of the Korean educational context provide an interesting case to investigate the linking mechanisms between families’ socioeconomic status (SES) and adolescents’ academic achievement, which may differ from the findings of Western countries. In addition, examining the association in light of parental involvement and different types of adolescent educational settings is an important and timely research problem. However, most other relevant previous studies have been conducted in the United States (Brooks-Gunn & Duncan, 1997; Caro, McDonald, & Willms, 2009; Wickrama & O’Neal, 2013), and less is known about this association in different cultural backgrounds. It is also unclear whether family resources directly influence adolescents’ academic achievement or indirectly influence their achievement through other family process variables. Based on family stress and family investment models, the present study examined whether family SES contributes to family processes and time-use patterns of adolescents and whether the potential mediational processes predict academic achievement of adolescents in Korean cultural context.

Family SES and Adolescents’ Academic Achievement

Numerous studies have found that family SES is associated with children’s poor academic achievement (Pears, Kim, & Fisher, 2008; Sirin, 2005). The family stress model (Conger et al., 1992; Conger & Elder, 1994) provides an explanation of this association: Family’s economic aspects, such as material hardship and economic pressure, measured by low income, unmet material needs and negative economic events, have adverse effects on parents’ behavior and emotions, which in turn can negatively influence their parenting practices (e.g., harsh, inconsistent, and uninvolved parenting) that can lead to negative developmental outcomes for their children, including lower academic performance. Studies based in the United States have consistently supported the suggested linking mechanisms (Barrera et al., 2002; Brody et al., 1994; Reising et al., 2012).

Unlike the family stress model, the family investment model proposes a more direct association among family economic resources, parental investment, and academic achievement (Becker & Tomes, 1994; Bradley & Corwyn, 2002; Duncan & Magnuson, 2003). This model does not primarily focus on the negative emotional aspects of parents or disrupted parenting, which may be caused by economic difficulties. Rather, the model proposes that parents who have more economic resources are able to make investments for the successful development of their children (e.g., arranging private tutoring or academic institutes for extracurricular studies); parents who have fewer resources invest their capital and income in more urgent family needs. Therefore, wealthier parents can likely support children directly through advanced or specialized tutoring or training.

It has been argued that the social and emotional development of adolescents is better explained by the family stress model and that the family investment model provides a better explanation of cognitive development (Conger, Conger, & Martin, 2010). In agreement with previous studies, we assumed that the family investment model would provide a better explanation of the association between family SES and adolescents’ educational achievement. However, because of the high level of educational zeal and aspiration of many Korean parents, parental characteristics may have their own right in the academic achievement of children. It can be assumed, therefore, that a lack of emotional and behavioral competencies of parents (e.g., depressive symptoms and uninvolved parenting reflected by lack of parental monitoring) may actively mediate the association between a family’s economic status and an adolescent’s academic outcomes, which could also support the family stress model.

Therefore, we proposed a combined model of family stress and family investment models, as shown in Figure 1. We investigated not only the direct association among families’ economic aspects, time-use patterns, and academic achievement of Korean adolescents (as suggested by the family investment model) but also the indirect associations through family processes (as suggested by the family stress model). This endeavor may contribute in evaluating the relative contributions of each model to the relationship between family SES and the academic achievement of adolescents. In doing so, we included parental education as another important marker of family SES, in addition to family economic difficulties, because of the proposal that parental education is considered by many contemporary investigators to be the canonical standard of SES (Conger et al., 2010).

Figure 1.

Conceptual model.

Family Characteristics and Adolescents’ Time Use in Out-of-School Activities

Adolescents’ time-use patterns in out-of-school activities are important proximal factors for estimating their home educational settings and parental involvement in the education of their children, which are closely related to academic success (Schneider & Lee, 1990; Won & Han, 2010). For instance, adolescents who spend more time in structured learning-oriented activities, such as after-school programs, private tutoring, and institutes that are likely to be facilitated by their parents, may spend more time in academic environments than in nonacademic settings, such as playing with peers, using the Internet, or watching television, since time is a limited resource.

Previous studies have demonstrated that a family’s SES is strongly associated with an adolescent’s choice of out-of-school activities. Children whose parents had higher educational attainment spent significantly more time reading, while children whose parents were not college graduates spent significantly more time watching television than children whose parents had more education (Bianchi & Robinson, 1997). Another study found that children and adolescents with more educated parents were more likely to spend time in studying and reading, while spending less time watching television than children with less-educated parents (Hofferth & Sandberg, 2001; Wright, Price, Bianchi, & Hunt, 2009). Family income was correlated positively with the time that adolescents spent on homework (Wright et al., 2009) and negatively with television viewing (Hofferth & Sandberg, 2001).

Family processes, such as parental depression and parenting behaviors are also closely related to adolescents’ out-of-school time-use patterns. Higher maternal depressive symptoms were associated with more television viewing as well as more viewing of programs with age-inappropriate content among children (Conners, Tripathi, Clubb, & Bradley, 2007). Parental involvement has been found to be negatively related to heavy television viewing among children. Jago and associates (2011) found that the likelihood of children watching television more than 4 hours per day was 3.3 times higher in families in which permissive mothers imposed few restrictions on their children’s behavior. Disengaged parenting predicted poor academic performance and heavy television viewing 5 years later (Davison, Francis, & Birch, 2005).

Some cultural studies conducted in the Korean context provide similar results. Parents’ years of education were negatively associated with intense Internet use, watching television, and PC gaming (Jung, Kim, Lin, & Cheong, 2005). Higher household incomes and more years of education attained by parents affected the expenditure of money and time spent on private tutoring (Kim & Lee, 2010), which means more time that was spent in academic environments. Parental restriction on the Internet and game use were associated with less time that was spent on the Internet among children (Park, 2011).

Although the predictors of family stress and family investment models are closely associated with adolescents’ time-use patterns, less is known about the continuous processes involving time-use patterns stemming from family SES and processes that contribute to adolescents’ academic achievement. In this regard and in light of findings from previous literature, adolescents’ time-use patterns after school in learning-oriented and nonstudying activities were included in the combined model as important proximal variables of different educational settings that may vary depending on family SES. As shown in Figure 1, we expected that family SES may be either directly or indirectly associated with time-use patterns of adolescents, as supposed by family investment and family stress models, respectively.

Adolescents’ Time Use and Academic Achievement

It is well established that adolescents’ time-use patterns in out-of-school activities are closely associated with the academic achievement of adolescents. According to the time-displacement hypothesis, television viewing and heavy Internet use can take time away from intellectual activities that involve learning opportunities (Shin, 2004) and is associated with poor academic achievement (Romer, Bagdasarov, & More, 2013). Engagement in homework (Leone & Richards, 1989) and literacy-oriented activities, such as reading books and visiting the library (Powell, Peet, & Peet, 2002), were positively associated with school performance. Academically unsuccessful boys spent more time watching television and playing video games than did their academically high-achieving peers (Madden, Bruekman, & Littlejohn, 1997).

Studies conducted in the Korean cultural context show that high-achieving students were more likely to spend their extra time in reading books and getting private tutoring and were less likely to play with their friends (C. Park & Park, 2006). Reading books and doing homework were positive predictors of academic achievement, while watching television, PC gaming, and using the Internet were negative predictors of achievement (Won & Han, 2010). Interestingly, playing with friends was negatively associated with academic achievement among high achievers, while such conduct was positively associated among low achievers (Won & Han, 2010).

Drawing on these findings, the present study is designed to examine a range of out-of-school activities in which early adolescents invest their time after school. These activities include television viewing, Internet use, playing outside with peers, studying at home, and attending private academic institutes or being tutored at home privately in supplementary study. Although the classification of these activities is inconsistent among researchers, structured activities were generally defined as learning-oriented activities, such as doing schoolwork, going to the library and playing sports, while unstructured activities were defined as other activities, such as watching television, sleeping, and playing with friends (Dotterer, McHale, & Crouter, 2007; Larson & Verma, 1999). Therefore, we categorized studying at home (including private tutoring) and attending academic institutes as structured learning-oriented activities. We categorized other activities, such as television viewing, Internet use, and playing with peers as unstructured nonacademic activities. We expected structured learning-oriented activities to enhance academic achievement and unstructured nonactivities to be associated with lower achievement.

The Current Study

Building on the family stress and family investment models, the present study aimed to analyze the relationships between family SES and the academic achievement of adolescents using longitudinal data from a national representative sample of young Korean adolescents. We hypothesized that family SES predicts family processes, such as parental depressive symptoms and monitoring, which are in turn linked to two different after-school time-use patterns of adolescents: structured learning-oriented time use and unstructured nonacademic time use. We expected that while learning-oriented time-use patterns predict higher academic achievement, nonacademic time use is associated with lower achievement. Additionally, we examined the existence of a direct influence of family SES on adolescents’ academic achievement after accounting for the linking mechanisms of family processes and the time-use patterns of adolescents.

Method

Sample and Data

Data for this study came from a nationally representative sample of young adolescents participating in the Korea Welfare Panel Study (KOWEPS) that was conducted by the Korea Institute for Health and Social Affairs (KIHASA), which is a government-funded research agency. KOWEPS is a comprehensive longitudinal survey that has been conducted every year since 2006 and that traces economic activity, receipt of social security benefits, and other demographic characteristics of household members. Since 2006, separate surveys have been administered every 3 years to collect information on young adolescents’ academic achievement, school environment, educational resources, and parental support.

In 2006, the baseline (Wave 1) data were derived from a survey of 14,469 adult respondents and 750 young adolescent respondents between the ages 9 and 11 years (fourth-, fifth-, and sixth-grade elementary school students) from 7,072 households. KOWEPS used a two-phase sampling method. First, the sample was stratified by region, city, and type of residence. Next, 3,500 households in which the income level was below 60% of the median sample income were selected, and another 3,500 households in which the income level was above 60% of the median sample income were selected. The second (Wave 2) and fourth (Wave 4) waves of data were collected in 2007 and 2009 (Wave 2: n = 13,478; Wave 4: n = 12,661), respectively. An additional survey of young adolescents between ages 12 and 14 years (seventh-, eighth-, and ninth-grade middle school students) was conducted during Wave 4 (n = 608). We used the data from adolescents and parents in two-parent families who participated in Wave 1, Wave 2, and Wave 4 of the study (N = 441). An attrition analysis was conducted to test for differences among adolescents who provided data for the various waves. t test results confirmed that the differences in background variables, such as family SES and parental characteristics between attritors and stayers were not statistically significant.

Measurements

Family socioeconomic status measure

Using the data gathered from parents during Wave 1 in 2006, we identified two components of family SES: economic pressure and parental education attainment.

Family economic pressure

Family economic pressure was quantified using multiple indicators of material hardship and poverty. During Wave 1, parents responded to eight dichotomously scored (0 = no, 1 = yes) material hardship experience items that asked whether any member of the household experienced food insecurity; could not afford house rent and had to move; had difficulty in paying utility bills; had ever had his or her telephone, electricity, or water disconnected; had difficulty in paying his or her children’s educational expenses; had difficulty in paying medical/health insurance; or had a delinquent credit account. The sum of the responses to these eight items yielded an internally consistent index of material hardship (α = .76) that ranged from 0 to 8, with higher scores reflecting greater economic hardship. Family poverty was measured using a dichotomous categorization that indicated whether the household income level was below (scored 1) or above (scored 0) 60% of the sample’s median income.

Parental education

Ordinal-level education categories for mothers and fathers were generated based on the parents’ questionnaire in Wave 1 concerning their highest levels of formal education attainment: 0 = never went to school, 1 = elementary school or less, 2 = more than elementary school but did not graduate from middle school, 3 = more than middle school but did not graduate from high school, 4 = high school graduation or its equivalent but less than a college degree, 5 = graduated from a college or university, and 6 = professional training beyond a 4-year college or university degree. The values of maternal and paternal education levels served as an index of parental education in each family.

Parent measures

Parental depressive symptoms were measured using the parents’ survey in Wave 2 (2007), and parental monitoring was measured using the children’s and adolescents’ survey in Wave 4 (2009).

Parental depression symptoms

Participants responded to 11 items from the Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977). These items assessed the frequency of parents’ feelings of distress (e.g., “felt depressed and sad,” “felt lonely”) during the past week on a scale ranging from 0 (never or rarely) to 3 (most of the time or more than 6 days per week). Using exploratory factor analysis, Radloff (1977) found that the CES-D has four factors: depressed affect, positive affect, somatic complaints, and interpersonal problems. This four-factor model was replicated and widely accepted in subsequent studies (Knight, Williams, McGee, & Olaman, 1997; Nguyen, Kitner-Triolo, Evans, & Zonderman, 2004), suggesting that the four-factor model resulted in a significantly better fit, and the four dimensions of the CES-D seem to be more informative in assessing depressive symptoms (Zhang et al., 2011). This study quantified parental depressive symptoms by categorizing parents’ responses based on the four-factor model: depressed affect (3 items) = felt depressed, sad, lonely; positive affect (2 items) = felt as good as others, enjoyed life; somatic complaints (4 items) = had poor appetite, restless sleep, felt everything was an effort, could not get going; and interpersonal problems (2 items) = people were unfriendly, people disliked me. We developed separate indices for mothers and fathers. Each depressive symptom factor had an adequate internal consistency: Fathers: .75 (depressed), .69 (positive), .70 (somatic), and .73 (interpersonal); Mothers: .80 (depressed), .73 (positive), .72 (somatic), and .73 (interpersonal).

Parental monitoring

The children and adolescents who were surveyed in Wave 4 answered questions about their perceptions of parental monitoring. These questions were adopted from parental monitoring scales that were used in previous studies on parental monitoring (Cernkovich & Giordano, 1987; Weintraub & Gold, 1991). Adolescents answered the following questions: (a) In my free time away from home, my parents know where I am and who I am with; (b) My parents know when I am back home; (c) My parents know what I am doing when I am away from home; and (d) My parents call me when I am at home alone without any adults. Each item was rated on a scale ranging from (0) not at all to (3) all the time. These items were used as multiple indicators; they had loadings of .75, .76, .75, and .54, respectively. The constructed index was also reliable (α = .79). Higher values of parental monitoring served as an index of more involved parenting.

Adolescent measures

Measures of adolescents’ time use in out-of-school activities and school achievement were taken from Wave 4 (2009).

Time use in out-of-school activities

Five out-of-school activities were assessed to investigate adolescents’ time-use patterns after school: (a) watching television, (b) using the Internet, (c) studying at home (including both doing homework and being tutored at home privately for supplementary study), (d) attending private academic institutes, and (e) playing outside with peers (movie watching, shopping, visiting computer cafes, and karaoke). Due to the lack of continuous measurement for some activities, the study was unable to acquire information on the absolute amount of time that adolescents spent for each activity. Instead, we tried to identify the primary activities that young adolescents engaged in after school by generating dichotomous variables for all five activities. First, two types of media use, watching television and Internet use, were assessed with questions that asked how many hours the participants usually spent on each media per day. Adolescents who scored one standard deviation above the sample average on television viewing and Internet use were categorized as belonging to the “primarily watching television” and “primarily using Internet” after-school activity groups. The sample mean and standard deviation of television viewing ( $\bar{X}$ = 2.09 hours, SD = 1.43 hours) and Internet use ( $\bar{X}$ = 1.62 hours, SD = 1.19 hours) were similar to the results of previous time-use studies on media that were conducted with young adolescents (Hancox, Milne, & Poulton, 2005). Categorizing one standard deviation above the sample average as “primary” media-use behaviors also produced results compatible with the previous literature on excessive media time use (American Academy of Pediatrics, 2001; Gomez et al., 2007). Next, in order to assess the other three activities, studying at home, attending private academic institutes, and playing with peers, questions that asked adolescents to indicate their most frequent after-school activities during weekdays were used. If adolescents chose any of those three activities as their most frequent activities that they engaged in after school, they were categorized as belonging to “primarily studying at home,” “primarily attending academic institutes,” and “primarily playing with peers” groups. Then, we categorized those five time-use activities into structured and unstructured activities. First, “structured learning-oriented activities” were generated by adding the two dichotomous variables of primary time spent in (a) studying at home and (b) attending academic institutes. The variable ranged from 0 to 2, and higher values reflected more time spent in structured extracurricular academic activities. “Unstructured nonacademic activities” were generated by adding the three dichotomous variables of primary time spent (a) watching television, (b) using the Internet, and (c) playing with peers. The variable ranged from 0 to 3, and a higher value meant more time spent in unstructured nonacademic activities.

Academic achievement

Average scores from the adolescents’ self-evaluations of their academic achievement during the past year in mathematics, Korean, English, and overall subjects were used to quantify the academic achievement measurements. Questions assessed adolescents’ honest opinions of the academic achievement on scales that ranged from 0 to 4: 0 = very low achievement, 1 = below average achievement, 2 = average achievement, 3 = more than average achievement, and 4 = very high achievement.

Analytic Approach

In order to evaluate the conceptual model (see Figure 1), we began by examining the descriptive statistics and zero-order correlations among the study variables. We then conducted structural equation modeling (SEM) in order to evaluate possible linking mechanisms between the families’ SES variables and adolescents’ academic achievement by using Mplus 7. For justification of our final conceptual model, several SEM models were tested, and their model fit indices were compared. We used the Comparative Fit Index (CFI ≥ .95; Bentler & Bonett, 1980) and Root Mean Square Error of Approximation (RMSEA ≤ .06; Steiger & Lind, 1980) to evaluate the model fit. The significance of the indirect effects was calculated via bootstrapping method (Preacher & Hayes, 2008). Bias-corrected 95% confidence intervals (CI) were computed using 10,000 bootstrapped resamples for each indirect estimate. CIs that do not contain a zero value indicate a significant effect. Missing data were accounted for using the Full Information Maximum Likelihood (FIML) procedure.

Results

Table 1 reports means, standard deviations, and ranges of the major study variables. The mean material hardship experience of families was low ( $\bar{X}$ = .58, SD = 1.22), and 24% of the sample reported that their household income was below 60% of the median household income. The mean educational attainment level of both parents was above middle school graduation. On average, adolescents showed “above average achievement” in all subjects.

Table 1.

Descriptive Statistics of Study Variables (N = 441).

Variable	$\bar{X}$	SD	Range
Family SES
Material hardship	0.58	1.22	0-8
Family poverty	0.24	0.43	0-1
Father education	3.51	1.22	0-6
Mother education	3.24	0.93	1-6
Father depressive symptoms
Depressed	0.96	1.58	0-9
Positive	1.19	1.53	0-6
Somatic	1.98	2.21	0-12
Interpersonal	0.21	0.71	0-6
Mother depressive symptoms
Depressed	1.06	1.69	0-9
Positive	1.26	1.67	0-6
Somatic	1.90	2.23	0-12
Interpersonal	0.16	0.62	0-6
Parental monitoring
Know whereabouts	1.73	0.92	0-3
Know when back home	2.03	0.83	0-3
Know what they do	1.74	0.88	0-3
Call home	1.90	0.94	0-3
Adolescent out-of-school time use
Structured learning-oriented activities	0.77	0.61	0-2
Unstructured nonacademic activities	0.41	0.63	0-3
Adolescent academic achievement
Mathematics	2.11	1.28	0-4
Korean	2.53	0.92	0-4
English	2.13	1.25	0-4
Overall	2.15	0.92	0-4

Note. SES = socioeconomic status.

The results of zero-order correlations among study variables for the data were consistent with our expectations: The families’ SES variables were significantly correlated with young adolescents’ academic achievement, ranging from −.35 to .34. Maternal depression was correlated with parental monitoring (−.19), but not with paternal depression. Academic achievement was significantly correlated with parental monitoring (ranging from .18 to .30) and adolescents’ time-use patterns (ranging from −.25 to .29; Table 2). Considered together, the pattern and strengths of the correlations among the study constructs support additional tests related to the proposed model.

Table 2.

Pearson’s Correlation Coefficients Among Study Variables (N = 441).

	1	2	3	4	5	6	7	8	9	10	11
1. Economic pressure	—
2. Father education	−.40***	—
3. Mother education	−.35***	.67***	—
4. Father depression	.52***	−.06	−.15**	—
5. Mother depression	.57***	−.18**	−.18**	.39***	—
6. Parental monitoring	−.12	.18**	.14*	−.01	−.19**	—
7. Structured learning-oriented activities	−.37***	.21**	.21***	−.07	−.15**	.18***	—
8. Unstructured nonacademic activities	.12	−.23***	−.23***	.09	.04	−.24***	−.31***	—
9. Mathematics	−.19**	.23***	.23***	−.01	−.07	.24***	.29***	−.21***	—
10. Korean	−.06	.18***	.18***	−.01	−.08	.18**	.11*	−.13**	.48***	—
11. English	−.35***	.34***	.34***	.01	−.08	.20***	.28***	−.15**	.54***	.48***	—
12. Overall	−.23**	.28***	.28***	−.01	−.09	.30***	.22***	−.25***	.68***	.66***	.64***

p < .05. **p < .01. ***p < .001.

To justify that the conceptual model shown in Figure 1 postulated valid hypotheses and provided the best model fit for the data, we tested statistical comparisons of different models that are subsets of the final conceptual model. Table 3 shows that Model 4, which is the conceptual model in Figure 1, has the best fit to the data in terms of a significantly decreased chi-square value (Δχ² = 116.80 with Δdf = 20) with the highest CFI and the lowest RMSEA. The results of the nested model comparisons support the existence of the indirect effects from family SES to adolescent academic achievement (Wickrama, Lorenz, Conger, Matthews, & Elder, 1997). Thereafter, we tested significance of the mediation of family process and adolescents’ time-use activities using bootstrapping. Although the indirect effect of family SES on academic achievement through all possible mediators (i.e., parental depression, parental monitoring, and time-use activities) was not significant, several partial indirect paths were significant. The results indicated that adolescents’ unstructured nonacademic activities significantly mediated the effects of father’s educational level on overall achievement of adolescents (b = .020, CI = [.002, .059]). In addition, the indirect effects of parental monitoring on overall achievement via unstructured nonacademic activities (b = .038, CI = [.011, .090]) and on math achievement via structured learning-oriented activities (b = .048, CI = [.003, .140]) were significant.

Table 3.

Summary of Model Fit Comparisons.

Model	Path	χ²	df	Δχ²(Δdf)	CFI	RMSEA
Model 1	Family SES → Academic achievement	356.59	183	—	.939	.046
Model 2	Family SES → Parental depression → Academic achievement	343.73	177	12.86(6)	.942	.046
Model 3	Family SES → Parental depression → Parental monitoring → Academic achievement	313.71	171	30.07(6)	.950	.044
Model 4	Family SES → Parental depression → Parental monitoring → Adolescent time use → Academic achievement	196.91	151	116.80(20)	.984	.026

Note. CFI = Comparative Fit Index; RMSEA = Root Mean Square Error of Approximation; SES = socioeconomic status.

Figure 2 provides the standardized coefficients of the tested model. The model showed an adequate fit to the data; χ²(151) = 196.909, CFI = .984, Tucker-Lewis Index (TLI) = .976, RMSEA = .026. All path coefficients presented in Figure 2 were statistically significant at α = .05 level except for the dashed paths from mothers’ education to mothers’ depression (b = −.01, p = .88) and fathers’ depression to parental monitoring (b = .06, p = .44).

Figure 2.

Linking family socioeconomic characteristics to adolescent academic achievement.

Overall, the findings supported our hypotheses. As expected, family SES was related to academic achievement. Family economic pressure was associated with higher depressive symptoms in both mothers (b = .60, p < .001) and fathers (b = .59, p < .001). In turn, maternal depressive symptoms predicted lower parental monitoring (b = −.19, p < .05). For parental monitoring, positive associations were found with all academic achievement indicators (Overall: b = .22, p < .001; Korean: b = .12, p < .05; English: b = .13, p < .05; Mathematics: b = .17, p < .01). In addition, parental monitoring was negatively associated with time spent in unstructured nonacademic activities (b = −.21, p < .001) and positively associated with structured learning-oriented activities (b = .14, p < .05). More specifically, higher parental monitoring subsequently predicted less time spent in nonacademic activities, such as watching television, using the Internet, and playing with friends, and more time spent in extracurricular academic activities, such as getting private tutoring at home or attending private academic institutes. While unstructured nonacademic time use was associated with lower achievement in overall subjects (b = −.13, p < .01), structured learning-oriented time use was associated with higher achievement in mathematics (b = .19, p < .01).

The results also showed that the families’ SES was directly associated with adolescents’ academic achievement. Specifically, economic pressure was negatively associated with adolescents’ English achievement (b = −.42, p < .01). That is, adolescents who experienced higher economic pressure reported lower academic achievement than their peers who experienced less economic pressure. The mothers’ educational attainment was positively associated with higher overall achievement (b = .15, p < .05), while the fathers’ educational attainment was not related to adolescents’ academic achievement. Thus, our hypothesized indirect paths did not completely capture the association between family SES and adolescents’ academic achievement as the direct effect of the family SES remained statistically significant.

Furthermore, some direct paths from the families’ SES to adolescent’s time use were found. Higher family economic pressure was associated with less time spent in structured learning-oriented activities (b = −.46, p < .01). The fathers’ higher educational attainment level was associated with less time spent in unstructured nonacademic activities (b = −.20, p < .01). The results showed that parents who experienced economic pressure were less likely to support their children by direct supervision for engaging in homework, or via private tutoring and academic institutes. Parents who had more human capital with a higher educational level were more likely to support their children by supervising and restricting their children’s time-use patterns.

Discussion

Young adolescents’ educational success is an important indicator of socioeconomic attainment in their adulthood. Therefore, parents are often enthusiastically involved in their children’s academic progress across many societies. The Korean cultural context particularly stresses the importance of success in academics with a strong family bond; however, the academic achievement of Korean children has substantial economic and social returns and is closely associated with the well-being of both the children and their parents (Sorensen, 1994). Many Korean parents equate their children’s success in school with their own achievements and allocate their families’ economic and human resources in order to maximize educational opportunities for their children. With this particular cultural context of Korea in mind, we have extended the current literature on family SES and children’s academic achievement that has focused on education in the United States by identifying linking mechanisms.

Through a longitudinal design, we tested a SEM model of the links between families’ SES and academic achievement among Korean adolescents across a 3-year period. Similar to findings in the United States, our empirical analysis found negative relationships between family SES measures and parental depression in Korea: Higher economic pressures on the family and lower educational attainment of the parents were associated with higher parental depression. Then, parents’ depressive symptoms were associated with lower parental monitoring, which, in turn, predicted lower academic achievement of adolescents in all tested subjects.

The results of our empirical analysis suggested several important points for discussion. The significant effect of a family’s SES on an adolescent’s academic achievement through parental characteristics raises concerns about double-jeopardy. For example, parents who lack human and economic resources may not only experience low psychological well-being but may also face the negative cognitive outcomes of their children, which may substantially influence their well-being in return. Moreover, children from families with a low SES may not only be exposed to unfavorable family environments, such as insufficient monitoring and interactions with parents, but may also experience low academic success at school, which may aggravate their weak relationships with their parents through failing to meet parental expectations.

Possibly because of the high level of cultural educational zeal of Korean parents, parental monitoring was directly associated with academic success, showing its importance in the academic achievement of adolescents. However, parental monitoring was also indirectly associated with academic achievement through adolescents’ time-use activities. Adolescents who spent more time in private tutoring and academic institutes and spent less time in unstructured media use or playing with friends showed higher test scores in mathematics and showed better achievement overall. The significant effect of parental monitoring on the academic success of their children through structured learning-oriented time-use activities and the substantial association between parental monitoring and their SES highlights potential educational inequalities and intergenerational transmission of socioeconomic adversity. With more economic resources, higher SES parents are more likely to be able to afford high-quality and specialized private tutoring and academic institutes, which meet their children’s specific needs, than lower SES parents. This kind of parental investment for additional academic opportunities, in turn, likely leads to improved academic achievement of higher SES children. In addition, higher SES parents may also have different types of resources in terms of time, because both parents may not have to work full-time or can hire someone to take care of their children. Compared with working-class parents, parents from a middle or higher SES may be effective at monitoring children’s academic progress through managing and arranging their after-school time use and interacting more with their children, private tutors, and schoolteachers.

Another interesting point worth noting is that different time-use activities are likely to be associated with different academic achievements. Specifically, adolescents’ structured time use in learning-oriented extracurricular activities was associated with improved achievement in mathematics. This result may reflect the unique cultural aspects of Korea in that mathematics is considered one of the most important subjects by parents and children. The curriculum of primary and secondary education in the Korean public educational system mainly focuses on four subject areas, Korean, mathematics, English, and science/social studies, which are assessed on the college entrance exam (known as the College Scholastic Ability Test [CSAT]). This entrance exam is well known for its competitiveness, and students are discriminated more by their mathematics scores than by their scores in other subjects (Park & Leung, 2004). Moreover, in comparison with other Western countries, the mathematics curriculum is more demanding in East Asian countries (Leung, 2006), which requires an advanced level of mastery. Therefore, the perceived importance and difficulties of mathematics may motivate Korean parents to seek the services of private tutoring and academic institutes for mathematics.

Lastly, one purpose for this study was to respond to a gap in the literature by integrating two theoretical perspectives, the family stress and family investment models. Earlier studies competitively employed only one of these models, which either suggested that family stress predictors explained social emotional development of children better or the family investment predictors explained children’s cognitive development better (Conger et al., 2010). Conducting test of the combined family stress and family investment models was suggested in order to determine whether the two models construct a theoretical alliance (Conger et al., 2010). Our joint-test model supports the integrated model of “family stress and investment,” because a family’s SES is linked to academic achievement either directly or indirectly. Our results supported the family stress model and showed linking pathways from a family’s SES to an adolescent’s academic achievement through parental characteristics and time-use activities. The results also supported the family investment model and showed more direct pathways between family’s economic pressure and structured learning-oriented activities and between fathers’ educational levels and unstructured nonacademic activities, which suggested a lack of parental support through economic and human resources to provide specialized tutoring or to restrict nonacademic time-use activities. These direct paths posit that parents who have greater resources make investments for their children and that parents with fewer resources spend their capital on immediate family needs. Moreover, our study showed that family stress processes are better predictors of academic achievement for Korean adolescents, again, supporting that both frameworks should be understood as a comprehensive model, and the joint-test model of the present study provides better explanations than the two independent models.

Our findings provided support for the proposed research hypotheses. Nevertheless, the study has some limitations. First, our findings were derived from adolescents’ subjective reports of their academic achievement. More objective measures, such as GPAs or class ranking, may be better indices for the assessment of adolescents’ academic achievement. Second, adolescents’ time-use variables were measured as degrees of involvement in each activity rather than an absolute amount of time that adolescents spent on each pursuit because of a lack of instruments for the assessment of continuous time use. Inevitably, this challenge caused a reduction of information on adolescents’ time-use behaviors. The content of time-use activities also should be considered in future research, because the quality of each time-use behavior may exert influence on academic achievement under the assumption that the negative linear relationship between time use in unstructured nonacademic activities and lower academic achievement cannot be certain without this information. The continuous time-use measurements and analysis of content of time use may be useful indicators that complement the measures that were used in the present study to provide a fuller picture of time-use patterns and academic achievement. In addition, even though we allowed a time lag between the predictors and the outcome with longitudinal data, we could not include prior levels of mediating and outcome variables in the tested model because of the lack of information from earlier waves. As a consequence, the key path coefficients may show inflated estimates. This limitation may reduce the benefits of using a longitudinal design.

Despite the limitations of this research, its findings have practical implications. Our findings on the substantial association between a family’s SES and time spent in private tutoring or in academic institutes raise the problem of educational inequality. Income inequality and polarization increased in the 2000s in Korea (Kwack & Lee, 2007), and the socioeconomic gap in private tutoring also increased recently (Byun & Kim, 2010). Tackling the low academic achievement of adolescents and the growing influence of the socioeconomic backgrounds of families on academic success is a difficult and complex problem. However, as social concerns about weakening chances for academic success for students from lower SES families have grown, more governmental efforts to support these adolescents to accrue human capital are required than ever before. Therefore, the results of this study can be used to demand more comprehensive research and start a policy debate to address the inequality of educational opportunities and develop multifaceted interventions to curb polarization of educational changes. In addition, our study showed that rather than adolescents’ time that was spent in different educational settings, parental involvement in children’s lives exerted stronger influence on academic performance. This result may provide useful information in understanding the changing educational environment in the United States, because private tutoring has greatly expanded in Western countries, including the United Studies (Bray, 2006). Excessive investment in private educational settings may undermine accumulating family assets and create chronic stress for family members. On the other hand, intervention programs that are directed at increasing parental interest and attention in children’s lives, such as school activities and educational progress, may strengthen students’ chances for academic success, particularly those students who come from lower SES families.

Footnotes

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.

Author Biographies

Dayoung Bae is a doctoral student at the Department of Human Development and Family Science, University of Georgia. Her current research interests center on neighborhood context and socioeconomic inequality in health and socioeconomic attainment of adolescents and young adults.

K. A. S. Wickrama is a professor at the University of Georgia at the Department of Human Development and Family Science. His research focuses on social determinants of health and health inequality across the life course, racial/ethnical inequalities in mental and physical health, and application of advanced statistical methods to social epidemiology.

References

American Academy of Pediatrics. (2001). Children, adolescents, and television. Pediatrics, 107, 423-426.

Barrera

Prelow

H. M.

Dumka

L. E.

Gonzales

N. A.

Knight

G. P.

Michaels

M. L.

. . . Tein

(2002). Pathways from family economic conditions to adolescents’ distress: Supportive parenting, stressors outside the family, and deviant peers. Journal of Community Psychology, 30, 135-152.

Becker

G. S.

Tomes

(1994). Human capital and the rise and fall of families. In Becker

G. S.

(Ed.), Human capital: A theoretical and empirical analysis with special reference to education (3rd ed., pp. 257-298). Chicago, IL: The University of Chicago Press.

Bentler

P. M.

Bonett

D. G.

(1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88, 588-606.

Bianchi

S. M.

Robinson

(1997). What did you do today? Children’s use of time, family composition, and the acquisition of social capital. Journal of Marriage and Family, 59, 332-344.

Bradley

R. H.

Corwyn

R. F.

(2002). Socioeconomic status and child development. Annual Review of Psychology, 53, 371-399.

Bray

(2006). Private supplementary tutoring: Comparative perspectives on patterns and implications. Compare, 36, 515-530.

Brody

G. H.

Stoneman

Flor

McCrary

Hastings

Conyers

(1994). Financial resources, parent psychological functioning, parent co-caregiving, and early adolescent competence in rural two-parent African-American families. Child Development, 65, 590-605.

Brooks-Gunn

Duncan

G. J.

(1997). The effects of poverty on children. The Future of Children, 7(2), 55-71.

10.

Byun

S. Y.

Kim

K. K.

(2010). Educational inequality in South Korea: The widening socioeconomic gap in student achievement. Research in the Sociology of Education, 17, 155-182.

11.

Caro

D. H.

McDonald

J. T.

Willms

J. D.

(2009). Socio-economic status and academic achievement trajectories from childhood to adolescence. Canadian Journal of Education, 32, 558-590.

12.

Cernkovich

S. A.

Giordano

P. C.

(1987). Family relationships and delinquency. Criminology, 24, 295-321.

13.

Conger

R. D.

Conger

K. J.

Elder

G. H.

Lorenz

F. O.

Simons

R. L.

Whitbeck

L. B.

(1992). A family process model of economic hardship and adjustment of early adolescent boys. Child Development, 63, 526-541.

14.

Conger

R. D.

Conger

K. J.

Martin

M. J.

(2010). Socioeconomic status, family processes, and individual development. Journal of Marriage and Family, 72, 685-704.

15.

Conger

R. D.

Elder

G. H.

(1994). Families in troubled times: Adapting to change in rural America. Hawthorne, NY: Aldine de Gruyter.

16.

Conners

N. A.

Tripathi

S. P.

Clubb

Bradley

R. H.

(2007). Maternal characteristics associated with television viewing habits of low-income preschool children. Journal of Child and Family Studies, 16, 415-425.

17.

Davison

K. K.

Francis

L. A.

Birch

L. L.

(2005). Link between parents’ and girls’ television viewing behaviors: A longitudinal examination. The Journal of Pediatrics, 147, 436-442.

18.

Devaux

Sassi

Church

Cecchini

Borgonovi

(2011). Exploring the relationship between education and obesity. OECD Journal, 2011, 121-145.

19.

Dotterer

A. M.

McHale

S. M.

Crouter

A. C.

(2007). Implications of out-of-school activities for school engagement in African American adolescents. Journal of Youth and Adolescence, 36, 391-401.

20.

Duncan

G. J.

Magnuson

K. A.

(2003). Off with Hollingshead: Socioeconomic resources, parenting, and child development. In Bornstein

M. H.

Bradley

R. H.

(Eds.), Socioeconomic status, parenting, and child development (pp. 83-106). Mahwah, NJ: Lawrence Erlbaum.

21.

Gomez

L. F.

Parra

D. C.

Lobelo

Samper

Moreno

Jacoby

. . . Borda

(2007). Television viewing and its association with overweight in Colombian children: Results from the 2005 National Nutrition Survey: A cross sectional study. International Journal of Behavioral Nutrition and Physical Activity, 4(1), 41-48.

22.

Hancox

R. J.

Milne

B. J.

Poulton

(2005). Association of television viewing during childhood with poor educational achievement. Archives of Pediatrics & Adolescent Medicine, 159, 614-618.

23.

Harper

Marcus

Moore

(2003). Enduring poverty and the conditions of childhood: Lifecourse and intergenerational poverty transmissions. World Development, 31, 535-554.

24.

Hofferth

S. L.

Sandberg

J. F.

(2001). How American children spend their time. Journal of Marriage and Family, 63, 295-308.

25.

Jago

Davison

K. K.

Thompson

J. L.

Page

A. S.

Brockman

Fox

K. R.

(2011). Parental sedentary restriction, maternal parenting style, and television viewing among 10- to 11-year-olds. Pediatrics, 128, e572-e578.

26.

Jeynes

(2011). Parental involvement and academic success. New York, NY: Routledge.

27.

Jung

J. Y.

Kim

Y. C.

Lin

W. Y.

Cheong

P. H.

(2005). The influence of social environment on Internet connectedness of adolescents in Seoul, Singapore and Taipei. New Media & Society, 7, 64-88.

28.

Kim

Lee

J. H.

(2010). Private tutoring and demand for education in South Korea. Economic Development and Cultural Change, 58, 259-296.

29.

Knight

R. G.

Williams

McGee

Olaman

(1997). Psychometric properties of the Centre for Epidemiologic Studies Depression Scale (CES-D) in a sample of women in middle life. Behaviour Research and Therapy, 35, 373-380.

30.

Kohler

Aldridge

Christensen

L. M.

Kilgo

(2012). Issues in education: Tiger moms: Five questions that need to be answered. Childhood Education, 88, 52-53.

31.

Kwack

S. Y.

Lee

Y. S.

(2007). The distribution and polarization of income in Korea, 1965-2005: A historical analysis. Journal of Economic Development, 32(2), 1-39.

32.

Larson

R. W.

Verma

(1999). How children and adolescents spend time across the world: Work, play, and developmental opportunities. Psychological Bulletin, 125, 701-736.

33.

Leone

C. M.

Richards

(1989). Classwork and homework in early adolescence: The ecology of achievement. Journal of Youth and Adolescence, 18, 531-548.

34.

Leung

F. K. S.

(2006). Mathematics education in East Asia and the West: Does culture matter? In Leung

F. K. S.

Graf

Lopez-Real

F. J.

(Eds.), Mathematics education in different cultural traditions: A comparative study of East Asia and the West (pp. 21-46). New York, NY: Springer.

35.

Madden

Bruekman

Littlejohn

K. V.

(1997). A contrast of amount and type of activity in elementary school years between academically successful and unsuccessful youth. Education Resources Information Center. Retrieved from http://eric.ed.gov/?id=ED411067

36.

Masten

A. S.

Roisman

G. I.

Long

J. D.

Burt

K. B.

Obradović

Riley

J. R.

. . .Tellegen

(2005). Developmental cascades: Linking academic achievement and externalizing and internalizing symptoms over 20 years. Developmental Psychology, 41, 733-746.

37.

Mullis

I. V.

Martin

M. O.

Foy

Arora

(2012). TIMSS 2011 international results in mathematics. Chestnut Hill, MA: TIMSS & PIRLS International Study Center.

38.

Nguyen

H. T.

Kitner-Triolo

Evans

M. K.

Zonderman

A. B.

(2004). Factorial invariance of the CES-D in low socioeconomic status African Americans compared with a nationally representative sample. Psychiatry Research, 126, 177-187.

39.

OECD. (2010). PISA 2009 results: Executive summary. Retrieved from http://www.oecd.org/pisa/pisaproducts/pisa2009/pisa2009keyfindings.htm#Executive_summary

40.

Park

(2011). Effects of home environment on internet use and dependence of children and adolescents. Selected Papers of Internet Research, 1, http://spir.aoir.org/index.php/spir/article/view/28/30

41.

Park

(2006). Factors affecting Korean students’ achievement in TIMSS 1999. In Howie

S. J.

Plomp

(Eds.), Contexts of learning mathematics and science (pp. 177-194). New York, NY: Routledge.

42.

Park

Byun

S. Y.

Kim

K. K.

(2011). Parental involvement and students’ cognitive outcomes in Korea focusing on private tutoring. Sociology of Education, 84, 3-22.

43.

Park

Leung

F. K. S.

(2004, January). Factors contributing to East Asian students’ high achievement in mathematics: The case of Korea. APEC Summit Conference, Beijing, China.

44.

Park

Y. S.

Kim

(2006). Family, parent-child relationship, and academic achievement in Korea. In Kim

Yang

Hwang

(Eds.), Indigenous and cultural psychology: Understanding people in context (pp. 421-443). New York, NY: Springer.

45.

Pears

K. C.

Kim

H. K.

Fisher

P. A.

(2008). Psychosocial and cognitive functioning of children with specific profiles of maltreatment. Child Abuse & Neglect, 32, 958-971.

46.

Powell

D. R.

Peet

S. H.

Peet

C. E.

(2002). Low-income children’s academic achievement and participation in out-of-school activities in 1st grade. Journal of Research in Childhood Education, 16, 202-211.

47.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879-891.

48.

Radloff

L. S.

(1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401.

49.

Reising

M. M.

Watson

K. H.

Hardcastle

E. J.

Merchant

M. J.

Roberts

Forehand

Compas

B. E.

(2012). Parental depression and economic disadvantage: The role of parenting in associations with internalizing and externalizing symptoms in children and adolescents. Journal of Child and Family Studies, 22, 335-343.

50.

Romer

Bagdasarov

(2013). Older versus newer media and the well-being of United States youth: Results from a national longitudinal panel. Journal of Adolescent Health, 52, 613-619.

51.

Schneider

Lee

(1990). A model for academic success: The school and home environment of East Asian students. Anthropology & Education Quarterly, 21, 358-377.

52.

Shin

(2004). Exploring pathways from television viewing to academic achievement in school age children. The Journal of Genetic Psychology, 165, 367-382.

53.

Sirin

S. R.

(2005). Socioeconomic status and academic achievement: A meta-analytic review of research. Review of Educational Research, 75, 417-453.

54.

Sorensen

C. W.

(1994). Success and education in South Korea. Comparative Education Review, 38, 10-35.

55.

Steiger

J. H.

Lind

J. M.

(1980, June). Statistically based tests for the number of common factors. Paper presented at the Annual Meeting of the Psychometric Society, Iowa City, IA.

56.

Weintraub

K. J.

Gold

(1991). Monitoring and delinquency. Criminal Behavior and Mental Health, 1, 268-281.

57.

Wickrama

K. A. S.

Lorenz

F. O.

Conger

R. D.

Matthews

Elder

G. H.

Jr. (1997). Linking occupational conditions to physical health through marital, social, and intrapersonal processes. Journal of Health and Social Behavior, 38, 363-375.

58.

Wickrama

K. A. S.

O’Neal

C. W.

(2013). Family of origin, race/ethnicity, and socioeconomic attainment: Genotype and intraindividual processes. Journal of Marriage and Family, 75, 75-90.

59.

Won

S. J.

Han

(2010). Out-of-school activities and achievement among middle school students in the US and South Korea. Journal of Advanced Academics, 21, 628-661.

60.

Wright

V. R.

Price

Bianchi

S. M.

Hunt

B. R.

(2009). The time use of teenagers. Social Science Research, 38, 792-809.

61.

Zhang

Fokkema

Cuijpers

Smits

Beekman

(2011). Measurement invariance of the center for epidemiological studies depression scale (CES-D) among Chinese and Dutch elderly. BMC Medical Research Methodology, 11(1), 74-83.