According to Confucian philosophy, education is the means for upward social mobility for both the individual and his or her family. Therefore, Chinese parents are highly involved in the lives of their adolescent children, paying special attention to their school functioning. Chinese parents also tend to be more restrictive in their parenting practices than North American parents (Kelley and Tseng, 1992; Chao, 1994) and use more psychologically controlling strategies (instead of autonomy supportive strategies; Cheung and Pomerantz, 2011). Additionally, Chinese parents are more likely to encourage children’s connectedness and less likely to encourage autonomy (Liu et al., 2005) than their North American counterparts. Chinese youth, influenced by interdependent cultural values, are less likely to view adolescence as a time of increased independence and individuation from parents (Qu et al., 2016). They also tend to anticipate gaining autonomy at an older age compared to American adolescents (Feldman and Rosenthal, 1991).

From the perspective of SDT, since the need for personal autonomy is universal (Deci and Ryan, 2000), parental autonomy support is posited to function similarly in both collectivist and individualist cultures (Marbell and Grolnick, 2013). “Even in interdependent society, when parents take children’s perspectives, allow them to voice their opinions, and provide choices, it is associated with positive outcomes” (Marbell and Grolnick, 2013, p. 89). Several cross-cultural studies have supported this view, showing that higher autonomy support predicted better psychological and academic functioning for students in the United States, as well as in collectivist Asian cultures, including China (Hasebe et al., 2004; Vansteenkiste et al., 2005; Supple et al., 2009).

However, cultural relativists maintained that certain parenting strategies may function differently in different cultural settings. Consistent with this perspective, research has shown that Chinese, Chinese American, and Asian American adolescents interpret parental control differently than their Caucasian peers. Instead of perceiving control as a violation of their individualism and privacy, they tend to view it as an expression of love and care (Lam, 2003; Cheung and McBride-Chang, 2008). Similarly, some adolescents in collectivist cultures (e.g., Ghana) regarded certain autonomy granting questionnaire items negatively, interpreting them as lack of parental support (Marbell and Grolnick, 2013). Therefore, parental autonomy granting and control may have different meanings and functions in Chinese society, where interdependence and collectivism are valued (Bao and Lam, 2008).

Additionally, non-significant relationships between autonomy granting and school functioning have also been documented among Chinese adolescents (Cheung and Pomerantz, 2011; Cheung et al., 2016), suggesting that there may be potential moderating effects that have not been investigated. Theoretically, cultural values and familial factors may interact to predict youth outcomes (Bronfenbrenner and Morris, 2006). In other words, adolescents with different levels of cultural values may benefit from different types of parenting styles. Considering the large within-group variation on cultural values, it is possible that these intracultural variations may moderate the relationship between parental autonomy granting and adolescents’ school functioning. Based on our knowledge, no published studies have examined this moderation effect among Chinese youth. The current study seeks to advance current understanding of the role of parental autonomy granting on school functioning by examining the moderating effect of cultural values.

Conformity to parental expectations is another concept related to autonomy granting. For example, Chinese adolescents showed more willingness to adhere to their parents’ academic expectations compared to their American peers (Chen and Lan, 1998). Although some Western researchers expressed concerns that too much conformity might prevent adolescents from developing social competency and autonomy (Allen et al., 1989), Chinese adolescents who have high levels of conformity to their parents actually had higher school engagement and stronger school motivation (Shen, 2011). Additionally, parent–adolescent conflict is another important family variable that has been found to impact youth outcomes in China, including school adjustment difficulties. For example, Shek (1997, 2002) postulated that conflict with parents might act as a potential stressor affecting adolescents’ emotional well-being and impacting their school adjustment.

Ecological Systems Theory (Bronfenbrenner, 1992; Bronfenbrenner and Morris, 2006) proposes that behaviors are influenced by multiple systems (individual, family, school/classroom, culture) as well as their interactions. Few studies have explored how cultural values might impact school functioning, particularly at both the individual and classroom level. One way to measure cultural values is on the continuum of independence and interdependence. According to self-construal theory (Markus and Kitayama, 1991), there are differences in the cultural orientation of the self between people in the East (e.g., China and Japan) and those in the West (e.g., United States and Canada). Individuals in the East tend to have higher interdependent self-construal, emphasizing their relationship to the social group and considering the attainment of group goals and harmony to be of importance. On the other hand, people in the West tend to have higher independent self-construal, with a strong focus on the individual’s identity and uniqueness, emphasizing the pursuit of personal goals and achievements (Cross, 1995; Singelis et al., 1999). In general, collectivism and interdependence are valued in Chinese society and have been linked to positive youth outcomes, such as lower aggression (Li et al., 2010).

Additionally, Chinese society also values a strong sense of familism. Few studies have investigated the effects of familism on students’ educational outcomes in China or how this cultural value might interact with parenting to predict adolescents’ school functioning. Research in the United States has found that Asian American youth, including those of Chinese descent, have a stronger sense of family obligation (one type of familism) than European American youth (Tseng, 2004). Furthermore, higher emphasis on family obligation was linked to stronger academic motivation, stronger educational aspirations, and more time studying among Asian American adolescents (Fuligni et al., 1999; Tseng, 2004). There are also considerable intracultural variations among adolescents, which deserve attention from researchers in order to understand individual differences in parenting and school functioning within a given context. Based on our knowledge, no published studies have examined the relationship between intracultural variations in cultural values and school functioning among Chinese adolescents.

Furthermore, no studies have examined classroom-level cultural values as predictors of adolescents’ school functioning in China. Different from their American counterparts, Chinese adolescents typically spend their school days with the same group of peers in the same classroom (instead of switching classrooms for different subjects). Consequently, a significant amount of socialization occurs in the classroom, resulting in unique classroom behavioral norms and values. Tracking students (assigning students into different classrooms) based on their grades is also a common practice in Chinese schools, potentially leading to different classroom norms even within the same school. Our study will advance previous research by examining the effects of cultural values and their interaction with parenting at both the individual and classroom level on adolescents’ school functioning.

Various students’ characteristics, such as gender, age, only-child status and self-esteem, also have been found to contribute to school outcomes. Some researchers showed that Chinese girls tend to have better academic achievement than boys (Shen, 2011; Liu et al., 2014), while others did not find these gender differences (Schwartz et al., 2013). Moreover, older Chinese students tend to have lower GPAs (grade point averages) compared to younger students (Shen, 2011), and self-esteem has been found to positively relate to school motivation and grades (Tsai et al., 2001; Shen, 2011).

Due to the one-child policy, most adolescents in China are the only-child at home. Parents place high expectations on their adolescent to succeed academically. This has presented a unique phenomenon for adolescent development. It is possible that Chinese parents have increased financial means and resources to support their only-child’s academic achievement compared to parents who have multiple children. Findings relating to the only-child status are less consistent. For instance, Chen et al. (1994) found that only-children and children with siblings did not differ significantly on normative school behaviors (moral, intellectual, physical areas), scholastic excellence, and academic achievement (in Chinese language and math) in elementary school. In a more recent study, Zhou et al. (2016) found that Chinese only-children have similar math achievement scores compared to their peers with siblings; however, they scored lower on other cognitive measures, such as processing speed and working memory. Due to these inconsistent findings, we will examine the role of individual variables on school functioning in the Chinese context.

The current study was guided by the SDT framework (Grolnick et al., 1997; Deci and Ryan, 2000) and ecological systems theory (Bronfenbrenner and Morris, 2006). Using multilevel modeling, the goal of the current study was to examine the influence of individual (e.g., demographic characteristics, self-esteem), familial (e.g., parental autonomy granting, conformity to parental expectations, parent–adolescent conflict), and cultural factors (e.g., familism, interdependence, independence) on Chinese students’ school functioning (e.g., school adjustment and grades). The second goal was to examine how cultural values moderate the relationship between parental autonomy granting and adolescents’ school functioning. Specifically, the study addressed two main research questions:

To obtain a representative sample, we collected data from a stratified random sample of 2,864 students (6th through 12th grade) from 55 classrooms, in 13 schools in Shanghai, China in 2007. The average age of the sample was 15.52 years (SD = 1.62), with the majority being female (51.5%) and the only-child in their family (86.1%). The ethical review committee at Shanghai Academy of Social Sciences approved this study. The passive consent method was used; whereby, parents were notified about the research and were given the opportunity to withdraw their children from study participation. Adolescents then gave assent to participate in this study. Students completed the survey at school, containing several measures and a demographic questionnaire (age, gender, grade, only-child status). No identifying information was collected.

We collected data using the following measures to answer our research questions related to how cultural values (independence, interdependence, and familism), as well as other individual (self-esteem, only-child status) and familial factors (parental autonomy granting, conformity to parental expectations, parent–adolescent conflict) relate to students’ school functioning.

Multilevel modeling was performed using SAS PROC GLIMMIX, due to the nested nature of the data, whereby students were nested within classrooms, classrooms nested within schools, and schools nested within school districts. Three random intercepts were included, with the assumption that there were variations at the classroom, school, and district level when covariates were controlled. All variables used in the analysis, except the three cultural values (independent and interdependent self-construal and familism), were centered at their grand means. For each cultural value, two separate variables were created to represent the within-classroom (referred to as individual-level) and between-classroom (referred to as classroom-level) effects. The within-classroom predictors were centered at the classroom mean. For instance, within-classroom independent self-construal for a student was derived from subtracting the classroom’s average score from his/her individual score. The between-classroom predictors were calculated using the average score of the students in the classroom, which were centered at their grand mean to indicate the average differences between classrooms. To examine the interaction effect, six interaction terms between parental autonomy granting and individual-level and classroom-level cultural values were created.

Table 1 shows the descriptive statistics and Table 2 displays the correlations between the variables. There were significant correlations between school adjustment and all other variables, except conformity to parental expectations. Grade rank was significantly correlated with self-esteem, parental autonomy granting, and parent–adolescent conflict. Cultural values (independence, interdependence, and familism) positively correlated with each other and with parental autonomy granting.

Results showed that male students had significantly lower school adjustment (b = -0.06, SE = 0.02, p < 0.01) and had worse grade ranking (b = -0.36, SE = 0.05, p < 0.001) than female students. Only-child status positively predicted better grade ranking (b = 0.16, SE = 0.07, p < 0.05) but not school adjustment (b = 0.03, SE = 0.03, p > 0.05). Self-esteem significantly predicted both students’ school adjustment (b = 0.04, SE = 0.01, p < 0.001) and their grade ranking (b = 0.08, SE = 0.02, p < 0.001).

For the familial variables, parental autonomy granting (b = 0.06, SE = 0.02, p < 0.001) significantly predicted students’ school adjustment but did not predict grade ranking. Parent–adolescent conflict negatively predicted both students’ school adjustment (b = -0.03, SE = 0.00, p < 0.001) and grade ranking (b = -0.04, SE = 0.01, p < 0.001). Contrary to our hypothesis, conformity to parental expectations (b = -0.09, SE = 0.04, p < 0.01) negatively predicted grade ranking. Students who had higher levels of conformity to their parents reported worse grades. For cultural values, individual-level interdependent self-construal (b = 0.02, SE = 0.01, p < 0.001) and individual-level familism (b = 0.05, SE = 0.02, p < 0.01) significantly predicted students’ school adjustment.

Due to the high correlation between independent and interdependent self-construal, we examined the moderation models separately for each predictor. Results partially supported our hypothesis concerning the interaction effect, showing that parental autonomy granting interacted with individual-level independent self-construal to predict grade rank (b = 0.06, SE = 0.02, p < 0.01; Tables 3, 4). As illustrated in Figure 1, for students who reported higher independent self-construal (1 SD above the mean), more parental autonomy granting resulted in better grade ranking. On the other hand, for students who reported lower independent self-construal (1 SD below the mean), more autonomy granting resulted in lower grade ranking. None of the other interactions were significant.

Our results showed that parental autonomy granting benefits adolescents who value a strong sense of independence. Moreover, parental autonomy granting has an adverse effect for adolescents who value independence to a lesser degree. Therefore, educational psychologists, teachers, parent educators, and other helping professionals who are working with families can assess and assist parents to adjust their parenting practices based on the cultural values of their adolescents in order to optimize youth outcomes.

Furthermore, it is concerning that adolescents who reported higher levels of conflict with their parents reported more negative academic outcomes, which has been consistently documented in the literature. One possible implication is that schools or youth programs can develop curricula to teach parents and youth conflict resolution skills to help them learn appropriate ways to handle conflicts. These important life skills will not only help adolescents improve their family relationships but also may indirectly increase their academic performance. Parenting training may also help parents understand the challenges and stress adolescents experience while also teaching them ways to help their adolescents navigate these challenges.

This study contributes to current understandings of factors influencing Chinese adolescents’ school functioning. There are several limitations that need to be acknowledged. First, all data were based on adolescents’ self-report, and shared-method variance is of concern. Using adolescents’ perception of parenting practice may still produce useful results, because research has found that there were moderate but significant correlations between Chinese early adolescents’ and their mothers’ reports of parenting (Cheung et al., 2016). However, future studies should also collect data from parents and teachers to strengthen these findings. Second, the data were cross-sectional in nature. Studies utilizing longitudinal designs are needed to investigate the effects of these variables on youth outcomes over time. Third, in addition to autonomy support, future research should examine other parenting practices, such as psychological control, which has been found to interact with autonomy granting to predict adolescents’ outcomes (Kunz and Grych, 2013). Moreover, grade rank was a one-item measure of students’ last exam rank; therefore, it might not be a comprehensive assessment of students’ academic achievement. Future studies should collect students’ grades from school records or teachers’ reports to supplement grade ranking data. Finally, although not the focus of this study, it is important to mention that student–teacher relationships significantly contribute to school adjustment among adolescents (Longobardi et al., 2016). Future studies should examine how this variable may interact with cultural values to predict youth’s academic outcomes.