Even though there are longitudinal studies examining the change of adolescents’ life satisfaction, inconsistent findings have generally been showed in the literature. For example, Shoshani and Slone (2013) found that Israeli adolescents showed lower levels of life satisfaction in their eighth grade than in the seventh grade. A similar decrease pattern in life satisfaction was also reported in Spanish adolescents (Gomez-Baya et al., 2018), Korean adolescents (Jung and Choi, 2017), and American adolescents (Martin et al., 2008). In contrast, Lewis et al. (2011) found that US middle school students’ life satisfaction significantly increased over 5 months. In a 2-year longitudinal study, Salmela-Aro and Tuominen-Soini (2010) reported that life satisfaction increased during the transition from Grade 9 to upper secondary or vocational education among Finnish adolescents. In addition, some other studies showed a stable pattern of life satisfaction during adolescence. For example, Marques et al. (2011, 2013) reported non-significant changes in life satisfaction in Portuguese adolescents over 1- and 2-year intervals, respectively. Bratt (2015) found that life satisfaction was stable over 3 years among Norwegian middle school students.

Under cross-cultural backgrounds, the results of change in adolescents’ life satisfaction have also been reported (e.g., Cosma et al., 2020; Marquez and Long, 2020). For example, based on the data from the International Health Behavior survey between 2002 and 2018 among adolescents in 36 countries, Cosma et al. (2020) found adolescents from 13 countries had a decreasing trend in life satisfaction (e.g., Greece, Austria, and Canada), and adolescents from the other 13 countries had increases in life satisfaction (e.g., Romania, Croatia, and Lithuania), whereas the adolescents from the rest 10 countries were stable in life satisfaction (e.g., Germany, France, and Italy). Based on the data from the Program for International Student Assessment, Marquez and Long (2020) analyzed the change trends of life satisfaction in 15-year-old adolescents between 2015 and 2018 in 46 countries. In their article, a decline trend of life satisfaction among adolescents in 39 out of 46 countries (e.g., France, Germany, and Russia) was reported. Adolescents in seven countries (e.g., Thailand, Spain, and Italy) had a stable level of life satisfaction between 2015 and 2018. Only South Korean adolescents showed an increase in life satisfaction during the period.

As for studies in Chinese context, most of them have been conducted among Hong Kong and Taiwanese adolescents. For example, Leung et al. (2004) found that the global life satisfaction of Hong Kong adolescents significantly declined from the beginning to the end of the 7th grade. Also in Hong Kong adolescents, Shek and Liang (2018) reported that life satisfaction showed a declining trend in a 6-year period. Jhang (2018) found that life satisfaction declined across 2 years among Taiwanese junior high school students in poverty. In addition, two inconsistent findings were reported for the adolescents from mainland China. One was conducted by Wang and Zhang (2012), and a decrease pattern in life satisfaction from Grade 7 to Grade 9 was reported. The other was performed by Nie et al. (2019) and they found life satisfaction displayed a stable pattern across 2 years among high school students.

To sum up, all of those above-mentioned inconsistent research findings suggest more longitudinal studies need to be conducted to explore the development trajectories in life satisfaction among adolescents.

Compared with studies exploring links between personality and life satisfaction in adults, there are fewer studies conducted in adolescent samples (Anglim et al., 2020). Furthermore, the majority of studies that examined the association between personality and life satisfaction in adolescents were in light of the Big Five Model. For instance, neuroticism has consistently been shown to be the strongest predictor of life satisfaction and was negatively related to life satisfaction (e.g., Suldo et al., 2015; Xu et al., 2017). Extraversion and conscientiousness were found to be positively associated with life satisfaction (e.g., Suldo et al., 2015; Weber and Huebner, 2015). As for openness, most empirical studies reported a non-significant relationship between openness and life satisfaction (e.g., Jovanović, 2019; Kim et al., 2019). Only a few found a positive link between openness and life satisfaction (e.g., Suldo et al., 2015). The relationship between agreeableness and life satisfaction was also inconsistent in the literature. Some showed a positive link between agreeableness and life satisfaction (e.g., Jovanović, 2019), whereas others reported non-significant relationships (e.g., Marcionetti and Rossier, 2016).

Despite such research findings about personality and life satisfaction in adolescents, with very few exceptions (e.g., Weber and Huebner, 2015), previous studies have mostly focused on the relationships between personality and global life satisfaction, so that little is known about the relationships between personality and specific domains of life satisfaction. For example, among US adolescents, Weber and Huebner (2015) found that functions of personality were varied in different domains of life satisfaction. Specifically, neuroticism negatively predicted all five domains of satisfaction (i.e., family life, friendships, school experiences, self, and living environment). Extraversion was positively related to satisfaction with friendships, self, and living environment. Openness was positively associated with satisfaction with family life, school experiences, and self. Agreeableness was only positively linked to satisfaction with family life. Conscientiousness was positively related to all life satisfaction domains except for satisfaction with friendships.

As shown in the above literature review, the relationship between personality (mainly in terms of the Big Five dimensions) and life satisfaction including global and specific domains has been explored mainly in western backgrounds. However, because the Big Five Model is derived from the western contexts, it may not be sufficient to capture some non-western personality features in other cultural contexts (Yang, 2006; Cheung et al., 2008; Fan et al., 2011). Accordingly, the association between personality and life satisfaction in non-western backgrounds may show very distinguished patterns from those in western backgrounds. Furthermore, the levels of global life satisfaction and the levels of satisfaction in various domains are not necessarily consistent (e.g., Jovanović et al., 2017), and the patterns of associations between global or different domain-specific satisfaction and personality are different (e.g., Lent et al., 2005; Weber and Huebner, 2015). Therefore, more empirical studies in non-western settings may further increase our knowledge about the relations of personality to both global and domain-specific life satisfaction across cultures.

For example, in the Chinese context in which Confucian philosophy is highlighted, there are a few culture-specific personality contents that have not been covered in the Big Five Model, such as harmony, face, Renqing, and family orientation (Cheung et al., 2008). Those Chinese indigenous personality attributes in adolescents have been systematically and effectively assessed by the factor of interpersonal relatedness of the Cross-Cultural (Chinese) Personality Assessment Inventory for Adolescents (CPAI-A), which is derived from Chinese settings and has been validated across eastern and western cultural settings (Fan et al., 2011; Cheung et al., 2013). In addition, the CPAI-A also includes universal independent personality factors such as social potency, emotional stability, and dependability that are correspondingly correlated with those factors in the Big Five model (Cheung et al., 2001, 2003; Lin and Church, 2004).

Empirical studies have examined the association between the personality dimensions of CPAI and life satisfaction in the Chinese samples (e.g., Chen et al., 2006; Zhang et al., 2011). For example, in a sample of Hong Kong adolescents, Ho et al. (2008) reported that certain indicators of social potency, emotional stability, and interpersonal relatedness positively predicted life satisfaction. Among junior secondary school students, Xie et al. (2016) reported positive relationships of emotional stability, dependability, and interpersonal relatedness to global life satisfaction. In addition, they also found that different domains of life satisfaction were predicted by different personality dimensions. However, all those aforementioned studies are cross-sectional design and did not provide evidence for the contributing roles of CPAI dimensions on the development trajectories of adolescents’ life satisfaction.

According to Baumrind (1991), three distinct styles could be identified based on the demandingness (also referred to as behavioral control) and responsiveness (also referred to as warmth) of parenting: authoritative, authoritarian, and permissive parenting. Authoritative parenting emphasizes clear standards and support and is characterized by being both demanding and responsive based on children’s needs and capabilities. Authoritarian parenting emphasizes order and obedience and is characterized by demanding parental behaviors but without responsiveness. Permissive parenting has more responsiveness than demandingness, and parents with this style allow children’s self-regulation and do not accentuate authority. This model of parenting styles has been widely examined across cultures in the past 40 years (e.g., Xie et al., 2016).

For many years, studies regarding the relationships of parenting styles to life satisfaction have mostly been conducted in children (Di Maggio and Zappulla, 2014). Recently, researchers have begun to notice the impact of parenting practices on life satisfaction in adolescence (e.g., Coccia et al., 2012). Empirical studies provided evidence on the links between parenting styles and adolescents’ life satisfaction across Western (e.g., Milevsky et al., 2007) and non-Western contexts (e.g., Abubakar et al., 2015).

Higher authoritative parenting was often found to contribute to higher levels of life satisfaction in both Western and non-Western cultures (e.g., Milevsky et al., 2007; Xie et al., 2016). However, the influence of authoritarian parenting was shown to be different across cultures. In the United States (e.g., Milevsky et al., 2007), Romanian (e.g., Gherasim et al., 2017), and Russian contexts (e.g., Gherasim et al., 2017), authoritarian parenting was related to lower levels of life satisfaction. However, Xie et al. (2016) reported a positive link between authoritarian parenting style and school satisfaction in a group of Chinese adolescents. In other contexts, such as Indonesian (e.g., Abubakar et al., 2015) and French (e.g., Gherasim et al., 2017) contexts, no significant relationship between authoritarian and life satisfaction was found.

As for permissive parenting style, Xie et al. (2016) and Lavrič and Naterer (2020) reported that permissive parenting was positively associated with life satisfaction in Chinese adolescents and Romanian youths, respectively. Raboteg-Saric and Sakic (2014) found that Croatian adolescents with permissive mothers and fathers are more satisfied with their life than those with authoritarian mothers and fathers. However, negative associations between permissive parenting and life satisfaction among Albanian, Bosnian, Croatian, and Kosovo youths were reported in the literature (Lavrič and Naterer, 2020).

In addition, it has been shown that the functions of maternal parenting style and paternal parenting style on offspring’s life satisfaction are different (e.g., Milevsky et al., 2007). For example, Abubakar et al. (2015) reported that Indonesian adolescents’ life satisfaction was only positively predicted by paternal authoritativeness, but not by maternal authoritativeness. Milevsky et al. (2007) found that US adolescents with maternal authoritative style had higher life satisfaction than those with maternal permissive styles, whereas no significant difference was found in life satisfaction between adolescents with paternal authoritative style and those with paternal permissive styles.

The influence of parenting styles on Chinese adolescents’ life satisfaction was supported in the literature. For example, in addition to the study by Leung et al. (2004) and Xie et al. (2016) reported that maternal restrictiveness (also known as demandingness) was negatively associated with Hong Kong adolescents’ family, self, and friend satisfaction 8 months later; and maternal concern (also known as responsiveness) was positively correlated with family, self, and friend satisfaction 8 months later. Xiang et al. (2017) reported that parental autonomy support positively predicted adolescents’ life satisfaction, and psychological control negatively predicted adolescents’ life satisfaction. In a longitudinal study with 1 year interval, Shek (2007) found that paternal and maternal psychological control negatively predicted Hong Kong adolescents’ life satisfaction. Gao et al. (2021) reported that paternal and maternal warmth positively related to adolescents’ life satisfaction 8 months later. Zhu and Shek (2021) found that trajectories of paternal and maternal behavioral control positively predicted trajectory of adolescents’ life satisfaction, and paternal and maternal psychological control negatively predicted trajectory of adolescents’ life satisfaction during the 6-year high school period. However, with the exception of Zhu and Shek’s (2021) work, most of these studies are cross-sectional or short-term longitudinal design, and the antecedents to the development trajectories of life satisfaction cannot be well examined in Chinese adolescents up to now. Meanwhile, none of them has explored functions of maternal and paternal parenting styles on different aspects of life satisfaction separately.

In response to the abovementioned research gaps for the contributions of personality and parenting styles in Chinese adolescents, the present three-wave longitudinal study aimed at the following three points. The first was to investigate the stability and change in life satisfaction in adolescents from mainland China. Based on the previous research findings (e.g., Wang and Zhang, 2012; Shek and Liang, 2018), we expected decreasing trends in life satisfaction during the present study period. Second, the influence of both universal and indigenous personalities on adolescents’ development trajectories in life satisfaction was explored. Given the findings regarding the significant links between CPAI dimensions and life satisfaction (e.g., Xie et al., 2016), we expected that higher levels of social potency, dependability, emotional stability, and interpersonal relatedness would relate to higher levels of adolescents’ life satisfaction.

Third, in terms of Baumrind’s (1991) framework of parenting styles, the contributions of parenting styles to the development trajectories of life satisfaction during adolescence were investigated as well. Consistent with previous relevant findings (e.g., Xie et al., 2016; Gherasim et al., 2017), I hypothesized that perceived authoritative and permissive parenting would positively relate to life satisfaction. Due to the inconsistent research findings in the literature (e.g., Abubakar et al., 2015; Xie et al., 2016), no specific hypothesis was made regarding the influence of authoritarian parenting on life satisfaction. In addition, given previous supporting on the different effects of paternal parenting and maternal parenting on adolescents’ life satisfaction, (e.g., Abubakar et al., 2015), I further hypothesized that paternal and maternal parenting styles would have different functions on Chinese adolescents’ life satisfaction.

Participants in this study were students from four public junior high schools in Shanghai, China. The initial sample consisted of 710 students (M_age = 11.39, SD_age = 0.53; 53.2% girls) in Grade 6. Six hundred sixty-one students in Grade 7 (93.10%) participated in the second wave of data collection, and 646 (90.10%) students in Grade 8 participated in the third wave of data collection. All participants received their personality profiles at the end of the longitudinal study as reward. Results from attrition analyses indicated non-significant differences among all study variables (i.e., parenting styles, personality, and life satisfaction) at Grade 6 between students who participated in all three waves and those who participated in only one or two waves: Wilks’ Λ = 0.97, F (16, 596) = 1.01, p = 0.45. Consent forms were obtained from both participants and their parents. The majority (90.6% fathers and 87.4% mothers) of the participants’ parents had a high school or higher degree.

A power analysis was conducted to determine whether the sample size of this study is enough to do the following analysis using G^∗power. According to previous study regarding the influences of personality and parenting styles on life satisfaction (e.g., Xie et al., 2016), the R² was set to 0.30, the power was set to 0.80 (Faul et al., 2009), alpha was set to 0.05, the number of predictors was set to 10. Results of the power analysis suggested to collect data on a sample of 389 participants, which is lower than the actual number of participants (N = 710) in this study.

The scalar invariance is a prerequisite for comparing means across different time points (Klimstra et al., 2018). In order to employ latent growth models, the scalar invariance of life satisfaction was examined. Because item parceled solution, relative to the individual item solution, resulted in less bias in estimates of structural parameters (Bandalos, 2002), items under the dimensions with more than five items (i.e., family satisfaction, school satisfaction, and self-satisfaction) were parceled into three parcels using random assignment (Little et al., 2002). Overall, evidence for scalar invariance in global and four domain-specific life satisfactions (except for self-satisfaction) was found, which implies that valid conclusions from growth curve models based on observed variables could be drawn. Therefore, latent growth models based on observed variables were employed to explore the change trajectories of global, family, friend, health, and school satisfaction.

As for the self-satisfaction, although full scalar invariance was not supported, evidence for the partial scalar invariance (i.e., with at least one of the intercepts constrained to be equal across time) can still support the comparison of latent means across time (Steinmetz, 2013). Therefore, latent growth models based on latent variables with partial scalar invariance constraints that were imposed on the confirmatory factor analysis models for measurement invariance were employed to explore the change trajectories of self-satisfaction.

The skewness (ranged from −1.37 to 0.33) and kurtosis (−0.69 to 1.25) values of the research variables indicated the normal distributions of the data (Kline, 1998). Zero-order correlational analysis was conducted. Repeated measure multivariate analysis of variance (MANOVA) was used to examine the time and gender differences of life satisfaction. Another MANOVA was used to examine the gender difference in personality and parenting styles. Latent growth models were employed to examine the change trajectories of life satisfaction with the maximum likelihood estimator. Missing data were handled in Mplus using full information maximum likelihood estimation. Because intercept and linear slope were examined with a three-wave data set in the current study, the degree of freedom (df) should be 1. However, as argued by researchers (Kenny et al., 2014; Taasoobshirazi and Wang, 2016), it may be problematic and possibly misleading to use Root Mean Square Error of Approximation (RMSEA) as a fit index when df equals 1. As such, only Comparative Fit Index (CFI) and Standardized Root Mean Square Residual (SRMR) were fit indices referred to the latent growth model analysis. Regression paths were then added into the growth models to examine the influence of personality and parenting styles on the growth patterns of life satisfaction using Mplus.

In terms of the common method bias (CMB), Harman’s single factor was used to test the potential limitation of the self-reported personality, parenting styles, and life satisfaction. The total variance for a single factor is 10.60%, suggesting an acceptable rate that is lower than 50% (Podsakoff et al., 2003).

Descriptive statistics of the study variables, their correlations, and Cronbach’s alphas are presented in Tables 1 and 2. The overall effects of time, gender, and their interactions on life satisfaction in three waves were examined through mixed repeated MANOVA. According to Cohen’s (1988) guidelines for interpreting F-test effect size (small = 0.01, medium = 0.059, and large = 0.138), only time difference, Wilks’ Λ = 0.85, F (12, 589) = 8.70, p = 0.00, η² = 0.15 was further considered. The overall effect of gender on personality and parenting styles was examined through another MANOVA. According to the Cohen’s (1988) guidelines, gender difference was not considered in the following analyses. Correlation results showed that, in general, social potency, dependability, emotional stability, interpersonal relatedness, and authoritative parenting style were positively correlated with life satisfaction. Authoritarian parenting style was negatively correlated with life satisfaction. Permissive parenting style was positively correlated with family, self, and global satisfaction.

Latent growth models were conducted to examine the linear changes in life satisfaction. The model fit indices (i.e., CFI and SRMR) indicated adequate data fit for all linear growth models except for self-satisfaction (Table 3). The latent change model for self-satisfaction was not convergent. Therefore, no model fit indices for self-satisfaction were reported. Results revealed that life satisfaction (except for self-satisfaction) had medium-high initial levels that significantly decreased over time. The significant negative intercept-slope correlations in friend, school, and global satisfaction indicated that adolescents who initially scored higher reported a sharper decrease with time in the four aforementioned life satisfactions.

Five models were then tested with the growth patterns of life satisfaction regressing on personality and parenting styles to examine the predicting roles of personality and parenting styles in global and four domain-specific life satisfactions except for self-satisfaction (see Table 4). All five models indicated good fit index with χ²/df smaller than 5, CFI larger than 0.90, and RMSEA smaller than 0.08, SRMR smaller than 0.08 (Hu and Bentler, 1999; Kline, 2005).

Results indicated that dependability, interpersonal relatedness, and paternal and maternal authoritative styles positively predicted the initial level of family satisfaction, whereas paternal and maternal authoritarian style negatively predicted the initial level of family satisfaction. As for friend satisfaction, social potency, interpersonal relatedness, and paternal authoritative style positively predicted adolescents’ initial level of friend satisfaction, whereas paternal authoritarian style negatively predicted the initial level of friend satisfaction. The slope of friend satisfaction was positively predicted by emotional stability and paternal authoritarian style, whereas negatively predicted by social potency. The initial level of health satisfaction was positively predicted by social potency and emotional stability. The initial level of school satisfaction was positively predicted by dependability, emotional stability, interpersonal relatedness, and paternal authoritative style. The initial level of global satisfaction was positively predicted by dependability and interpersonal relatedness. The slope of global satisfaction was positively predicted by emotional stability, whereas negatively predicted by interpersonal relatedness.

These results indicated that social potency, emotional stability, and paternal authoritarian style moderated the growth of friend satisfaction; emotional stability and interpersonal relatedness moderated the growth of global satisfaction. I then conducted simple slope analyses to examine the moderating effects. As presented in Figure 1, students with higher social potency (+ 1 SD), lower emotional stability (−1 SD), or perceived lower paternal authoritarian style (−1 SD) displayed sharper decreasing trends in friend satisfaction than did those with lower social potency (−1 SD), higher emotional stability (+ 1 SD), or perceived higher paternal authoritarian style (+ 1 SD) (see Figures 1A–C, respectively). Students with higher interpersonal relatedness (+ 1 SD) or lower emotional stability (−1 SD) displayed sharper decreasing trends in global satisfaction than did those with lower interpersonal relatedness (−1 SD) or higher emotional stability (+ 1 SD) (see Figures 1D, E, respectively).

The regression results of self-satisfaction at three time points on parenting styles and personality were reported in Table 4. Results showed that social potency, emotional stability and paternal authoritative style positively predicted self-satisfaction at Time 1 and totally complained 25% variance. Self-satisfaction at Time 2 (R² = 0.24) was positively predicted by paternal authoritarian style, social potency, and emotional stability, whereas negatively predicted by interpersonal relatedness. Self-satisfaction at Time 3 (R² = 0.19) was positively predicted by social potency, dependability, and emotional stability.

Consistent with our hypothesis, adolescents displayed decreasing trends in both global and domain-specific life satisfactions as the school year progressed. This result reinforced previously reported evidence from Hong Kong (e.g., Shek and Liang, 2018), Taiwan (e.g., Jhang, 2018), mainland China (e.g., Wang and Zhang, 2012), and other countries (e.g., Marquez and Long, 2020). Such decreasing trends may be due to the high expectations (Shoshani and Slone, 2013) and schoolwork pressure (Cosma et al., 2020) during the middle school years in this highly formative era. The inconsistent findings with the stable pattern shown in mainland Chinese high school students (Nie et al., 2019) may indicate that the developmental trajectories in life satisfaction during early adolescence and late adolescence are different (e.g., Chen, 2020).

The results indicated that four CPAI dimensions significantly predicted both global and domain-specific life satisfaction in Chinese adolescents. The functions of these personality factors varied across different domains of life satisfaction. Such results supported the view that the global and domain-specific life satisfaction are distinguishable and may have different relationships with personality (Weber and Huebner, 2015).

Specifically, consistent with previous studies (e.g., Xie et al., 2016), adolescents who scored higher on social potency reported higher levels of satisfaction in friend and health domains at the initial stage and higher self-satisfaction at all three time points; no significant influence of social potency on the global life satisfaction was viewed (e.g., Chen et al., 2006). Unexpectedly, I found that individuals with higher levels of social potency experienced shaper decrease in friend satisfaction over time. Nevertheless, this result is similar to Li et al.’s (2019) research findings that Chinese adolescents with higher levels of social potency experienced increased loneliness from Grade 6 to Grade 8. It is possible that individuals who are extravertive and like to take the lead in making decisions would maintain larger social networks and thus have higher levels of friend satisfaction at the initial stage (e.g., Moran and Weiss, 2006; Harris and Vazire, 2016). However, as adolescents grow up, “good friends” are not only those who spend time together, but more those who can chat about beliefs, values, and ideologies with each other (Brown, 2004). Thus, it is conceivable that sharper decreasing trends in friend satisfaction for those with higher levels of social potency would be viewed during the research period.

Dependability was found to positively predict the initial levels of family, school, and global life satisfaction, which is in agreement with previous research (e.g., Chen et al., 2006; Xie et al., 2016). For example, Xie et al. (2016) reported positive associations between dependability and the global life satisfaction as well as the satisfaction in family and school domains. Dependable people who are high in responsibility and have clear life meaning tend to plan ahead and pursue meaningful goals (Cheung et al., 2008), and thus would be more likely to be relied on and have higher life satisfaction (Chen et al., 2006; Ho et al., 2010). Our current finding that dependability only predicted self-satisfaction at Time 3 may indicate that traits like responsibility and meaning in life are more important in determining one’s self-satisfaction at higher grades comparing to their contributions at lower grades.

Given the overlapping between dependability and conscientiousness (Cheung et al., 2001), the findings of present study are also, to some extent, consistent with previous studies in which a positive association between conscientiousness and life satisfaction was reported (e.g., Suldo et al., 2015; Weber and Huebner, 2015). For example, Weber and Huebner (2015) found that conscientiousness was positively related to satisfaction in family life and school experiences domains in US adolescents. With a sample of US high school students, Suldo et al. (2015) reported that conscientiousness positively predicted global life satisfaction.

Higher levels of emotional stability were associated with higher initial levels of satisfaction in health and school domains as well as higher self-satisfaction at all three time points, which substantiates previous empirical studies (e.g., Xie et al., 2016; Tian et al., 2017; Cohrdes and Mauz, 2020). For example, Tian et al. (2017) reported a positive association between emotional stability and subjective well-being in school (i.e., school satisfaction and school affect) in Chinese adolescent students. Although adolescents generally displayed a decrease trend in life satisfaction in the current study, it was found that highly emotionally stable individuals declined less from Grade 6 to Grade 8. These results suggested that emotional stability could serve as a protective factor that helps adolescents cope with stressful situations (e.g., Ho et al., 2013).

The positive predictions from interpersonal relatedness on the initial levels of adolescents’ family, friend, and school satisfactions are consistent with previous findings in Chinese adolescents. For example, Xie et al. (2016) found significant and positive influence of interpersonal relatedness on family, friend, and school satisfaction in a sample of junior secondary school students. These results provided empirical support for the idea that the interpersonal relationship orientation is associated with the quality of social relationships (Cheung et al., 2001) and leads to the satisfaction in interaction with others (Xie et al., 2016). Higher levels of interpersonal relatedness were also related to higher initial levels in global life satisfaction. This result is in agreement with previous research in which a positive association between interpersonal relatedness and adolescents’ global life satisfaction was reported (e.g., Ho et al., 2008; Xie et al., 2016). Such findings supported the idea that the emphasis on relationships with others can help to foster individuals’ life satisfaction (Blanca et al., 2018).

However, contrary to our hypothesis, individuals with higher levels of interpersonal relatedness experienced a shaper decrease in global life satisfaction over time and lower levels of self-satisfaction at Time 2. This result suggested that the tendency to maintain useful ties, avoid conflict, and contribute to the collective over the individual goals (features of interpersonal relatedness) may sometimes have negative consequences because individuals may give up their own interests and maintain even harmful relationships (Kim et al., 2018). The contribution of interpersonal relatedness to Chinese adolescents’ life satisfaction highlighted the importance of considering the indigenous personality when exploring factors that influence adolescents’ development in the specific cultural context.

Regarding the variations in life satisfaction as a function of maternal and paternal parenting styles, the present results overall indicated that parenting styles were related to domain-specific life satisfaction. However, in contrast with literature showing positive link between global life satisfaction and parental authoritative style in Western and non-Western cultures, and negative link between global life satisfaction and parental authoritarian style in Western cultures and positive link between global life satisfaction and parental authoritarian style in Chinese Context (e.g., Milevsky et al., 2007; Xie et al., 2016; Gherasim et al., 2017), parenting styles in the present study did not influence the global life satisfaction when the effect of adolescents’ personality is considered. One possible reason is that most previous studies did not consider the influence of individuals’ personality when exploring the influence of parenting styles on life satisfaction (e.g., Abubakar et al., 2015). Results in the current study implied that the influence of parenting on student development becomes less important as individuals’ demand for independence and autonomy increase during adolescence (Inguglia et al., 2015), and that personality traits were better predictors of life satisfaction than situational factors (Diener et al., 1999).

In the present study, adolescents who perceived their mothers or fathers as more authoritative reported higher family satisfaction at the initial stage, and those who perceived their mothers or fathers as more authoritarian reported lower family satisfaction at the initial stage. These findings are consistent with the previous studies that were conducted in Western contexts, showing that authoritative parenting style is associated with happier family life and authoritarian is related to lower family satisfaction (e.g., Givertz and Segrin, 2014). Such similar results may indicate that whether in Western or in non-Western contexts, authoritative parenting is preferred than authoritarianism in achieving a child perceived happy family life. An alternative explanation is that the present sample was recruited from Shanghai that has a relatively high exposure to the Western cultural context, and thus would display a similar relationship pattern with the findings in Western context. It was also noted that on family satisfaction, paternal and maternal parenting styles have a similar function.

In addition, paternal authoritative style was found to be associated with higher levels of friend, school, and self-satisfaction at the initial stage. These results were in accordance with the findings of previous studies in which authoritative style was shown to play an important role in adolescents’ interpersonal management (e.g., Shalini and Acharya, 2013), avoidance of social withdrawal (e.g., Sandhu and Sharma, 2015), school adjustment (e.g., Pinquart and Kauser, 2018), and general self-efficacy (e.g., Tam et al., 2012).

It should be noted that adolescents who perceived lower paternal authoritarian style reported higher levels of friend satisfaction at the initial stage and sharper decrease trends over 2 years. Although this result was unexpected, similar observations were made by Di Maggio and Zappulla (2014). In their study, Italian high school students (with the similar age to the participants in the present study at Time 3) who perceived their fathers as authoritarian reported higher levels of satisfaction with friends than those with authoritative fathers.

It was also found that adolescents who perceived higher paternal authoritarian parenting would be more likely to be satisfied of themselves. These findings hinted that lower authoritarian parenting may not always be a good thing for Chinese adolescents’ life satisfaction (Chao, 1994). Specifically, unlike in Western cultures where authoritarianism is construed as rejection, in the Chinese context that Confucian philosophy is highlighted, the authoritarian parenting may be perceived as training and monitoring. Chinese parents who engage the authoritarian parenting would supervise, monitor, and guide their children’s behaviors to fit the rules and standards of the society. Such guidance and monitoring are beneficial for the formation of children’s adaptive working model of self, such as seeing oneself as lovable, competent and worthy (Li et al., 2016). This result is also consistent with previous research which showing that parental behavioral control associated with higher life satisfaction among Chinese adolescents (e.g., Zhu and Shek, 2021).