Perceived School Performance, Life Satisfaction, and Hopelessness: A 4-Year Longitudinal Study of Adolescents in Hong Kong

Chinese society’s strong emphasis on school success is well known around the world. There is a traditional Chinese belief that “to be a scholar is to be the top of society” (xue er you ze shi, 學而優則仕). Driven by this thinking, Chinese students are expected to study hard and achieve academic excellence (Shek and Chan 1999). Such beliefs have created a highly selective and competitive education system in Hong Kong (Kwan 2010), which drives students to study hard and perform well to achieve good school results in the harsh public examination system. Under the influence of Confucianism and related Chinese philosophies, academic performance and school conduct are two closely related aspects of school performance (Ali 2011). In addition to achieving good grades, Chinese children are expected to have good school conduct (Li and Fung 2014).

In Hong Kong, students experience fierce school competition from early childhood, as their performance at each learning stage determines whether they will be accepted to study at a distinguished school in the next stage. Among different age groups, secondary school students experience the greatest stress because they must pass the Hong Kong Diploma of Secondary Education Examination to get into university, which is a conventional route to future success. In particular, the 3-3-4 scheme introduced in the academic structure reform (i.e., the new senior secondary school curriculum) implemented in Hong Kong in 2006 has increased the burden on students. Under the new system, students have only three instead of 4 years of senior secondary education to prepare for the university entrance examination. Students must work much harder to adapt to this new initiative and achieve academic excellence in a shorter time. Against this background, students are very likely to suffer from more emotional problems because of their stressful school experiences (Au and Watkins 1997; Long et al. 2012).

Researchers have defined hope as an individual’s overall perception that he or she can meet his or her goals (e.g., Snyder et al. 1991). Other researchers have defined hopelessness as a negative expectation toward oneself and the future (e.g., McLaughlin et al. 1996). Individual hopelessness may be reflected in unfavorable expectations about one’s future life (Beck et al. 1974). Abramson et al. (1989, p. 359) defined hopelessness as “an expectation that highly desired outcomes will not occur or that highly aversive outcomes will occur coupled with an expectation that no response in one’s repertoire will change the likelihood of occurrence of these outcomes.” Researchers have examined whether hope changes during adolescence. McKnight et al. (2002) reported that positive affect decreased over time in adolescents. However, Bolland (2003) found that age was not related to a sense of hopelessness in male adolescents, whereas age and hopelessness were negatively correlated for female adolescents. In view of the inconsistent research findings, there is a need to further examine this issue.

Daily events or chronic stress have considerable influence on individuals’ lives and, in particular, stressful life events cause intrapersonal problems for adolescents (McKnight et al. 2002). School performance is one of the precursors of hopelessness because school learning is the major developmental task of adolescents in Chinese culture (Huang 2014). As school performance is almost the sole standard for predicting future success, school performance is of great importance for Chinese adolescents and those surrounding them, such as parents and teachers (Huang 2014).

Adolescents’ lives are filled with a variety of positive and negative stressors (McKnight et al. 2002), the most important of which may be school performance because experiences at school are the most typical and influential sources of stress for adolescents (Ash and Huebner 2001). Perceptions of personal academic capabilities and academic self-image influence student well-being (Van Petegem et al. 2007) and school conduct is positively related to perceived academic performance (Roeser and Eccles 1998). If students worry about and have negative perceptions of their school performance, it will create negative affect (Long et al. 2012). In Chinese societies, adolescents usually display hopelessness if their school performance is unsatisfactory (Shek and Li 2014).

Poor perceptions of academic performance and school conduct are related to depressive feelings and negative psychological adjustment (Roeser and Eccles 1998). Studies have documented the relationship between perceived school performance and hopelessness. Typically, successful school experiences help students to develop a positive sense of well-being, whereas school failures not only increase the feeling of hopelessness, but also contribute to poorer perceptions of their academic self-image and school performance (Au and Watkins 1997). Pekrun et al. (2009) also found that scholastic ability was negatively related to hopelessness. Perceived school performance has been related to self-esteem and locus of control, and found to predict students’ emotions and behavioral problems such as depression and suicide attempts (Richardson et al. 2005). Some studies have found that students who perceive that they have little control over their examination performance are prone to high levels of hopelessness and unhappiness (e.g., Burić and Sorić 2012). It can be argued that when students feel incompetent in school, they can easily lose confidence and feel hopeless about moving forward into the future. Chinese students may be particularly likely to feel hopeless if they have poor perceptions of their school performance because of the strong emphasis on academic achievement in Chinese culture.

As an important construct of positive psychology, life satisfaction has been regarded as a stable indicator of personal well-being (Gilman et al. 2000) and psychological development (Goldbeck et al. 2007) in adolescence. Life satisfaction is a global cognitive evaluation of an individual’s life as a whole (Suldo and Huebner 2004). McCullough et al. (2000) found that the majority of adolescents in secondary schools had moderately high levels of life satisfaction. Some scholars have argued that age does not influence adolescent life satisfaction and the trend of life satisfaction is moderately stable in adolescence (e.g., Ash and Huebner 2001; Suldo and Huebner 2004). However, Gilman and Huebner (2003) argued that life satisfaction as a developmental phenomenon is not static but volatile for adolescents, because adolescence is one of the most difficult periods in human development, with intense cognitive, emotional, and social changes (Arnett 1999). Goldbeck et al. (2007) found that adolescent life satisfaction decreased over time in many countries, including Germany, Australia, and Poland. Michel et al. (2009) found that adolescents generally had a poorer quality of life than children, based on data from 12 European countries, including the United Kingdom, France, the Netherlands, and Sweden.

Ash and Huebner (2001) proposed that life experience and life satisfaction have a transactional relationship. For example, positive daily events lead to high life satisfaction (McCullough et al. 2000) and negative perceptions of school experience usually lead to low life satisfaction (Gilman and Huebner 2006). Empirically, Chow (2008) found a positive relationship between academic achievement and life satisfaction in adolescents. Suldo et al. (2008) found that problematic school behavior was associated with low levels of school life satisfaction and global life satisfaction.

In addition to the positive relationship between perceived school performance and life satisfaction, life satisfaction and psychological adjustment are interrelated (Gilman and Huebner 2006). Gilman and Huebner (2006) found that adolescents with high life satisfaction had virtually no psychological symptoms, whereas adolescents with low life satisfaction often had mental health problems of a clinical nature. Lewinsohn et al. (1991) found that low life satisfaction predicted subsequent psychological disorders. Suldo and Huebner (2006) found that high life satisfaction was related to better adaptive psychosocial functioning and fewer emotional and behavioral problems. Students with high global life satisfaction tended to experience less intrapersonal distress and a greater sense of hope than students with low life satisfaction (Gilman and Huebner 2006). Thus, high life satisfaction can be regarded as a potential protective factor against hopelessness (Heisel and Flett 2004).

McKnight et al. (2002) showed that adolescents who experienced more stressful events were less satisfied with life and consequently were more likely to have maladaptive internal responses (e.g., anxiety and depression). Sun and Shek (2010, 2012, and 2013) reported that adolescents with high levels of positive youth development were more satisfied with life and had fewer behavioral problems. These findings suggest that life satisfaction functions as a mediator in the relationship between positive youth development and problem behavior. However, there is limited scientific evidence on the mediating role of life satisfaction in the association between perceived school performance and hopelessness. Because school is a means of climbing the social ladder, it influences adolescent life satisfaction and serves as an important determinant of their hope about the future.

As a transitional stage in life, adolescence is filled with difficulties and challenges (Goldbeck et al. 2007). Adolescents may display different patterns of life satisfaction from adults. However, studies of life satisfaction have mainly focused on adults, even though life satisfaction is regarded as an important construct for people of all age groups (Silva et al. 2014). The same limitation is true for studies on hope, most of which have used adult samples (Gilman and Huebner 2006). To fill this research gap, we recruited a large number of Grade 7 secondary school students (Secondary 1) in 2009 and tracked them until Grade 10 (Secondary 4) to investigate whether adolescent perception of school performance is a determinant of hopelessness, with life satisfaction as a mediating factor. A review of the scientific literature revealed that no studies have yet considered the relationships between perceived school performance, life satisfaction, and hopelessness.

The goal of the study was to explore the developmental pathways of students’ perceptions of school performance, life satisfaction, and hopelessness, and the underlying associations between them. School performance was conceived as perceived academic performance and conduct in school. Several hypotheses were proposed:

The data were derived from an on-going 6-year longitudinal study investigating adolescent development in Hong Kong. This study used the data collected at Time 1, Time 2, Time 3, and Time 4. Twenty-eight public secondary schools were randomly selected and invited to participate in this large-scale project. The student participants’ details are reported in Table 1. The data from 2427 students (72.9 % of the 3328 students who completed the first assessment) who completed all four assessments were used in the analyses. Comparison of those participants who only took part in the first assessment and those who completed all assessments revealed that more boys than girls dropped out of this longitudinal study (Pearson x ²(1) = 41.073, p < .001). Regarding the main variables in this study, there was no significant difference between the life satisfaction ratings of drop-out students and retained students (t = −1.629, p > .05), but drop-out students had significantly lower perceptions of school performance (t = −7.104, p < .001) and higher levels of hopelessness (t = 5.284, p < .001).

The data collection procedure was the same at each time point. Written informed consent was obtained from the participants’ parents before data collection. Passive informed consent was also obtained from the participants. Trained research assistants were responsible for the administration of the questionnaires. They provided students with standardized instructions (e.g., research objectives, voluntary participation, and data confidentiality), answered their questions about the investigation, and maintained classroom discipline. The same self-report questionnaire (approximately 30 min) was used at each time point.

To avoid the potential problem of collinearity, the tolerance and variance inflation factor (VIF) values were examined. To measure the changes in perceived school performance, life satisfaction, and hopelessness at each time point, one-way repeated measures analyses of variance (ANOVA) were performed in SPSS 21. To test the mediating effect of life satisfaction on the relationship between perceived school performance and hopelessness longitudinally, the two-step modeling approach recommended by Anderson and Gerbing (1988) was adopted. The measurement model and structural model were successively tested using LISREL 8.7 (Jöreskog and Sörbom 2004). Due to the non-normal data distribution, the robust maximum likelihood (RML) estimation method was used. Various indicators were used to assess the goodness of fit of the model, including the comparative fit index (CFI; Bentler 1990), incremental fit index (IFI; Bollen 1989), and non-normed fit index (NNFI; Tucker and Lewis 1973). Goodness-of-fit requires that the values of these indicators are above .95 (Hu and Bentler 1999). Additionally, a value below .80 (Hu and Bentler 1999) for the root mean square error of approximation (RMSEA; Steiger 1990) is considered a good fit. Following the mediation test across the four time points, the mediating effect of life satisfaction was examined at each time point according to Baron and Kenny’s approach (1986). Bootstrapping was performed as this method renders measurement more accurate by replacing the original sample size n multiple times (Preacher and Hayes 2008). In this study, the resampling was repeated 10,000 times with the calculation of the bias-corrected (BC) confidence intervals (CIs), suggesting that the lower and upper bounds of the interval would be the 250th and 9751st estimates.

The mean values and standard deviations of perceived school performance, life satisfaction, and sense of hopelessness at the four assessments are reported in Table 2. According to Belsley et al. (1980), a tolerance value below 0.10 or a VIF above 10 implies multi-collinearity. The tolerance values in this study ranged from .854 to .875 and the VIF ranged from 1.171 to 1.142, thus multi-collinearity was not a problem.

To examine the changes in perceived school performance, life satisfaction, and hopelessness over time, one-way repeated measures ANOVAs were conducted. Perceived school performance (F = 220.883, p < .001, η² = .083) and life satisfaction (F = 119.957, p < .001, η² = .047) linearly decreased over time, whereas hopelessness showed a quadratic increasing trend in the same period (F = 7.250, p < .01, η² = .003). The effect sizes of the differences in the above three variables generally ranged from small to medium. Bonferroni post hoc comparisons indicated that adolescents had better perceptions of their school performance and higher life satisfaction at lower grades than at higher grades. Adolescent hopelessness significantly increased in early adolescence, although this trend remained relatively stable in middle adolescence (see Table 3). The findings provide support for hypotheses 1a and 1b.

The measurement model included 12 latent variables (perceived school performance, life satisfaction, and hopelessness at the four time points) and 52 observed variables. The measurement model showed a good model fit: x ²(1208, n = 2427) = 11,713, p < .001; CFI = .96; IFI = .96; NNFI = .95; and RMSEA = .060 (90 % CI: 0.059–0.061). All of the factor loadings of the observed variables on the latent variables were significant. The factor loadings ranged from .37 to .81 for perceived school performance, .54 to .89 for life satisfaction, and .57 to .91 for hopelessness. Across the four time points, perceived school performance, life satisfaction, and hopelessness were significantly correlated with each other (see Table 4). These findings provide support for hypotheses 2, 3, and 4.

The structural model (see Fig. 1) fitted the data very well: x ²(1247, n = 2427) = 12,023, p < .001; CFI = .96; IFI = .96; NNFI = .95; and RMSEA = .060 (90 % CI: 0.059–0.061). All of the autoregressive paths and hypothesized cross-lag paths were significant, suggesting that life satisfaction mediated the relationship between perceived school performance and hopelessness longitudinally. We also found significant indirect effects of perceived school performance (Time 1) on hopelessness (Time 4) through the mediator of life satisfaction at Time 2 [β = −.005, (BC 95 % CI: −.010, −.001)] and Time 3 [β = −.007, (BC 95 % CI: −.014, −.002)], after controlling for life satisfaction and hopelessness in the previous wave. Table 5 also shows that the mediating role of life satisfaction was established in all four assessments. “Zero” was not included in the bias-corrected confidence intervals, indicating that the indirect effects were significant. After controlling for life satisfaction, the direct effects of perceived school performance on hopelessness were still significant at each time point (ps < .001). This result implies that life satisfaction functioned as a partial mediator of the relationship between perceived school performance and hopelessness. Overall, adolescent life satisfaction mediated the relationship between perceived school performance and hopelessness in both a longitudinal and a cross-sectional manner. The findings provide support for Hypothesis 5.