Longitudinal Effects of Examination Stress on Psychological Well-Being and a Possible Mediating Role of Self-Esteem in Chinese High School Students

Abstract

Through using a latent growth curve model (LGCM), the present study investigated longitudinal relationships between examination stress, self-esteem, and psychological well-being in Chinese high school students. This paper presents results of a three-wave longitudinal study among 248 Chinese high school students who were followed over the course of one semester. The students completed questionnaires about once every 2 months from the beginning to the end of a school semester for a total of three questionnaires including the shorten version of Academic Stress Scale, the Rosenberg Self-Esteem Scale and the Chinese version of Psychological Well-Being Scale. The results obtained from latent growth curve modeling (LGCM) showed that the initial examination stress level negatively predicted the students’ initial level of psychological well-being. Also, changes in examination stress over time negatively predicted changes in psychological well-being. In addition, self-esteem can mediate the effects of examination stress on psychological well-being: first, initial level of examination stress can influence the initial level of psychological well-being via self-esteem; second, examination stress at Time 1 predicted psychological well-being at Time 3 mediated by self-esteem at Time 2. These findings contributed the theoretical explanation about the effect of stress in damaging psychological well-being and the mediating mechanism of self-esteem. There are also some practical implications on improving psychological well-being among the high school students through reducing the levels of examination stress.

Introduction

Psychological well-being (PWB) refers to that individuals have a positive self-regard, positive relationships with others, purpose in life, and are able to develop own potential through personal growth and choose/create environments that are suitable to one’s psychic condition, self-determination, and independence (Ryff and Singer 2008). With the development of positive psychology, psychological well-being has been recognized as a highly relevant research topic in the area of social psychology because it has been found to contribute to a range of critical outcomes in life, including increased social support, greater life satisfaction, improved physical health (Ryff and Singer 2008), and better academic performance (Norvilitis and Reid 2012; Topham and Moller 2011).

Among the factors that affect psychological well-being, perceived stress has received considerable attention. Lazarus and Folkman (1984) defined stress as “a particular relationship between the person and the environment that is appraised by the person as taxing or exceeding his or her resources, and as endangering his or her well-being.” In order to address the effects of stressors on well-being as well as their possible mediators, Lazarus and Folkman (1984) proposed a stress process model. In this model, they emphasized two central processes, cognitive appraisal and coping, as critical mediators between stress and immediate or long-range outcomes. When encounter stressful event that is relevant to their well-being, people will make cognitive appraisals. They first evaluate whether the encounter (stressful event) is harmful or beneficial for their well-being (first cognitive appraisal) and then evaluate whether there are resources to overcome harm or to improve the benefit (second cognitive appraisal) (Folkman et al. 1986). For the second process, coping is defined as a personal effort to manage demands, which is appraised as exceeding the person’s resource. When the available resources can not satisfy the demands, individual may develop strategies to search available resources that would help to decrease their stress (Folkman et al. 1986). How people appraise events can make a difference in their coping strategies, and then result in various outcomes. However, when stress continues increasing and coping resources still cannot meet demands, the buffering psychological resources would be threatened (Lee et al. 2013). Based on this model, previous studies examining relationships between stress and well-being indicated that social stress condition and coping resources were important for well-being (e.g., Eisenbarth 2012; Chao 2011).

Furthermore, researchers have identified several resources that may mediate the relationship between stress and well-being including psychological resource and social resource according to the stress process model (Goode et al. 1998). In previous literature, self-esteem has been employed as a robust psychological resource to establish a positive view of the self (e.g., Hobfoll 1989). When confronted with stressful events, people who have positive view of themselves are less likely to experience overwhelmed than people who do not, because people with positive self-view would see themselves as being able to cope with various problems (DeLongis et al. 1988). Without higher self-esteem, stressors accumulate in life can overwhelm individual’s available coping resources to support mental health (Taylor and Stanton 2007). Moreover, Hobfoll (1989) further developed the stress process model and more directly emphasized that people strive to reserve, protect, and build resources for keeping psychological health. It is the potential or actual loss of available resources, such as self-esteem, resilience and self-efficiency that threatening to them and leading to unhealthy psychological status (Hobfoll 1989). Therefore, it is reasonable to expect that stress is closely associated with psychological well-being, and that self-esteem may play some intervening roles in the relationship between stress and psychological well-being.

Adolescence, as an important stage in the development of personality and learning, is also an important period for the development of psychological well-being. Being one of the most important factors that affect adolescents’ psychological well-being, examination stress is one of the most important kinds of academic stress perceived by teenagers, which has received considerable attention. It is well known that in some Eastern cultures (e.g., China), examination performance is a key criterion for evaluating students and an important developmental indicator of successful school lives among adolescents. That is, in order to become a good student, individuals should struggle to achieve a good grade on important examinations, so that stress may easily stem from the process of academic examinations preparation. Previous studies have shown that changes in examination stress are significantly correlated with changes in levels of anxiety as measured by salivary cortisol (Singh et al. 2011). Similarly, Li et al. (2010) demonstrated that examination stress has obvious impacts on emotions of individuals. Therefore, examination stress can be expected to have a negative impact on psychological well-being in adolescent students. However, despite a large number of studies focusing on the relationship between examination stress and positive affect (Srivastava 2015), only a few studies have examined the patterns of the relationships between changes in the levels of examination stress and psychological well-being over time, and almost no studies have examined the possible mediating mechanism of examination stress on psychological well-being.

Drawing on previous literature about the relationship between stress and psychological well-being, the current study employed a longitudinal study design to explore the patterns of relationship between changes in the levels of examination stress and psychological well-being over time and further explored the possible mediating role of self-esteem in the relationship between examination stress and psychological well-being in adolescent students.

Theoretical Framework and Hypotheses

Examination Stress and Psychological Well-Being

Examination stress is physical and mental tension caused by the imbalance between an actual or perceived requirement and coping ability when people are in the process of adapting to a specific examination. Examination stress focuses exclusively on examinations, and is different from “academic stress” which refers to a wider range of school activities (Putwain 2007). There is an abundance of evidence to suggest that changes in levels of examination stress co-occur with a number of physiological and psychological changes, including changes in emotion (Singh et al. 2011; Weekes et al. 2006), cognitive function (Kofman et al. 2006; Lewis et al. 2008), even personality characteristics (Rohrmann et al. 2003; Wadee et al. 2001). For example, Weekes et al. (2006) have observed that examination stress is a significant trigger of elevations in negative mood states such as anxiety which is associated with a rise in cortisol level. Previous researches also showed that examination stress was influenced by many factors, such as gender (e.g., Lewis et al. 2008; Rohrmann et al. 2003), age (e.g., Yang 2016), and school environmental factors (e.g., Ortuño-Sierra et al. 2015). Although some studies showed no gender difference in examination stress in laboratory experiments (e.g., Lewis et al. 2008; Rohrmann et al. 2003), studies in real school situation demonstrated significant gender differences on examination stress, with female students being more likely to report higher levels of examination stress (e.g., Eman et al. 2012). In terms of age, the students from 10th grade to 12th grade faced different levels of examination stress in Chinese school environment. Yang (2016) found that the 12th grade students experienced the highest stress for the coming college entrance examination. Examination stress did not show differences between the 10th and 11th grade in school environment (Yang 2016). It is interesting that in the real school environments, students even appear to feel examination stress as soon as they entered the school. The closer to important exams, the more stress they are likely to have (Mohapatra et al. 2012). In sum, it has been well established that examination stress widely exists and may have significant effects on various aspects of daily lives in adolescent students (Srivastava 2015).

Examination stress, as a type of academic stress often experienced by adolescent students, is one cause of perceived stress. Based on the stress process model as well as previous research, examination stress could be expected to be associated with psychological well-being through at least two ways. First, the overall level of examination stress can be expected to have a negative impact on psychological well-being. Martin and Ickovics (1987), who conducted a longitudinal study on army wives, found that both military life stress and marital stress have important independent negative effects on the level of general psychological well-being. Similarly, Edwards (2004) found that a higher level of stress could predict a lower level of well-being. Second, changes in examination stress overtime can be expected to be negatively associated with changes in psychological well-being. Previous researchers have found that reduced perceived stress tends to increase well-being (Schwartz et al. 1992). Similarly, Rusch et al. (2014) found that a decrease in stress predicted a better sense of well-being at follow-up. Examination stress usually comes from the preparing for academic examinations. It is also a typical type of stress for youths and can be expected to have a deleterious effect on an adolescent’s well-being. Therefore, based on these previous studies, we developed two hypotheses:

Hypothesis 1:

The overall level of examination stress is a significant predictor of the level of psychological well-being in adolescent students.

Hypothesis 2:

Changes in examination stress across time negatively predict changes in psychological well-being in adolescent students.

Mediation Role of Self-Esteem

Self-esteem reflects a person’s overall subjective emotional evaluation of his or her own worth. It is a judgment of oneself as well as an attitude toward the self. Self-esteem encompasses beliefs (e.g., “I am competent”) and emotions such as triumph, despair, pride, and shame (Lopez and Snyder 2004). According to the literature reviewed above, self-esteem as a psychological resource (Hobfoll 1989), should play an important role between stress and well-being. Notably, some research confirmed a moderating role of self-esteem between stress and well-being and revealed that having high level of self-esteem alleviated the negative effects of acculturative stress against psychological well-being (Gardner and Parkinson 2011; Kim et al. 2014). Some other research demonstrated that self-esteem can also be an important predictor of perceived stress (Albertsen et al. 2010). However, it is more likely to expect a mediating role of self-esteem in the relationship between stress and well-being on the basis of stress process model (Lazarus and Folkman 1984), which indicated some buffering resource that could be used to mediate stress-well-being relationship. Some empirical research, which took self-esteem as a psychological resource, also provided compelling evidence for this mediating effect (e.g., Choi 2010; Lee et al. 2013). For example, using a longitudinal data, Pearlin et al. (1981) demonstrated that stress could erode positive concepts of self (e.g., including self-esteem and mastery). The diminished self-concepts leave one especially vulnerable to experiencing mental issues. Lee et al. (2013) also revealed that self-esteem mediated the relationship between work-related stress and depression. In general, self-esteem can be threatened when an individual considers stress to be uncontrollable or highly disturbing, and threatened self-esteem can consequently undermine well-being.

In line with the rationale reviewed above, the expected mediating role of self-esteem played in the relationship between stress and well-being can also be expected in a real school context. Empirical studies have also well established the relationship between examination stress which is a typical kind of perceived stress, self-esteem and psychological well-being. On the one hand, a large body of research has consistently revealed a close link between examination stress and self-esteem. Learning activities are a key component of adolescent school lives, and examination performance is one of the most important criteria for evaluating students. To get a reasonable academic performance and to be evaluated as good students, most adolescent students are expected to struggle during the semester to get good grades on their academic examinations. Accordingly, the closer a student gets to important examinations when adolescent students are busy preparing for the exam, the more likely they are to encounter difficulties. Consequently, the more likely they are to experience stress and frustration. In the process of preparing for some of the more important examinations that have a significant implication on individual’s later life, a greater level of perceived stress should be expected. From another perspective, researchers have established that experiencing stress and frustration will influence an individual’s emotions and sense of self-worth and will further affect the individual’s self-esteem (Baldwin and Hoffmann 2002; Kernis 2005). Therefore, examination stress could be a major source of stress and could therefore be closely related to the sense of self-value. Thus, examination stress may lead to lower self-esteem in adolescent students. Previous cross-sectional studies have revealed a significant link between perceived stress and self-esteem, specifically that greater perceived stress is associated with lower self-esteem in adolescent students (Nikitha et al. 2014; Uba et al. 2013). Using a longitudinal design, Murray et al. (2013) also found that an increase in stress predicted a reduction in self-esteem in an adolescents’ sample.

On the other hand, some theorists have insisted that self-esteem has some effects on psychological well-being. A positive attitude toward one’s self should lead to self-enhancement and then to happiness; that is, there should be a significantly positive relationship between self-esteem and psychological well-being (Dogan et al. 2013; Rosenberg et al. 1995). In line with these findings, empirical research has confirmed that self-esteem is a significant predictor of psychological well-being (Cheng and Furnham 2003) and has consistently found that the higher the self-esteem, the greater the psychological well-being (Paradise and Kernis 2002; Schilling 2015).

Based both on theoretical and empirical literature about the relationship between examination stress and self-esteem as well as between self-esteem and psychological well-being, it is reasonable to predict that self-esteem could be a potential mediator in the relationships between examination stress and psychological well-being. Drawing on the aforementioned theoretical and empirical research, a third hypothesis regarding the mediating role of self-esteem was proposed.

Hypothesis 3:

Self-esteem mediates the relationship between examination stress and psychological well-being in adolescents.

The Current Study

The current study used latent growth curve modeling to examine developmental trends in examination stress and psychological well-being, incorporating a three-wave longitudinal design. The participants in the current study were senior grade one students at a high school and completed questionnaires (see below section of Measures) three times with an interval of about 2 months during an entire semester (from the beginning of March to the end of June in 2015). Based on the three-wave data, we explored the relationship between examination stress and psychological well-being in two ways. First, we explored the relationship between the initial level of examination stress and the initial level of psychological well-being primarily by looking for an intercept–intercept correlation in the latent growth model. Second, we explored the relationship between a change in examination stress and a change in psychological well-being primarily by looking for a slope–slope correlation in the latent growth model. Latent growth modeling was also used to investigate the mediating role of self-esteem in the relationship between examination stress and psychological well-being.

Methods

Participants

The participants were recruited from the first year class at a senior high school in 2015. In this study, participants completed the questionnaires three times with an interval time of about 2 months throughout the entire term from March to June 2015. A total of 532 participants completed the questionnaire at Time 1, 356 completed the questionnaires at Time 1 and Time 2, and 248 completed the questionnaires at all three times.

Data from the 248 participants were included in the final analyses, of which missing data were estimated using the full information maximum likelihood (FIML) method. Of the participants, 116 were male (mean age = 15.92 ± 0.51 years), and 132 were female (mean age = 15.83 ± 0.57 years). There was no significant gender difference with respect to age (t = 1.39, df = 246, p > .05).

Measures

Psychological Well-Being

Psychological well-being was measured with the Chinese version of the 24-item Scales of Psychological Well-Being (Ryff 1989), as revised by Cheng and Chan (2005), for measuring psychological well-being. This scale has relatively acceptable internal consistency coefficients (Cheng and Chan 2005). An example item is: “In general, I feel I am in charge of the situation in which I live”. After reversely scoring the negative items and summing the items, a high score indicates high overall psychological well-being. In the current study, Cronbach’s alpha was 0.89 at Time 1, 0.90 at Time 2, and 0.88 at Time 3.

Examination Stress

Six items were extracted from the Academic Stress Scale to assess the levels of examination stress. The academic stress scale (Kohn and Frazer 1986) is a 35-item scale designed to measure students’ perceived academic stressors. Respondents report their level of stress under 35 different stressful situations (e.g., examinations, excessive homework, missing class) using a 5-point Likert scale (1 = not at all stressful, 2 = rarely stressful, 3 = sometimes stressful, 4 = fairly stressful, 5 = extremely stressful). The scores are obtained by summing the scores for each item. The academic stress scale has satisfactory internal consistency as measured by Cronbach’s alpha and by split-half reliability (.92 and .86, respectively) (Kohn and Frazer 1986). The aim of the current study was to measure students’ perceived examination stress, we therefore only used the items that related to examinations. Six items that are designed to assess the perceived stress from examination situations and that always included the word “examinations” were extracted to form a new “examination stress scale” to assess the students’ perceived levels of examination stress. Two strategies were used to examine the validity of the new examination stress scale and validate the items on the new scale. First, to assess their content validity, a panel of five experts in psychology assessed the validity of these items. The panel members were provided with the description (definition) of examination stress and were asked to choose between the 35 items those items which best fit the operational definition of examination stress. All the six items that we had previously chosen were selected by the experts, indicating that those items have adequate content validity and discriminant validity. Second, in order to assess the construct validity and reliability, a pilot study was conducted, in which, an Exploratory Factor Analysis (EFA) and a Confirmatory Factor Analysis (CFA) were employed. For the EFA, we approached 500 students (mean age = 15.85, SD = 0.54, 258 girls, 232 boys) and asked them to complete the academic stress scale. An EFA with a varimax rotation was conducted on these 500 students. Kaiser’s rule was used to determine the number of factors. We found that out of all 35 items, the six items on the examination stress scale that were chosen belonged to the same dimension. For the CFA, another 350 students were approached (mean age = 15.87, SD = 0.55, 182 girls, 168 boys) and completed the questionnaire. The CFA results revealed reasonable fit indices, which indicated that the six items of the examination stress scale belonged to one dimension, χ2(9) = 44.451, CFI = 0.954, TLI = 0.924, RMSEA = 0.087, SRMR = 0.040. Results of the EFA and CFA supported that the six items measured a common trait-examination stress. The scale had good internal consistency as measured by Cronbach’s alpha (0.85 at Time 1, 0.84 at Time 2, and 0.81 at Time 3) in the current study.

Self-Esteem

Self-esteem was measured with the Rosenberg Self-Esteem Scale (RSES) (Rosenberg 1965). The RSES is a self-report scale that contains 10 items, each of which is assessed using a 4-point scale that ranges from 1 (strongly disagree) to 4 (strongly agree). It contains an equal number of positively (e.g., “I take a positive attitude toward myself”) and negatively (e.g., “I certainly feel useless at times”) worded items. After reversal scoring the negative items and summing the item results, a high score indicates high overall self-esteem. To ensure that the score of the scale was an indicator of participants’ state self-esteem in that particular week and was sensitive to changes in an individual’s self-esteem, we stressed that the participants should assess each item according to their general feelings over the past 7 days. Cronbach’s alpha was 0.87 in previous studies (Carrard et al. 2012) and was 0.87 at Time 1, 0.90 at Time 2, and 0.86 at Time 3 in the current study.

Socio-Demographic Variables

The participants completed a self-reported questionnaire including students’ socio-demographics variables such as age, gender, sibling status, and family income.

Procedure

In this study, the participants completed the questionnaires three times throughout the entire term (from the beginning of March to the end of June in 2015). The first wave (Time 1) was on a Monday 2 weeks before the first school exam at the very beginning of that semester. Because the exam was at the beginning of a new semester, the students did not have time to get well adapted from holiday to their school lives. Everything in the class tends to seem very difficult at the beginning because of the lack of background knowledge in each discipline. Therefore, the first exam should be the most objectively and subjectively difficult one in the semester, and it seems to be generally accepted as the hardest. In addition, a student’s performance on this examination will determine the evaluations and attitudes of teachers, parents, and classmates as well as their social status and the quality of their lives throughout the following semester. Thus, the level of examination stress is expected to be much higher than that at any other time in the semester. The second wave (Time 2) was on a Monday 2 weeks before the mid-term examination. The mid-term exam is generally carried out in the middle of the semester, primarily to examine the students’ quality of learning during the first half of the semester. The mid-term exam seems easier than the first one because of some accumulation of knowledge in each discipline, so we supposed that the examination stress is lower than in the first wave. The third time (Time 3) was a Monday 2 weeks before the final exam. The students had gone through a full term of study by that time, and most of them are well-prepared for the final exam. The final exam is thought to be less difficult because of good preparation, greater proficiency of knowledge about the subjects, and students’ skill in applying knowledge before the final examination. In addition, we conducted the third wave of questionnaires at a time that was close to the beginning of the summer holiday, so it is reasonable to assume that the perceived examination stress is the lowest for the adolescent students at this time.

At each time point before the questionnaires were distributed to students in their classrooms, female researchers explained the procedure and assured the participants that the study was conducted purely for research purposes and that participation was voluntary. The students were free to decline participation without any negative consequences. All samples were administered as an anonymous self-report survey, however, students needed to write down an identifying number that allowed us to match each participant’s data from the three times the questionnaires were filled out. All the participants in the study received a small gift as a reward (e.g., a notebook, worth around 20 Chinese yuan).

Data Analysis

In the current study, latent growth curve modeling (LGCM) was used for all analyses. LGCM models individual growth trajectories for observed variables over time by specifying latent intercept and slope variables for each construct of interest (Duncan et al. 2013). The scores from Time 1, Time 2, and Time 3 were used as the observed variables. For each model, the intercept was centered relative to the scores at the baseline so that the intercept represented the initial status of the growth curve. The linear slope represents the functional form of the growth trajectory across three time points.

A series of models was tested in the following sequence. First, the growth trajectory of examination stress, self-esteem, and psychological well-being were examined separately. For each construct, as in Fig. 1, a model containing a latent intercept factor and a latent growth factor, in which the intercept was set at the baseline, was estimated. The three loadings of the latent growth factors were set to 0, 1, and 2. Second, the effects of academic examination stress on psychological well-being were examined by the parallel process models shown in Fig. 2. This model coded two latent variables for LGCM of examination stress as a predictor of the latent growth factors in psychological well-being. Third, two approaches were applied to examine the mediation hypothesis. One was to test whether stress at Time 1 predicted well-being at Time 3, mediated by self-esteem at Time 2. The other one was to examine a mediation model (see Fig. 3) with the latent growth factors of self-esteem. In this model we first used a causal-step strategy to inspect the specific paths, and used the bootstrap method to test the indirect effects (MacKinnon 2008; Preacher and Hayes 2008; Zhao et al. 2010). A significant indirect effect indicates that the asymmetric confidence intervals did not include zero. In addition, many studies have found that demographic variables, such as gender (Kling et al. 1999), age (Twenge and Campbell 2001), family income (Zhang and Postiglione 2001), and sibling status (Fleming and Courtney 1984; Rosenberg 1965), affect self-esteem in adolescents. Thus, the impacts of those demographic variables on self-esteem were controlled during the mediation model.

Fig. 1
figure1

Latent growth model for all variables. x1–x3, one variable from Time 1 to Time 3 (e.g., examination stress from Time 1 to Time 3); Ix, the intercept for this variable (e.g., examination stress intercept); Sx, the slope of this variable (e.g., examination stress slope)

Fig. 2
figure2

Parallel growth mode for examination stress and psychological well-being. x1–x3, examination stress from Time 1 to Time 3; Ix, examination stress intercept; Sx, examination stress slope; y1–y3, psychological well-being from Time 1 to Time 3; Iy, psychological well-being intercept; Sy, psychological well-being slope

Fig. 3
figure3

Mediation model. Ix, examination stress intercept; Sx, examination stress slope; Iy, psychological well-being Intercept; Sy, psychological well-being slope; Iz, self-esteem intercept; Sz, self-esteem slope

All the models were estimated using Mplus 7 (Muthén and Muthén 2012). Maximum likelihood estimation with robust standard errors (Satorra and Bentler 1994) was employed for the LGCMs. To allow for analysis of the data containing missing values, all the LGCMs were estimated using full information maximum likelihood (FIML), which assumes that data are missing at random. The model fit was assessed by χ2, the Tucker–Lewis index (TLI), the comparative fit index (CFI), the root-mean-square error of (RMSEA), and the standardized root-mean-square residual (SRMR).

Results

To capture a general idea of the data, we performed a preliminary analysis. Tables 1 and 2 presents the correlations and descriptive statistics for all the variables. The mean levels of all variables decreased from Time 1 to Time 3. The results showed that psychological well-being had a significantly negative correlation with examination stress across three times.

Table 1 Descriptive statistics for all variables
Table 2 Correlations statistics for all variables

Univariate Growth Curves

We first estimated univariate growth curves for each variable. Linear, nonlinear, and quadratic slopes were tested for each variable. We found that the data has a perfect fit for the linear one, but not fit either the non-linear or quadratic models. Therefore, Fig. 1 shows the linear slope model.

Because all of the growth curves showed significant variances in the intercept and the slope terms (see Table 3), none of these variables were eliminated from further analyses. Furthermore, these significant variances indicate the presence of meaningful individual differences in study constructs in both the initial level and the rate of change over the study period. The mean value of the linear slope for each study variable indicates the average rate of linear change for that variable over time, and the variance of the linear slope shows the variability around that average rate.

Table 3 Univariate growth curve

The respective linear univariate LGCMs for all variances in Fig. 1 demonstrated an adequate to excellent fit to the data (see model fit indicators in Table 3). The statistical significance of the mean value of the linear slope suggests that the mean level of all the variables decreased significantly from Time 1 to Time 3.

The results also revealed significant negative relationships between the intercepts and their corresponding slopes for the three variables (see covariance in Table 3). The values of covariance ranged from − 5.212 (p < .01) to − 33.799 (p < .01), with all p's < .05, indicating that individuals who had higher initial levels for all three variables tended to decrease at steeper rates (e.g., have more negative slopes) in these respective constructs.

Parallel Growth Model

To examine the relationship between examination stress and psychological well-being, we built a parallel growth model which simultaneously estimated the latent growth for both examination stress and psychological well-being.

Model 2, as presented in Fig. 2, is a parallel growth model, in which the latent growth factors of examination stress are independent variables and those of psychological well-being are dependent variables. A parallel growth model which provided an excellent fit to the data was estimated, χ2(9, N = 239) = 2.769, CFI = 1.000, TLI = 1.026, RMSEA = 0.000, SRMR = 0.020.

The results showed that the intercept for examination stress negatively and significantly predicted the initial levels of psychological well-being (β = − 0.444, p < .001). This supports hypothesis 1. Also, changes in examination stress from Time 1 to Time 3 negatively predicted changes in psychological well-being from Time 1 to Time 3 (β = − 0.23, p < .05), which is consistent with hypothesis 2. The intercept for examination stress did not predict the slope of psychological well-being (β = 0.074, p > .05).

Mediation Analyses

To examine hypothesis 3, in which self-esteem was expected to mediate the relationship between examination stress and psychological well-being, two mediation analysis strategies were applied.

In the first analysis strategy, we tested whether stress at Time 1 predicted well-being at Time 3, mediated by self-esteem at Time 2. Our results showed that self-esteem at Time 2 played a partial mediation role between examination stress at Time 1 on psychological well-being at Time 3. Examination stress at Time 1 had a significant negative effect on self-esteem at Time 2 (β = − 0.25, SE = 0.077, p = .001). The path from self-esteem at Time 2 to psychological well-being at Time 3 was significant and positive (β = 1.811, SE = 0.155, p < .001). The path from examination stress at Time 1 to psychological well-being at Time 3 was significant and negative (β = − 0.399, SE = 0.186, p < .05). The total effect was significant (β = − 0.258, p < .001), and the total indirect effect was also significant (β = − 0.137, p = .001). Asymmetric confidence limits showed a significant mediated effect (lower limit = − 0.218, upper limit = − 0.056). This suggests that self-esteem at Time 2 partially mediated the effect of examination stress at Time 1 on psychological well-being at Time 3.

Furthermore, in order to verify that it is self-esteem and psychological well-being influenced by examination stress, but not self-esteem and psychological well-being influence examination, we tested a second mediation model with psychological well-being at Time 1 as independent variable, self-esteem at Time 2 as mediating variable, and examination stress at Time 3 as outcome variable. However, we did not find mediated effect in this model. Psychological well-being at Time 1 had a significant positive effect on self-esteem at Time 2 (β = 0.195, SE = 0.020, p < .001). The path from self-esteem at Time 2 to examination stress at Time 3 was not significant (β = − 0.071, SE = 0.073, p = .331). The path from psychological well-being at Time 1 to examination stress at Time 3 was not significant (β = − 0.036, SE = 0.025, p = .147). The total effect was significant (β = − 0.176, p < .05), but the total indirect effect was not significant (β = − 0.049, p = .330). Asymmetric confidence limits, which did include 0, also showed a non-significant mediated effect (lower limit = − 0.147, upper limit = 0.050). These two results provided evidence showing that examination stress was a predictor of psychological well-being, mediated by self-esteem.

In the second analysis approach, based on the results above, we added the latent growth factors of self-esteem as mediators in our model. As we mentioned before, gender, age, family income, and sibling status can affect self-esteem, we therefore added participants’ socio-demographic characteristics as time-invariant covariates to control the impact of those demographic variables on self-esteem. This model was found to fit data well, χ2(53) = 122.561, CFI = 0.944, TLI = 0.924, RMSEA = 0.075, SRMR = 0.079.

In the path from the intercept for examination stress to the intercept for psychological well-being via the intercept for self-esteem, the intercept for examination stress had a significant negative effect on the intercept for self-esteem (β = − 0.398, SE = 0.100, p < .001). The path from the self-esteem intercept to the psychological well-being intercept was significant and positive (β = 2.592, SE = 0.160, p < .001). Therefore, support was found for a mediating path from the intercept for examination stress to the intercept for psychological well-being via the intercept for self-esteem. The estimate of the indirect effect was significant (β = − 0.331, p < .001). Asymmetric confidence limits also showed a significant mediated effect (lower limit − 0.477, upper limit − 0.184). The direct effect of the intercept for examination stress on the intercept for psychological well-being in the model was significant (β = − 0.160, SE = 0.055, p < .01). This suggests that self-esteem partially mediated the effect of the intercept for examination stress on the intercept for psychological well-being (see Table 4).

Table 4 Total and specific indirect effects of examination stress on psychological well-being via self-esteem

On the other hand, the path from the examination stress slope to the self-esteem slope was not significant (β = − 0.249, SE = 0.679, p > .05). The path from the self-esteem slope to the psychological well-being slope was significant (β = 3.209, SE = 2.370, p < .05). The estimate of the indirect effect was not significant (lower limit = − 1.709, upper limit = 1.127). Therefore, there was no mediated effect in this path. The results above partially support Hypothesis 3 (see Table 4).

To further confirm our hypothesis, we swapped the dependent variable and independent variable to form a new model. In the new model, psychological well-being was the independent variable, examination stress was the dependent variable, and self-esteem was a mediator. We employed the same procedure to estimate the new model again while fixing the growth factor variance of self-esteem to zero to establish convergence of the model. The results showed that the model fit the data well, χ2(53) = 125.507, CFI = 0.943, TLI = 0.923, RMSEA = 0.075, SRMR = 0.065. However, the estimate of the indirect effect, the intercept for psychological well-being to the intercept for examination stress via the intercept for self-esteem, was not significant (value = − 0.044, SE = 0.351, p > .05). The estimate of the indirect effect, the slope for psychological well-being to the slope for examination stress via the slope for self-esteem, was not significant (value = 0.412, SE = 1.145, p > .05). This shows that this modified model was not established and was less appropriate than the original unmodified one.

Discussion

In the present study, we examined three hypotheses using LGCM. First, univariate growth curves for examination stress and psychological well-being were assessed using three waves of data, indicating that the mean levels of examination stress, self-esteem, and psychological well-being decreased significantly from Time 1 to Time 3. The results also revealed that individuals with higher initial levels of examination stress, self-esteem, and psychological well-being tended to decrease at steeper rates. Second, a parallel growth model was employed to explore the relationship between examination stress and psychological well-being in two ways. The results indicated that initial levels of examination stress negatively predicted initial levels of psychological well-being and that changes in examination stress across time predicted significant, reversed changes in psychological well-being. Specifically, individuals who decreased at steeper rates in examination stress from Time 1 to Time 3 tended to decrease at gradual rates in psychological well-being across the time span. Third, the mediation analysis suggested that the self-esteem intercept partially mediated the effect of initial levels of examination stress on initial levels of psychological well-being. In the following sections, we will discuss these findings in more detail.

First, the mean levels of examination stress, self-esteem and psychological well-being all decreased significantly from Time 1 to Time 3. As we assumed before conducting the study, the mean levels of examination stress decreased, primarily because of the differences that students perceived about the timing and content of the three exams. However, findings that psychological well-being and self-esteem decreased significantly across time were unexpected. A possible reason could be that our measure of psychological well-being is omnibus, and did not measure a particular aspect of psychological well- being (e.g., academic or examination psychological well-being). This omnibus well-being could be affected by many factors, including individual factors and social factors (Bakhshi et al. 2015), not merely by examination stress. For example, empirical evidences suggest that social support from teachers, peers, and parents was an important factor that contributed to the levels of psychological well-being (Chu et al. 2010). As for the decreased self-esteem, we consider that school environmental change could be an essential reason. In transitioning from middle school to high school, adolescents have to move to new school environment. Students are exposed to different peer groups and influences, and may encounter increased academic rigor and expectations and fewer emotional supports (Hutson et al. 2016). Experience of frustration with difficulties in relationships and learning can easily switch into negative judgment and attitude toward themselves at this stage of adolescence, which may damage self-esteem (Rose 2010). Therefore, self-esteem and psychological well-being overall showed a downward trend.

Second, our results showed in Univariate Growth Curves analysis suggest significantly negative relationships between the intercepts and their corresponding slopes for the three variables, indicating that individuals who had higher initial levels for all three variables tended to decrease at steeper rates (e.g., have more negative slopes). Group dynamics theory could be used to explain such results, which indicates that students’ behavior is influenced by the unique characteristics of a group and the school environment (Lind and Lind 1997). Students at the same high school and same grade, as a group shared the same norms and exhibited similar physical and mental behavior (Lawson et al. 2006). Our sample was a combination of Grade one high school students just graduated from various middle schools. Even though students from different middle school had vastly different mental state, the psychological state of the students will regress to the average level of a group after experiencing the same school atmosphere. Applying it to our results explanation, individuals who had higher level of examination stress, self-esteem and psychological well-being would decrease more to the average level of a group. Therefore, in the meantime, those who scored high on examination stress, self-esteem and psychological well-being at the initial time would experience a steeper decrease in respective constructs over time.

Third, the initial levels of examination stress significantly predicted the initial levels of psychological well-being, a finding which agreed with hypothesis 1. This result is consistent with previous studies (Arulrajah and Harun 2000; Munir et al. 2015; Siddique and D’Arcy 1984; Zhong 2009) that showed that the more an individual perceives stress, the lower their psychological well-being. In addition, changes in examination stress across time predicted a significant, reversed trend in the changes in psychological well-being, which is also consistent with Hypothesis 2. That is, students who decreased steeply in stress have a less pronounced decrease in well-being. These results are in line with previous studies (Rusch et al. 2014; Schwartz et al. 1992) in that reduced perceived stress tends to increase well-being. General models of stress posit that perceiving a situation as being threatening or beyond one’s coping resources causes stress and leads to a negative affect (Lazarus and Folkman 1984). The rationale for the link between stress and mental-health outcomes follows from the underlying assumption of social epidemiological research, which considers mental illness as a pattern of human reaction set in motion by stressful experiences (Dohrenwend and Dohrenwend 1974; Mechanic 1983; Susser 1981). It has been suggested that social stress, actual or perceived, tends to act as a precursor to psychological well-being and exercises a strong direct impact on physical and mental health. Based on these theories, in our study, the levels of examination stress that the students had tended to act as precursors to psychological distress and to lead to changes in psychological well-being.

Fourth, findings that initial levels of self-esteem partially mediated the effect of initial levels of examination stress on initial levels of psychological well-being and self-esteem at Time 2 played a partial mediation role between examination stress at Time 1 support Hypothesis 3. The experience of stress has been shown to negatively influence well-being by decreasing the individual’s positive appraisal of the quality of life and self-esteem because they may focus on the problem or the emotions related to the problem, thus limiting their coping ability (Hinton and Earnest 2010). Self-esteem is an important psychological construct in adolescence and is associated with stress. It has been theorized that the inability to cope with events outside an individual’s control can lead to feelings of inadequacy and worthlessness (Youngs et al. 1990), which are obvious indicators of low self-esteem. A positive attitude toward one’s self could lead to self-enhancement, which could lead to happiness, which has a positive relationship with psychological well-being (Dogan et al. 2013; Rosenberg et al. 1995). In contrast, individuals with low self-esteem are more likely to feel anxious, depressed, hostile, lonely, embarrassed, jealous, ashamed, guilty, hurt, shy, and generally upset compared to people with high self-esteem (Leary 2005). To sum up, when experiencing high examination stress, adolescent students may focus on the examination and the stress related to the examination may cause them to feel inadequate and worthless, which would lower their self-esteem. In turn these adolescent students would be more likely to feel anxious, depressed, ashamed, guilty, as well as generally upset, which would finally lead to a reduction in psychological well-being. Because self-esteem partially mediated the effect, there must exist other factors that played a mediating role. Since we found that the self-esteem slope did not mediate the effect of the examination stress slope on the psychological well-being slope, we may conclude that with further research other psychological constructs may be found which have an even greater impact on the relationship between examination stress and psychological well-being.

Additionally, when analyzing the mediating role of self-esteem in the relationship between stress and psychological well-being, we uncovered a reciprocal relationship between self-esteem and psychological well-being. These results are in line with previous studies, which confirmed that self-esteem was a significant predictor in maintaining psychological health, functioning, and well-being during adolescence (e.g., Cheng and Furnham 2003; Schilling 2015), as well as that well-being prospectively predicted subsequent levels of self-esteem in turn (e.g., Barendregt et al. 2016). These results indicate that the relationship between stress and well-being may not be unidirectional but reciprocal. Specifically, as a mediator, self-esteem can positively influence psychological well-being. Increased psychological well-being can also in turn have a beneficial impact on self-esteem under some other mechanism that different from that revealed in current study (e.g., psychological well-being can help build self-confidence to one’s life), which help establish a positive view of the self and build self-esteem in the long run.

Implications

This study has important theoretical implications by expanding research about the relationship between stress and psychological well-being among young people. To our knowledge, this is the first longitudinal study examining the causal relationship between examination stress and psychological well-being as well as their mediators. Our study added to previous cross-sectional studies (e.g., Rusch et al. 2014) on the associations between perceived stress and psychological well-being by providing robust evidence showing that examination stress may influence adolescents’ psychological well-being through self-esteem. In addition, the application of latent growth curve models in this study further helped to determine the mediating role of self-esteem in the relationship between initial level of examination stress and psychological well-being, though there is no mediating result between trends in examination stress and psychological well-being. These findings enrich the research field about the influencing factors and mechanisms underlying the psychological well-being of adolescents.

Findings of the present study have relevant practical implications with respect to the design of intervention for adolescents’ psychological well-being. The most important finding of this study is that self-esteem plays a mediating role between examination stress and psychological well-being. As such, adolescents with high levels of examination stress were likely to have lower self-esteem, which in turn contributed to lower psychological well-being. These findings suggested that school, together with parents, should pay attention to the examination stress that students may have. In order to maintain and improve adolescents’ psychological well-being, teachers and parents should not put too much pressure on adolescents, but should give students more guidance and suggestions regarding how to cope with examination stress. Furthermore, this study also provided clinically relevant suggestions on how interventions may help to improve students’ psychological well-being. That is, according to the results of present study, for those students with high examination stress, engaging in boosting self-esteem which is a very important psychological resource, may not only buffer the negative effect of examination stress on psychological well-being, but also contribute directly to students’ psychological well-being in the long run.

Limitations

A limitation with respect to the measurement should be noted. Specifically, our measure for psychological well-being is omnibus and there might exist other factors that can affect the general sense of well-being, but were not included in this study. For example, adolescent stress in the life course is marked by physical and psychological maturation, as well as by changing social roles and environments (Meadows et al. 2006). The stresses emanating from the family, school, and peer groups have different contents for adolescence (Meadows et al. 2006). Examination stress is just one of them that adolescence may face in school environment. Other stress, such as relationship stress from peers, academic expectation stress from parents and teachers, and parent-adolescent conflict stress, can also have a serious impact on psychological well-being (Huang 2014). Accordingly, there may exist some other variables that may mediate the relationship between examination stress and psychological well-being among young people. According to stress process model (Lazarus and Folkman 1984), some social resource such as social support and social network (Donnelly 2010), as well as some psychological resources such as self-esteem, mastery and self-efficiency (Pearlin et al. 1981), might play a possible mediated role between stress condition and psychology well-being. It should be very important to identify the factors affecting the general sense of well-being and intervening variables between examination stress and psychological well-being for future theoretical and empirical research.

References

  1. Albertsen, K., Rugulies, R., Garde, A. H., & Burr, H. (2010). The effect of the work environment and performance-based self-esteem on cognitive stress symptoms among Danish knowledge workers. Scandinavian Journal of Public Health,38(3), 81–89.

    Google Scholar 

  2. Arulrajah, A. A., & Harun, L. M. H. (2000, July). Relationship of psychological wellbeing with perceived stress, coping styles, and social support amongst university undergraduates. Paper presented at the national conference on Graduate Research in Education, Malaysia.

  3. Bakhshi, A. B., Sharifi, T., & Ghasemi, P. M. (2015). Assessment the relationship of psychological well-being with religiosity attitude, self-esteem and psychological hardiness by equation structural modeling among students of Islamic Azad and public universities in Shahrekord. Journal of Health Promotion Management,4(3), 60–69.

    Google Scholar 

  4. Baldwin, S. A., & Hoffmann, J. P. (2002). The dynamics of self-esteem: A growth-curve analysis. Journal of Youth and Adolescence,31(2), 101–113.

    Google Scholar 

  5. Barendregt, C. S., van der Laan, A. M., Bongers, I. L., & Van Nieuwenhuizen, C. (2016). Longitudinal relation between general well-being and self-esteem: Testing differences for adolescents admitted to secure residential care and after discharge. International Journal of Offender Therapy and Comparative Criminology,60(16), 1836–1855.

    Google Scholar 

  6. Carrard, I., Crepin, C., Ceschi, G., Golay, A., & Van der Linden, M. (2012). Relations between pure dietary and dietary-negative affect subtypes and impulsivity and reinforcement sensitivity in binge eating individuals. Eating Behaviors,13(1), 13–19.

    Google Scholar 

  7. Chao, R. C. L. (2011). Managing stress and maintaining well-being: Social support, problem-focused coping, and avoidant coping. Journal of Counseling & Development,89(3), 338–348.

    Google Scholar 

  8. Cheng, S. T., & Chan, A. C. M. (2005). Measuring psychological well-being in the Chinese. Personality and Individual Differences,38(6), 1307–1316.

    Google Scholar 

  9. Cheng, H., & Furnham, A. (2003). Attributional style and self-esteem as predictors of psychological well being. Counselling Psychology Quarterly,16(2), 121–130.

    Google Scholar 

  10. Choi, M. K. (2010). The relations of children’s stress, self-esteem, and subjective well-being. Journal of the Korean Home Economics Association,48(10), 65–75.

    Google Scholar 

  11. Chu, P. S., Saucier, D. A., & Hafner, E. (2010). Meta-analysis of the relationships between social support and well-being in children and adolescents. Journal of Social and Clinical Psychology,29(6), 624–645.

    Google Scholar 

  12. DeLongis, A., Folkman, S., & Lazarus, R. S. (1988). The impact of daily stress on health and mood: Psychological and social resources as mediators. Journal of Personality and Social Psychology,54(3), 486–495.

    Google Scholar 

  13. Dogan, T., Totan, T., & Sapmaz, F. (2013). The role of self-esteem, psychological well-being, emotional self-efficacy, and affect balance on happiness: A path model. European Scientific Journal,9(20), 31–42.

    Google Scholar 

  14. Dohrenwend, B. S., & Dohrenwend, B. P. (Eds.). (1974). Stressful life events: Their nature and effects. New York: Wiley.

    Google Scholar 

  15. Donnelly, E. A. (2010). Occupationally related stress exposures and stress reactions in the emergency medical services. Tallahassee: Florida State University.

    Google Scholar 

  16. Duncan, T. E., Duncan, S. C., & Strycker, L. A. (2013). An introduction to latent variable growth curve modeling: Concepts, issues, and application. London: Routledge.

    Google Scholar 

  17. Edwards, J. A. (2004). Stress and well-being in the workplace: A longitudinal cross-lagged structural equation modelling investigation. Doctoral dissertation. Middlesex University, London.

  18. Eisenbarth, C. (2012). Does self-esteem moderate the relations among perceived stress, coping, and depression? College Student Journal,46(1), 149–157.

    Google Scholar 

  19. Eman, S., Dogar, I. A., Khalid, M., & Haider, N. (2012). Gender differences in test anxiety and examination stress. Journal of Pakistan Psychiatric Society,9(2), 85–90.

    Google Scholar 

  20. Fleming, J. S., & Courtney, B. E. (1984). The dimensionality of self-esteem: II. Hierarchical facet model for revised measurement scales. Journal of Personality and Social Psychology,46(2), 404–421.

    Google Scholar 

  21. Folkman, S., Lazarus, R. S., Dunkelschetter, C., Delongis, A., & Gruen, R. J. (1986). Dynamics of a stressful encounter: Cognitive appraisal, coping, and encounter outcomes. Journal of Personality and Social Psychology,50(5), 992–1003.

    Google Scholar 

  22. Gardner, D. H., & Parkinson, T. J. (2011). Optimism, self-esteem, and social support as mediators of the relationships among workload, stress, and well-being in veterinary students. Journal of Veterinary Medical Education,38(1), 60–66.

    Google Scholar 

  23. Goode, K. T., Haley, W. E., Roth, D. L., & Ford, G. R. (1998). Predicting longitudinal changes in caregiver physical and mental health: A stress process model. Health Psychology,17(2), 190–198.

    Google Scholar 

  24. Hinton, R., & Earnest, J. (2010). Stressors, coping, and social support among women in Papua New Guinea. Qualitative Health Research,20(2), 224–238.

    Google Scholar 

  25. Hobfoll, S. E. (1989). Conservation of resource: A new attempt at conceptualizing tress. American Psychologist,44(3), 513–524.

    Google Scholar 

  26. Huang, Y. T. (2014, January). Cumulative and interaction effects of risk factors on depressed mood among adolescents in Taiwan. Paper presented at the 8th annual conference of the Society for Social Work and Research, San Antonio, TX.

  27. Hutson, A., Bleland, K., & Douglass, J. (2016). School-Connect intervention impact on high school students’ discipline referrals and academic outcomes. Austin: Agile Analytics.

    Google Scholar 

  28. Kernis, M. H. (2005). Measuring self-esteem in context: The importance of stability of self-esteem in psychological functioning. Journal of Personality,73(6), 1569–1605.

    Google Scholar 

  29. Kim, E., Hogge, I., & Salvisberg, C. (2014). Effects of self-esteem and ethnic identity: Acculturative stress and psychological well-being among Mexican immigrants. Hispanic Journal of Behavioral Sciences,36(2), 144–163.

    Google Scholar 

  30. Kling, K. C., Hyde, J. S., Showers, C. J., & Buswell, B. N. (1999). Gender differences in self-esteem: A meta-analysis. Psychological Bulletin,125(4), 470–500.

    Google Scholar 

  31. Kofman, O., Meiran, N., Greenberg, E., Balas, M., & Cohen, H. (2006). Enhanced performance on executive functions associated with examination stress: Evidence from task-switching and Stroop paradigms. Cognition and Emotion,20(5), 577–595.

    Google Scholar 

  32. Kohn, J. P., & Frazer, G. H. (1986). An academic stress scale: Identification and rated importance of academic stressors. Psychological Reports,59(2), 415–426.

    Google Scholar 

  33. Lawson, H., Quinn, K., Miller, R., & Hardiman, E. (2006). Mental health needs and problems as opportunities for expanding the boundaries of school improvement. In R. J. Waller (Ed.), Fostering child and adolescent mental health in the classroom (pp. 293–309). Thousand Oaks, CA: Sage.

    Google Scholar 

  34. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.

    Google Scholar 

  35. Leary, M. R. (2005). Sociometer theory and the pursuit of relational value: Getting to the root of self-esteem. European Review of Social Psychology,16(1), 75–111.

    Google Scholar 

  36. Lee, J. S., Joo, E. J., & Choi, K. S. (2013). Perceived stress and self-esteem mediate the effects of work-related stress on depression. Stress and Health,29(1), 75–81.

    Google Scholar 

  37. Lewis, R. S., Nikolova, A., Chang, D. J., & Weekes, N. Y. (2008). Examination stress and components of working memory. Stress,11(2), 108–114.

    Google Scholar 

  38. Li, Q., Zhao, L., & Wang, W. (2010). The effect of examination stress on emotion and immunity of primary school student. Sichuan Mental Health,23(1), 30–31.

    Google Scholar 

  39. Lind, V. R., & Lind, V. R. (1997). The relationship between Hispanic enrollment and the classroom environment in secondary choral music programs. Doctoral dissertation. The University of Arizona, Tucson.

  40. Lopez, S. J., & Snyder, C. R. (Eds.). (2004). Positive psychological assessment: A handbook of models and measures. Washington, DC: American Psychological Association.

    Google Scholar 

  41. MacKinnon, D. P. (2008). Introduction to statistical mediation analysis. New York: Erlbeum and Taylor Francis Group.

    Google Scholar 

  42. Martin, J. A., & Ickovics, J. R. (1987). The effects of stress on the psychological well-being of Army wives: Initial findings from a longitudinal study. Journal of Human Stress,13(3), 108–115.

    Google Scholar 

  43. Meadows, S. O., Brown, J. S., & Elder, G. H. (2006). Depressive symptoms, stress, and support: Gendered trajectories from adolescence to young adulthood. Journal of Youth and Adolescence,35(1), 89–99.

    Google Scholar 

  44. Mechanic, D. (1983). Adolescent health and illness behavior: Review of the literature and a new hypothesis for the study of stress. Journal of Human Stress,9(2), 4–13.

    Google Scholar 

  45. Mohapatra, S., Panigrahi, S. K., & Rath, D. (2012). Examination stress in adolescents. Asian Journal of Paediatric Practice,16(1), 1–3.

    Google Scholar 

  46. Munir, T., Shafiq, S., Ahmad, Z., & Khan, S. (2015). Impact of loneliness and academic stress on psychological well being among college students. Academic Research International,6(2), 343–355.

    Google Scholar 

  47. Murray, K., Rieger, E., & Byrne, D. (2013). A longitudinal investigation of the mediating role of self-esteem and body importance in the relationship between stress and body dissatisfaction in adolescent females and males. Body Image,10(4), 544–551.

    Google Scholar 

  48. Muthén, B., & Muthén, L. (2012). Mplus version 7: User’s guide. Los Angeles, CA: Muthén & Muthén.

    Google Scholar 

  49. Nikitha, S., Jose, T. T., & Valsaraj, B. P. (2014). A correlational study on academic stress and self-esteem among higher secondary students in selected schools of Udupi district. Nitte University Journal of Health Science,4(1), 106–108.

    Google Scholar 

  50. Norvilitis, J. M., & Reid, H. M. (2012). Predictors of academic and social success and psychological well-being in college students. Education Research International,2012(3), 127–135.

    Google Scholar 

  51. Ortuño-Sierra, J., Aritio-Solana, R., Chocarro de Luis, E., & Fonseca-Pedrero, E. (2015). Early adolescence and stress in the school context: An examination of the student stress inventory-stress manifestation. Electronic Journal of Research in Educational Psychology,13(3), 607–630.

    Google Scholar 

  52. Paradise, A. W., & Kernis, M. H. (2002). Self-esteem and psychological well-being: Implications of fragile self-esteem. Journal of Social and Clinical Psychology,21(4), 345–361.

    Google Scholar 

  53. Pearlin, L. I., Menaghan, E. G., Lieberman, M. A., & Mullan, J. T. (1981). The stress process. Journal of Health and Social Behavior,22, 337–356.

    Google Scholar 

  54. 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(3), 879–891.

    Google Scholar 

  55. Putwain, D. (2007). Researching academic stress and anxiety in students: Some methodological considerations. British Educational Research Journal,33(2), 207–219.

    Google Scholar 

  56. Rohrmann, S., Netter, P., Hennig, J., & Hodapp, V. (2003). Repression–sensitization, gender, and discrepancies in psychobiological reactions to examination stress. Anxiety Stress and Coping,16(3), 321–329.

    Google Scholar 

  57. Rose, J. (2010, November). The effect of frustration on implicit self-esteem. Paper presented at the annual meeting of the Oklahoma Research Day, Cameron University, Lawton.

  58. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

    Google Scholar 

  59. Rosenberg, M., Schooler, C., Schoenbach, C., & Rosenberg, F. (1995). Global self-esteem and specific self-esteem: Different concepts, different outcomes. American Sociological Review,60(1), 141–156.

    Google Scholar 

  60. Rusch, N., Muller, M., Heekeren, K., Theodoridou, A., Metzler, S., Dvorsky, D., et al. (2014). Longitudinal course of self-labeling, stigma stress and well-being among young people at risk of psychosis. Schizophrenia Research,158(1), 82–84.

    Google Scholar 

  61. Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology,57(6), 1069–1081.

    Google Scholar 

  62. Ryff, C. D., & Singer, B. H. (2008). Know thyself and become what you are: A eudaimonic approach to psychological well-being. Journal of Happiness Studies,9(1), 13–39.

    Google Scholar 

  63. Satorra, A., & Bentler, P. M. (1994). Corrections to test statistics and standard errors in covariance structure analysis. In A. von Eye & C. C. Clogg (Eds.), Latent variables analysis: Applications for developmental research (pp. 399–419). Thousand Oaks, CA: Sage.

    Google Scholar 

  64. Schilling, K. (2015). Examining the role of self-esteem in the association between emotional vulnerability and psychological well-being. Master’s thesis. University of Dayton: Ohio.

  65. Schwartz, G., Schwartz, J., Nowack, K., & Eichling, P. (1992). Changes in perceived stress and social support over time are related to changes in immune function. Unpublished manuscript. University of Arizona and Canyon Ranch, Arizona, USA.

  66. Siddique, C., & D’Arcy, C. (1984). Adolescence, stress, and psychological well-being. Journal of Youth and Adolescence,13(6), 459–473.

    Google Scholar 

  67. Singh, R., Goyal, M., Tiwari, S., Ghildiyal, A., Nattu, S., & Das, S. (2011). Effect of examination stress on mood, performance and cortisol levels in medical students. Indian Journal of Physiology and Pharmacology,56(1), 48–55.

    Google Scholar 

  68. Srivastava, A. (2015). Examination stress among Indian adolescents: A perspective. International Journal of Management and Social Science Research Review,1(7), 17–21.

    Google Scholar 

  69. Susser, M. (1981). The epidemiology of life stress. Psychological Medicine,11(1), 1–8.

    Google Scholar 

  70. Taylor, S. E., & Stanton, A. L. (2007). Coping resources, coping processes, and mental health. Annual Review of Clinical Psychology,3, 377–401.

    Google Scholar 

  71. Topham, P., & Moller, N. (2011). New students’ psychological well-being and its relation to first year academic performance in a UK university. Counselling and Psychotherapy Research,11(3), 196–203.

    Google Scholar 

  72. Twenge, J. M., & Campbell, W. K. (2001). Age and birth cohort differences in self-esteem: A cross-temporal meta-analysis. Personality and Social Psychology Review,5(4), 321–344.

    Google Scholar 

  73. Uba, I., Yaacob, S. N., Talib, M. A., Mofrad, S., & Abdullah, R. (2013). Effect of self-esteem in the relationship between stress and substance abuse among adolescents: A mediation outcome. International Journal of Social Science and Humanity,3(3), 214–217.

    Google Scholar 

  74. Wadee, A. A., Kuschke, R. H., Kometz, S., & Berk, M. (2001). Personality factors, stress and immunity. Stress and Health,17(1), 25–40.

    Google Scholar 

  75. Weekes, N., Lewis, R., Patel, F., Garrison-Jakel, J., Berger, D. E., & Lupien, S. J. (2006). Examination stress as an ecological inducer of cortisol and psychological responses to stress in undergraduate students. Stress,9(4), 199–206.

    Google Scholar 

  76. Yang, M. Y. (2016). Research on the relationship among learning stress, resilience and test anxiety of senior student. Master’s thesis. Guizhou Normal University, China.

  77. Youngs, G. A., Rathge, R., Mullis, R., & Mullis, A. (1990). Adolescent stress and self-esteem. Adolescence,25(98), 333–341.

    Google Scholar 

  78. Zhang, L., & Postiglione, G. A. (2001). Thinking styles, self-esteem, and socio-economic status. Personality and Individual Differences,31(8), 1333–1346.

    Google Scholar 

  79. Zhao, X., Lynch, J. G., & Chen, Q. (2010). Reconsidering baron and kenny: Myths and truths about mediation analysis. Journal of Consumer Research,37(2), 197–206.

    Google Scholar 

  80. Zhong, L. F. (2009). Academic stress and subjective well-being: The moderating effects of perceived social support. Paper presented at the 16th international conference on Industrial Engineering and Engineering Management, Peking, China.

Download references

Acknowledgements

The authors would like to thank National Nature Science Foundation of China (31200778) and the Fundamental Research Funds for the Central Universities (15SZTZ01) for funding this research. We thank Drs. Rhoda E. and Edmund F. Perozzi for their extensive review and English language assistance on this paper.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Baoshan Zhang.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Xiang, Z., Tan, S., Kang, Q. et al. Longitudinal Effects of Examination Stress on Psychological Well-Being and a Possible Mediating Role of Self-Esteem in Chinese High School Students. J Happiness Stud 20, 283–305 (2019). https://doi.org/10.1007/s10902-017-9948-9

Download citation

Keywords

  • LGCM
  • Psychological well-being
  • Examination stress
  • Adolescents