Although previous studies of trauma have revealed that many people with traumatic experiences exhibit negative psychological outcomes, some positive psychological changes have also been reported posttrauma, including improvements in personal strength, relationships, and appreciation of life (e.g., Ying, Wang, Lin, & Chen, Reference Ying, Wang, Lin and Chen2016; Yoshida et al., Reference Yoshida, Kobayashi, Honda, Matsuoka, Yamaguchi, Homma and Tomita2016). Tedeschi and Calhoun (Reference Tedeschi and Calhoun1996) coined the term posttraumatic growth (PTG) to describe positive changes exhibited after struggling with traumatic events. PTG has recently attracted increasing research interest (e.g., García, Páez-Rovira, Zurtia, Martel, & Reyes, Reference García, Páez-Rovira, Zurtia, Martel and Reyes2014; Yuen, Ho, & Chan, Reference Yuen, Ho and Chan2014). In some studies, PTG has been considered a transcendent state compared with the pretrauma state (e.g., Zoellner & Maercker, Reference Zoellner and Maercker2006). Importantly, PTG is reported to be prevalent among survivors of various traumatic experiences (e.g., Cho & Park, Reference Cho and Park2013; Kira et al., Reference Kira, Aboumediene, Ashby, Odenat, Mohanesh and Alamia2013).

To elucidate the underlying mechanisms of PTG development, some researchers have focused on potential predictive factors (e.g., Achterhof et al., Reference Achterhof, Dorahy, Rowlands, Renouf, Britt and Carter2017; García et al., Reference García, Páez-Rovira, Zurtia, Martel and Reyes2014; Wu, Zhou, Wu, & An, Reference Wu, Zhou, Wu and An2015), reporting that hope is a significant predictive factor for PTG (e.g., Zhou, Wu, & Zhen, Reference Zhou, Wu and Zhen2018). Similarly, other trauma-related studies have also indicated that hope for life predicted increased PTG (e.g., Casellas-Grau, Ochoa, & Ruini, Reference Casellas-Grau, Ochoa and Ruini2017; Heidarzadeh, Dadkhah, & Gholchin, Reference Heidarzadeh, Dadkhah and Gholchin2016; Yuen et al., Reference Yuen, Ho and Chan2014). Janoff-Bulmans’ (Reference Janoff-Bulman2010) shattered assumptions theory has been an important framework for interpreting this phenomenon. In this theoretical model, it is assumed that traumatic experiences can challenge people's stable view of the world, others, and self. In turn, this can lead people to lose the sense of predictability and control of the self, others, or the world. People may then lose hope and feel a sense of hopelessness toward life (e.g., Glass, Flory, Hankin, Kloos, & Turecki, Reference Glass, Flory, Hankin, Kloos and Turecki2009; Zhou, Wu, & Wang, Reference Zhou, Wu and Wang2017), which can prevent the survivors of traumatic events from adopting positive strategies for coping with traumatic events and posttraumatic reactions, limiting the realisation of PTG (e.g., Ho et al., Reference Ho, Rajandram, Chan, Samman, McGrath and Zwahlen2011). In contrast, hope may shift individuals’ focus from negative events or emotional states (Scheier & Carver, Reference Scheier and Carver1985), and facilitate their ability to overcome stressful life situations by engaging coping behaviour and adaptive problem solving (Reff, Kwon, & Campbell, Reference Reff, Kwon and Campbell2005). A study of adolescents following an earthquake revealed that once they developed a sense of hope about their posttraumatic lives, they actively sought for positive and effective strategies to struggle with traumatic events, and reconstructed their understanding of the posttraumatic self, others, and the world, ultimately leading to the realisation of PTG (Zhou, Wu, Wang, & Tian, Reference Zhou, Wu, Wang and Tian2017).

Nevertheless, the predictive mechanisms of hope in relation to PTG remain unclear. However, relevant theories suggest that hope can promote PTG via other factors (e.g., Joseph & Linley, Reference Joseph and Linley2005; Snyder, Reference Snyder2002), among which cognitive reappraisal and acceptance may be two important mediating factors. Therefore, the current study sought to further explore the potential mediating effects of cognitive reappraisal and acceptance in the relationship between hope and PTG.

Cognitive reappraisal is an antecedent-focused coping strategy (e.g., Gross, Reference Gross2002), by which a person makes active efforts to repair a negative emotional state, as well as generating benign or positive interpretations or perspectives in relation to a stressful situation (e.g., Gross, Reference Gross1998). According to Snyder's (Reference Snyder2002) hope theory, hope is comprised of agency thinking and pathways thinking, and can enable people to adopt positive and functional strategies to realise their purpose. Cognitive reappraisal is among these strategies. Through cognitive reappraisal, traumatised people with a sense of hope may change their negative emotional states and thoughts regarding traumatic events (e.g., Benzein & Berg, Reference Benzein and Berg2005), shifting their focus from the negative aspects of trauma to its positive side (e.g., Nes & Segerstrom, Reference Nes and Segerstrom2006). In turn, this may enable people to make sense of the traumatic event and understand its meaning in their life, ultimately realising PTG (e.g., Zhou et al., Reference Zhou, Wu, Fu and An2015).

Acceptance of the posttrauma world may be a further important factor in the relationship between hope and PTG. Joseph and Linley's (Reference Joseph and Linley2005) organismic valuing theory of PTG suggests that people have an innate tendency to know their own optimal life direction. This may activate one's aspiration or hope in the pursuit of fulfillment (e.g., Ryan, Reference Ryan1995). Particularly after the experience of trauma, hope can help traumatised individuals to accommodate new trauma-related information within an existing model of the world (e.g., Joseph & Linley, Reference Joseph and Linley2005). Here, successful accommodation implies people's acceptance of reality (e.g., McCracken, Reference McCracken1998). Moreover, this acceptance requires people to change their previously held views of the self, others, and the world (e.g., Janoff-Bulman, Reference Janoff-Bulman2010), which can relieve their negative psychological responses (e.g., Åkerblom, Perrin, Fischer, & McCracken, Reference Åkerblom, Perrin, Fischer and McCracken2015), and increase the likelihood of experiencing positive changes (e.g., Baranoff, Hanrahan, Kapur, & Connor, Reference Baranoff, Hanrahan, Kapur and Connor2013) and posttraumatic adjustments (e.g., Hayes, Luoma, Bond, Masuda, & Lillis, Reference Hayes, Luoma, Bond, Masuda and Lillis2006), leading to the realisation of PTG.

Although cognitive reappraisal and acceptance may separately mediate the relationship between hope and PTG, their combined effects are currently unclear. According to the organismic valuing theory of PTG (Joseph & Linley, Reference Joseph and Linley2005), traumatic events can lead to a disparity between new trauma-related information and pre-existing expectations or beliefs (e.g., Janoff-Bulman, Reference Janoff-Bulman2010). This disparity may prompt people to deal with this new information, such as seeking to understand the event retrospectively (e.g., Janoff-Bulman, Reference Janoff-Bulman2010; Zhou, Wu, Fu, & An, Reference Zhou, Wu, Fu and An2015) through cognitive reappraisal. Once the comprehension of traumatic information is achieved, people can then accommodate (e.g., Joseph & Linley, Reference Joseph and Linley2005) and accept the posttraumatic reality, and in turn enhance their understanding of the traumatic world. Thus, it is likely that cognitive reappraisal can lead to the realisation of PTG by fostering increased acceptance.

Although several theoretical models have been proposed to describe the relationship between hope and PTG, the predictive utility of such a relationship has not been evaluated in adolescent survivors after natural disasters. Because adolescents are more susceptible to the impacts of disasters (e.g., Margolin, Ramos, & Guran, Reference Margolin, Ramos and Guran2010), this is a particularly important issue. However, the combined role of cognitive reappraisal and acceptance in the relationship between hope and PTG has not yet been evaluated in trauma-related studies. Importantly, adolescents’ cognitive capacity is not fully developed, meaning that trauma is more likely to have event-related cognitive effects, which may have an impact on the later development of psychosomatic responses. Therefore, studies of the mechanisms by which adolescents cope with trauma constitute are urgently needed to develop methods for promoting adolescent posttrauma adjustment.

Given this background, the aim of the present study was to examine the mediating role of cognitive reappraisal and acceptance in the relationship between hope and PTG. We sought to test the following three hypotheses: (1) hope exerts a direct and positive effect on PTG; (2) cognitive reappraisal and acceptance separately mediate the relationship between hope and PTG; and (3) acceptance mediates the relationship between cognitive reappraisal and PTG.

Methods

Participants and Procedures

The present study was conducted 2.5 years after the Ya'an earthquake in Lushan county in Sichuan province, which was the most severely affected area. We first contacted the local education authority and informed them of the aims and methods of the study, and indicated that we could provide psychological services if required by the respondents. With the approval of the local education authority, we focused on the only two schools in Lushan county. Under the approval of head teachers and with the help of psychological teachers, students from six classes in each school were ultimately selected to participate into our investigation. All students in the selected classes were attending school on the assessment date, and the assessment was conducted under the supervision of trained individuals with a master's degree in psychology. In the present study, we selected a total of 397 adolescent survivors. The mean age of the participants was 16.42 (SD = 1.76) years, and the age range was 13 to 20 years. Of the 397 participants, 154 (38.8%) were male and 242 (61.0%) were female; one participant did not report their gender. All participants experienced this traumatic event; 9.8% of the participants were trapped and 9.1% were physically injured during the earthquake.

The study was approved by the Research Ethics Committee of Beijing Normal University and conducted with the permission of the principals of the participating schools. In the selected classes, everyone who attended school on the date of the survey was recruited to participate. There were no exclusion criteria, and compensation was not provided. The purpose of the study and the voluntary nature of the students’ participation were highlighted before the survey was conducted, and written informed consent was obtained from school principals and classroom teachers.

Measures

Trauma exposure

The Traumatic Exposure Questionnaire, with 18 items (Wu, Zhang, Lin, & Zang, Reference Wu, Zhang, Lin and Zang2013), was used to assess the severity of earthquake exposure among adolescents. The one-factor questionnaire has 18 items, and each of the items in this questionnaire was rated on a 3-point scale, in which 3 represented experience myself, 2 represented know of others’ experience and 1 represented do not experience. This questionnaire has previously been found to have good reliability among adolescents who experienced the Ya'an earthquake (Zhou, Wu, & Zhen, Reference Zhou, Wu and Zhen2017). In the current study, the questionnaire exhibited good validity (χ²/df = 2.57, CFI = 0.93, TLI = 0.90, RMSEA (90% CI) = 0.063 [0.053, 0.073], SRMR = 0.070) and reliability (Cronbach's α = 0.79).

Hope

The State Hope Scale (SHS; Snyder et al., Reference Snyder, Sympson, Ybasco, Borders, Babyak and Higgins1996) was used to assess adolescents’ sense of hope. The SHS has six items, including two dimensions: agency thinking (e.g., ‘I try my best to pursue my own goals’) and pathways thinking (e.g., ‘I can think of many paths to realise my goals’). Each of these items was rated on an 8-point scale ranging from 1 (completely disagree) to 8 (completely agree). When using this scale, we translated the original English version of the scale into Chinese and reworded some items to clarify their meaning. In the current study, SHS exhibited good validity (χ²/df = 2.53, CFI = 0.99, TLI = 0.98, RMSEA (90% CI) = 0.062 [0.013, 0.111], SRMR = 0.014) and reliability (Cronbach's α = 0.85).

Cognitive reappraisal

The cognitive reappraisal subscale of the Emotion Regulation Questionnaire (revised Chinese version; Wang, Liu, Li, & Du, Reference Wang, Liu, Li and Du2007) was used to assess cognitive reappraisal. The Emotion Regulation Questionnaire has 10 items divided into two dimensions: cognitive reappraisal and expressive suppression. Every item was rated on a 7-point Likert scale, in which 1 represented completely disagree and 7 represented completely agree. In this study, the reliability of the questionnaire was good (α = 0.80), and it showed good validity (χ²/df = 3.50, CFI = 0.94, TLI = 0.91, RMSEA [90% CI] = 0.079 [0.063, 0.096], SRMR = 0.041). For the current study, we only used the cognitive reappraisal subscale, which showed good internal reliability (Cronbach's α = 0.82).

Acceptance

We used the acceptance subscale of the Cognitive Emotion Regulation Questionnaire (Chinese revised version; Zhu, Luo, Yao, Auerbach, & Abela, Reference Zhu, Luo, Yao, Auerbach and Abela2007). The Cognitive Emotion Regulation Questionnaire is a 36-item instrument designed to measure the following cognitive coping strategies: acceptance, positive reappraisal, self-blame, positive refocusing, rumination, catastrophising, refocus on planning, others-blame, and putting into perspective (e.g., Garnefski, Teerds, Kraaij, Legerstee, & van den Kommer, Reference Garnefski, Teerds, Kraaij, Legerstee and van den Kommer2004). All of these items were rated on a 5-point Likert scale, which ranged from 0 (completely disagree) to 4 (completely agree). The Chinese version of this questionnaire has been found to have good stability and reliability (e.g., Zhu et al., Reference Zhu, Luo, Yao, Auerbach and Abela2007). The current study revealed that the questionnaire had good internal reliability (Cronbach's α = 0.92) and acceptable validity (χ²/df = 2.29, CFI = 0.90, TLI = 0.89, RMSEA [90% CI] = 0.057 [0.053, 0.061], SRMR = 0.059). In the current study, we selected the acceptance subscale (e.g., ‘I have to accept reality’), which exhibited good internal reliability (Cronbach's α = 0.75).

Posttraumatic growth

A modified version of the Post-Traumatic Growth Inventory (Zhou, Wu, An, Chen, & Long, Reference Zhou, Wu, An, Chen and Long2014) was used to assess adolescents’ PTG. The modified PTGI consisted of 22 items, and included three dimensions: perceived changes in self, a changed sense of relationships with others, and a changed philosophy of life. All the items were scored on a 6-point scale, in which 0 represented no change and 5 represented very great degree of change. The score range of this questionnaire was 0–110. This questionnaire has been found to have good stability and reliability for Chinese adolescents after experiencing an earthquake (e.g., Zhou, Wu, An, Chen, & Long, Reference Zhou, Wu, An, Chen and Long2014). In the current study, we found that the reliability of the inventory was good (Cronbach's α = 0.89) and the validity was acceptable (χ²/df = 3.53, CFI = 0.90, TLI = 0.89, RMSEA [90% CI] = 0.080 [0.073, 0.086], SRMR = 0.052).

Data Analysis Strategies

Descriptive analyses and Pearson's correlations were first calculated to examine all the main measurements and their associations, respectively. Mplus 6.0 software (e.g., Muthén & Muthén, Reference Muthén and Muthén2010) was used to conduct data analysis. We first conducted an analysis of missing data on variables and found that missing data across all items were less than 2.3%. We then adopted Little's Missing Completely at Random test to assess the missing data, and found that the data were missing at random, χ²(35) = 45.59, p = .108. Maximum likelihood (ML) was then used to handle missing data.

As all the data were from the self-reported scales, to avoid possible contamination of common method bias in the study results, we balanced the item order and anonymised the measurement process. In addition, we used Harman's one-factor test (Podsakoff, Mackenzie, Lee, & Podsakoff, Reference Podsakoff, Mackenzie, Lee and Podsakoff2003) to assess bias. Eight factors with eigenvalues greater than 1 were extracted. These factors explained 65.88% of the variance. Nevertheless, the first factor accounted for 29.91% of the variance, which is less than 40% of the variance. These results indicated that there was no significant common method variance in this study.

Next, we constructed two models to examine our research questions: (1) a direct effects model of the structural path from hope to PTG; and (2) based on the direct model, we inserted mediators (e.g., acceptance and cognitive reappraisal) between hope and PTG, and added a predictive path from cognitive reappraisal to acceptance, establishing a multiple indirect effects model (see Figure 1). Additionally, considering the importance of traumatic exposure in the development of PTG, we controlled for traumatic exposure in the model for examining the effects of hope, acceptance and cognitive reappraisal on PTG. Lastly, bootstrap method was used to examine the indirect effects.

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

Figure 1 The multiple indirect effects model after controlling gender and age. Note: rectangle = observable variable, ellipse = latent variable. Self = perceived changes in self; Relation to others = changed sense of relationships with others; Life = changed philosophy of life

Results

Descriptive Statistics and Correlation Analysis

Table 1 shows the results of descriptive statistics and correlation analysis. The mean levels of trauma exposure, hope, acceptance, cognitive reappraisal, and PTG were 20.45, 29.25, 10.15, 28.42, and 62.69 respectively. Pearson's correlations were also calculated, revealing that gender had significant relationships with hope, cognitive reappraisal, and PTG. Age was non-significantly associated with other variables. Traumatic exposure was only related significantly to PTG, but not other factors. There were positive correlations between hope, acceptance, cognitive reappraisal, and PTG.

Table 1 Means, Standard Deviations, and Correlations Among Main Variables

Note: Gender was coded as 1 and 2 in this study, and 1 represents male, and 2 represents female.

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

Analysis of Mediating Effects

We followed two steps to examine the mediating effects of acceptance and cognitive reappraisal in the relationship between hope and PTG. The first was to examine the direct effects model with the path from hope to PTG, and the second was to assess the multiple indirect effects model of hope, acceptance, cognitive reappraisal, and PTG.

A direct effects model was first constructed to examine the direct effects of hope on PTG. The results revealed that the model fit the data well (χ²/df = 2.02, CFI = 0.996, TLI = 0.991, RMSEA (90% CI) = 0.054 [0.000, 0.104], SRMR = 0.012), and that hope had a direct and positive effect on PTG.

Next, after controlling for gender, age, and traumatic exposure, we inserted the mediators (acceptance and cognitive reappraisal) between hope and PTG, and added a predictive path from cognitive reappraisal to acceptance to establish a multiple indirect effects model (see Figure 1). This model fit the data well (χ²/df = 1.12, CFI = 0.998, TLI = 0.996, RMSEA (90% CI) = 0.017 [0.000, 0.049], SRMR = 0.015). Path analyses revealed that after controlling the gender and age, traumatic exposure had no significant effect on hope, cognitive reappraisal, or acceptance, but had a significant and positive effect on PTG. Hope had a significant and positive effect on acceptance and cognitive reappraisal, as well as on PTG. Cognitive reappraisal had a significant and positive effect on acceptance and PTG. Acceptance positively predicted PTG. Taken together, all of the variables in this model explained 30% of the variance of PTG. These results indicated that when age, gender, and traumatic exposure were controlled for, hope not only had a direct and positive effect on PTG, but also indirectly affected PTG through acceptance and cognitive reappraisal, as well as through a multiple path from cognitive reappraisal to acceptance. Then, the bootstrap method was used to examine the indirect effects, and the results indicated that the indirect effects value from cognitive reappraisal and acceptance was 0.05 and 0.01, respectively. The multiple indirect effect value from cognitive reappraisal to acceptance is 0.005. The indirect effect accounts for 16.7% of the total effect hope predicted PTG.