Leaving home, or uprooting, particularly for the first time, is stressful and is often marked by difficult experiences adjusting to the new environment. College is often the time when youth depart their familiar homeland and undergo psychological changes adjusting to a new environment. While studies have examined psychological distress of uprooting and homesickness (see the review by Scopelliti & Tiberio, 2012), how youth cope with these difficulties of being away from home is unclear. Gender and nationality have been variables in research on homesickness and uprooting stress during college transitions, but the results are inconclusive as to whether males or females struggle more (Poyrazli & Lopez, Reference Poyrazli and Lopez2007). Coping with uprooting stress is a major issue in a country like China as millions engage in interprovincial migration for purposes of education. The present study intends to examine the longitudinal effects of uprooting, coping, and psychological adjustment on recent non-local, first-year Chinese university students.
Uprooting as an Element of Social Change
Humans have always been mobile, but as the term ‘uprooting’ implies, we put down roots and make social connections when we settle in one place. Uprooting has been defined generally as the process when one leaves one place (forced or self-imposed) to relocate to another place (Brown, Reference Brown, Coehlo and Ahmed1980). While uprooting can occur at any time across one's lifespan, the first year transition to college has been regarded as major life event and has served as a subject of study for many decades (Sun, Reference Sun2015).
College students are not immune to the negative effects of uprooting (uprooting stress) during their adjustment process. In general, researchers have agreed that the experience of uprooting (e.g., feelings of separation or homesickness) is directly linked to increased levels of depression, loneliness, and anxiety (Fisher & Hood, Reference Fisher and Hood1988). Although Fisher and Hood (Reference Fisher and Hood1988) found that college women experience more uprooting difficulties, Brewin, Furham, and Howes (Reference Brewin, Furnham and Howes1989) found that college men were more homesick. Interestingly, the literature has found a high prevalence of uprooting stress at some point in college students’ transition for the majority of college students around the world, including the United Kingdom, the Netherlands, the United States, Turkey, New Zealand, Australia, and China (Fisher, 1986; Scharp, Paxman, & Thomas, Reference Scharp, Paxman and Thomas2016; Stroebe, Van Vliet, Hewstone, & Willis, Reference Stroebe, Vliet, Hewstone and Willis2002; Tao, Dong, Pratt, Hunsberger, & Pancer, Reference Tao, Dong, Pratt, Hunsberger and Pancer2000; Urquhart & Pooley, Reference Urquhart and Pooley2007). Research on geographic distance has also consistently found that college students whose families lived closer experience less homesickness (Stroebe et al, Reference Stroebe, Vliet, Hewstone and Willis2002; Tognoli, 2010). There is also a plethora of research that has attributed positive benefits to social support during university adjustment (see review by Hinderlie & Kenny, Reference Hinderlie and Kenny2002; Sun, Reference Sun2015; Tao et al., Reference Tao, Dong, Pratt, Hunsberger and Pancer2000).
While transition to college presents major changes to domestic students, international students have also been prone to uprooting difficulties, particularly as they cross cultures in pursuit of higher education. Cross-cultural adaptation studies provide a useful theoretical framework for researching the relationship between stress-coping-adaptation for these situations. Demes and Geeraert (Reference Demes and Geeraert2015) followed over 2,500 international studentsFootnote 1 for 2 years as they studied at various sites around the world and found a strong association between stress and psychological adaptation — that coping strategies reduced stress over time. While the authors focused on general acculturative stress (i.e., stress of adjusting to new culture), Ward (Reference Ward and Matsumoto2001) highlighted the linkage among stress, homesickness, and maladaptation across international students. The most recent and relevant empirical evidence pointing to the association among uprooting, coping, and adjustment stems from a study by Szabo, Jose, and Ward (Reference Szabo, Ward and Jose2016). Using a sample of international students from two cultural backgrounds (Asian and Western) in New Zealand, the longitudinal study found that uprooting stress was moderated by two coping constructs (explained in the next paragraph). Given that uprooting stress is considered uncontrollable due to the lack of familiarity in the new environment, primary coping exacerbated maladaptation and secondary coping buffered the negative effects on anxiety. This research study points out that university students (domestic and international) undergoing a college transition may experience difficulty in coping with the unknown, resulting in a higher chance of becoming distressed and feeling a sense of lack of control (Fisher, Reference Fisher1989).
Stress and Coping
The person-environment model is useful for understanding uprooting stress in first-year college transitions. According to Lazarus and Folkman's (Reference Lazarus and Folkman1984) transactional theory of stress, when an individual encounters a stressful situation, it is necessary to appraise the potential threat posed by the situation and evaluate available resources, or coping strategies. Coping is defined as how an individual changes and/or cognitively structures stress in the person-environment relationship (Lazarus & Folkman, Reference Lazarus and Folkman1984). In the reaction process to stress, Rothabum, Weisz, and Snyder (Reference Rothbaum, Weisz and Snyder1982) suggest that when a stress-provoking situation is encountered, individuals may exert primary control coping, aimed at taking control or action in the person-environment relationship, or use secondary control coping, aimed at adjusting one's perception or interpretation of the person-environment relationship. In other words, primary coping is a behaviour oriented towards managing the stressor, while secondary coping is a cognitive change oriented towards reframing the stressor and fitting oneself into the environment (Skinner, Edge, Altman, & Sherwood, Reference Skinner, Edge, Altman and Sherwood2003). Morling and Evered (Reference Morling and Evered2006) operationalised secondary coping as the strategy to adjust the self and accept the situation. In contrast, primary coping attempts to exert control of the situation. Both coping control strategies (primary and secondary) have been defined in terms of a fit-focused approach and are independent constructs (Morling & Evered, Reference Morling and Evered2006), meaning it is important to note that people can use either of these coping styles depending on which one fits the social context better.
Goodness of Fit: Matching Control Strategy
It should be obvious that the two coping strategies may not be equally effective in all situations. That is to say, primary coping is more applicable in situations where one has the opportunity and ability to change the circumstances (Zakowski, Hall, Klein, & Baum, Reference Zakowski, Hall, Klein and Baum2001), while secondary coping is better utilised when one has less control over their environment. According to the goodness-of-fit hypothesis, the correlation between one's appraisal of a situation and choice of a coping strategy will predict how well one adapts (Park, Folkman, & Bostrom, Reference Park, Folkman and Bostrom2001). Thus, the utilisation of a ‘situationally’ appropriate strategy has become extremely important to coping researchers (Zeidner & Saklofske, Reference Zeidner, Saklofske, Zeidner and Endler1996).
To date, many studies have fully supported the effectiveness of primary coping in high control situations, whereas secondary coping has been found to be maladaptive (Aldwin, Reference Aldwin2007; Compas, Connor-Smith, Saltzman, Thomsen, & Wadsworth, Reference Compas, Connor-Smith, Saltzman, Thomsen and Wadsworth2001). In low control situations, however, evidence as to the purported benefits of secondary coping is not so conclusive. Unfortunately, only a handful of studies have studied primary and secondary coping simultaneously (Forsythe & Compas, Reference Forsythe and Compas1987; Park et al., Reference Park, Folkman and Bostrom2001; Zakowski et al., Reference Zakowski, Hall, Klein and Baum2001). Nevertheless, as research has suggested secondary coping may be correlated with collectivistic cultural environments such as China (English, Zeng, & Ma, Reference English, Zeng and Ma2015), Chinese participants may be better served by secondary coping than primary coping. Kuo's (2012) study supports this claim, arguing that a person's coping strategy must be understood within the frame of their social context. Secondary coping suits Asian cultures where interpersonal harmony and self-adjustment are valued (Kuo, 2012).
When applied appropriately, the utilisation of secondary coping resulted in lower levels of anxiety for Chinese participants who lacked control over their stressors (Cheng, Cheng, Hui, & Lam, Reference Cheng, Hui and Lam1999). Alternatively, Chinese participants who deployed action-oriented strategies (primary coping) regardless of situational characteristics reported higher anxiety levels (Cheng, Cheung, Chio, & Chan, Reference Cheng, Cheung, Chio and Chan2013). Yet, secondary coping has been shown to differ within a cultural context and by the level of controllability. For example, Wang and Gan (Reference Wang and Gan2011) compared coping strategies in response to two events — a natural disaster and the Chinese college entrance exam (Gaokao), and found those who experienced a recent natural disaster reported more use of secondary coping, whereas the examination group (around 1,300 kilometres from the earthquake) reported more use of primary control engagement coping. Obviously, a natural disaster is an uncontrollable event that would most likely result in the utilisation of secondary coping to buffer the negative psychological effects. Due to the unpredictability of uprooting and leaving home behind for the first time, we argue for the benefits of secondary coping and harmful effects of primary coping on the stress-psychological adjustment relationship for college students in mainland China.
The Chinese Setting
Since the late 1970s, China's economic development has been led by millions of migrant workers who have helped transform villages into mega-cities (Fan, Reference Fan2008). Internal migration, literally called ‘floating population’ in Chinese, is often characterised as rural residents (of interior provinces) moving to (coastal) cities for work. China has the largest internal migration in the world, with 221 million or almost 20% of its country migrating (Wang, Li, Stanton, & Fang Reference Wang, Li, Stanton and Fang2010). Since 2010, only 37% of the more than 220 million migrants were moving for unskilled jobs, while pursuit of education rose sharply as a motivation for moving (Liang, Li, & Ma, 2015). From the 2000 to 2010 censuses, higher education admissions (first-year students) increased from 2.21 million to over 6.5 million (as reported in Liang et al., 2015). This influx of students to college campuses represents a rapid increase in young Chinese leaving home for the first time (Lin et al., Reference Lin, Li, Wang, Hong, Fang, Qin and Stanton2011).
While some research has examined this floating population, it has mainly centred on economic migrants, despite the fact that the trend of internal mobility has shifted to younger and educated ‘peasants’ seeking education opportunities and white collar jobs in coastal cities (Qiu, Xie, & Zhou, Reference Qiu, Xie and Zhou2004; Wang & Fan, Reference Wang and Fan2006). Unfortunately, no research has compared rural-to-urban versus urban-to-urban migration in either the context of economic or educational migration in China. Nevertheless, as the phenomenon of uprooting refers to leaving behind one's home and prior life, one would expect to find more uprooting stress for a rural-to-urban move than an urban-to-urban relocation.
Within the context of China, gender as a variable is too great for this article to adequately address. Historically, China has entrenched patriarchal values enshrined by the Confucian philosophy, best illustrated by the practice of foot binding, which did not die out until the early 20th century. Men and women were considered equal under communism, and Chairman Mao waws often quoted as saying ‘women hold up half the sky’. Nevertheless, contemporary demographic data (118 men per 100 women births in 2010) suggests that favouritism towards boys persists. Interestingly, women are well represented on college campuses, making up over 50% of enrolment (Weng, Li, & Foo Reference Weng, Li and Foo2016). In fact, the World Bank (2015) reported in 2013 that Chinese women were enrolled in higher education at a rate of 114 for every 100 men, and that this ratio appeared to be increasing dramatically since 2008.
Research on gender differences in Chinese adolescence has revealed inconsistent findings regarding coping strategies and psychological symptoms. Some studies have reported more active (primary) coping for women (Herman-Stahl & Petersen, Reference Herman-Stahl and Petersen1996; Seiffge-Krenke, Reference Seiffge-Krenke1993). Others have found no gender differences (Chan, Reference Chan1995; Kurdek, Reference Kurdek1987), and some studies have reported that men use more problem-solving (primary) strategies compared to women (Chan, Reference Chan1994; Seiffge-Krenke, Reference Seiffge-Krenke1993). One recent study by Yao et al. (Reference Yao, Xiao, Zhu, Zhang, Auerbach, McWhinnie and Wang2010) found Chinese male participants reported significantly higher mean scores on primary coping and secondary coping compared to Chinese females. In a review on homesickness, researchers previously suggested Chinese women experienced greater maladjustment due to leaving home (Scopelliti & Tibero, Reference Scopelliti and Tiberio2010; Thruber, 2007), yet women were also reported to have lower homesickness. Crystal et al. (Reference Crystal, Chen, Fuligni, Stevenson, Hsu, Ko and Kimura1994) found no gender differences for stress or depression in Taiwan, but women reported more somatic complaints compared to men. With this inconclusive stream of studies, it is difficult to make any predictions for gender differences, but it remains an important topic of study.
Purpose of Study and Predictions
The current study examined uprooting stress for educational internal migration in China by exploring the role of demographic variables on recently matriculated college students transitioning to a provincial capital city. We explored the role of hometown (urban vs. rural) and gender (female vs. male) due to Confucian values and the last several decades of social change in China (Bond, Reference Bond2010). On the basis of the literature review of uprooting stress, coping control strategies, and adjustment for Chinese internal migrations, apart from planning exploratory tests of the demographic variables mentioned above, we formulated the following three main hypotheses (H):
H1: Students from a rural residence will report higher levels of uprooting stress (H1a) and anxiety (H1b) compared with students from an urban residence.
H2: Uprooting stress will lead to greater anxiety at time 2.
H3. Coping strategies will moderate uprooting stress on anxiety over time: (H3a) primary coping will exacerbate the negative effects of stress on anxiety at time 2, while (H3b) secondary coping will buffer the stress and anxiety relationship. More secondary coping will lead to lower levels of anxiety.
Methodology
Procedure
At the beginning of the 2015 academic year, undergraduate students were recruited from three universities in eastern China to participate in a longitudinal study to examine the impact of uprooting stress on their transition to the first year in college. Three months later, the same students were given a follow-up questionnaire. All participation was voluntary and had no impact on their grades or standing in the university. Participants provided a code (birth month, father's birthday, and mother's maiden name) that was used to match participants across time 1 and time 2, but were otherwise granted anonymity.
Sample
Originally, 202 Chinese students with ages ranging from 18 to 28 (M = 20.8, SD = 2.02) agreed to participate in the present study. However, only 169 (72% female) students completed both time waves (90-day time lag) and were used in the present analyses; 52% of participants were from the countryside and had been living on campus at the university for less than 6 months.
In longitudinal analyses, attrition is always a major concern as participants may drop out for unknown reasons. The data were subjected to t-test analyses to examine the attrition rate by tested variables. Two hundred and two participants completed the first questionnaire, and 169 were matched at time 2 for a reasonably low attrition rate of 16%. Between groups t tests revealed two significant differences with those completing both questionnaires reporting higher anxiety, M = 2.75 (SD = .96) versus M = 2.32 (SD = .78); t(200) = 2.48, p = .04, Cohen's d = .35, and being older, M = 20.30 (SD = 2.02) versus M = 19.35 (SD = 1.84); t(200) = 2.69, p = .001, Cohen's d = .39, than those who only completed questionnaire at time 1. No other differences were found in attrition analyses.
Materials
The measurements used in the present study were adopted from Szabo et al. (Reference Szabo, Ward and Jose2016) and are explained below.
Psychological adjustment
The anxiety measure was drawn from Goldberg and Hiller's (Reference Goldberg and Hillier1979) General Health Questionnaire's (GHQ-28) anxiety subscale, which has been used in several studies measuring adjustment (Nakagawa & Daibo, Reference Nakagawa and Daibo1985). The measure asked individuals to respond to how they had been feeling in the past 2 weeks (e.g., ‘I have been feeling constantly under strain’ and ‘I have been under pressure’). A 7-point anchor Likert scale (1 = never to 7 = always) was used to indicate the level of anxiety perceived by participants. The internal reliability was .76 at time 1 and .80 at time 2.
Uprooting Stress
Uprooting stress, as first defined in Szabo et al. (Reference Szabo, Ward and Jose2016), included four items measuring the intensity of the stress from ‘feeling separated from friends and family from back home’, ‘leaving something of importance behind’, ‘saying goodbye to people from back home’, and ‘deciding what to bring to the new place of residence’. Participants were asked ‘In the last month, how often have you felt stressed from . . .?’ Each question was anchored on a 7-point Likert scale (1 = never to 7 = always) and was used to indicate the level of uprooting stress participants had as a result of leaving home. The Cronbach's alpha was slightly below the recommended threshold of .70, yet the inter-item and item-total correlations met the threshold for good internal consistency (Table 1).
Table 1 Correlations, Means, and Reliabilities of Variables
Note: Correlations: +.10, *p < .05, **p < .01, ***p < .001. Total means are reported. N = 169.
Primary and secondary coping
Primary and secondary coping were measured using the COPE Inventory (Carver, Scheier, & Weintraub, Reference Carver, Scheier and Weintraub1989). The four coping subscales (i.e., active coping, planning, positive reinterpretation, and acceptance) were composed of four items each. Participants gave responses to statements such as ‘I take direct action to get around the problem’ (active coping), ‘I make a plan of action’ (planning), ‘I look for something good in what is happening’ (positive reinterpretation), and ‘I accept the reality of the fact that it happened’ (acceptance) on a 4-point scale anchored by 1 = don't do this at all and 4 = do this a lot. Primary coping was measured by eight items consisting of active and planning coping. Secondary coping was measured by eight items and consisted of positive reinterpretation and acceptance dimensions. Both primary and secondary coping demonstrated good internal consistency (Table 1).
Data Analyses Plan
The preliminary analyses of the present longitudinal study consisted of zero-order correlations conducted to examine basic relationships among age, length of time at university, and core variables. We then tested for group differences using a multivariate analysis of variance (MANOVA) to investigate the rural-to-urban migration hypothesis and effects of gender. Finally, to test the goodness-of-fit hypothesis for effects of uprooting stress and coping on anxiety over time, a hierarchical regression analysis was conducted. In the regression model, step 1 included the longitudinal residual, anxietytime1. Step 2 included control variables (gender, residence type, time at university, and age). Uprooting stresstime1 was entered in step 3. In step 4, our moderators (primary and secondary coping) were added, and finally, the two interactions effects were included (uprooting stress × primary coping and uprooting stress × secondary coping) in step 5.
Data Analysis
Preliminary Analyses
Bivariate correlations (see Table 1) revealed a medium to strong positive association between anxietytime1 and anxietytime2 (r = .52, p = .001) as well as between primarytime1 and secondary coping strategies at time 1 (r = .56, p = .001). Uprooting stresstime1 was positively correlated with anxietytime1 (r = .48, p = .001) and correlated with anxietytime2 (r = .24, p = .01). Time at university was correlated with primary copingtime1 (r = .19, p = .02) and secondary copingtime1 (r = .14, p = .03). No other correlations were significant.
A two-by-two MANOVA was conducted to examine mean group differences on anxietytime1, uprooting stresstime1, and anxietytime2. This analysis allowed us to examine group differences and the possibility of an interaction effect (residence type × gender).
The results indicated a significant effect by gender, Wilk's λ =.95, F(3, 167) = 1.96, p = .01 η2 = 0.10. We found a marginally significant interaction for residence type by gender, Wilk's λ = 0.92, F(3, 167) = 1.05, p = .10, η2 = 0.09. Surprisingly, no significant main effect was found for residence type, F(3, 167) < 1.00, ns. All results are shown in Table 2.
Table 2 Testing Multivariate Analyses by Hometown and Gender on Adjustment Variables
Note: Residence type: 0 = rural, 1 = urban; Gender: 0 = female, 1 = male.
+ .10, *p < .05, **p < .01, ***p < .001.
Between-group mean analyses showed significant differences. Females reported significantly less anxietytime1, M = 2.46, SE = .13, anxietytime2, M = 2.50, SE =.12, and uprooting stress, M = 2.22, SE = .09, compared with males, anxietytime1, M = 2.89, SE =.16, F(1,167) = 4.53, p = .03, η2 = .03; anxietytime2, M=2.99, SE = .15, F(1, 167) = 6.34, p = .01, η2 = .04; uprooting stress: M = 2.57, SE =.11, F(1, 167) =6.32, p = .01, η2 = .04.
A second MANOVA (residence type × gender) was conducted to examine mean group differences by coping styles (primary copingtime1, and secondary copingtime1). The results indicated no significant main effect by gender, F(2,168) < 1.00, ns; no effect was found for residence type, F(2, 168) < 1.00, ns; and no interaction residence type by gender was found, F(2,168) < 1.00, ns. Between-group mean analyses are shown in Table 3.
Table 3 Testing Multivariate Analyses By Hometown and Gender on Coping Styles
Note: Residence type: 0 = rural, 1 = urban; Gender: 0 = female, 1 = male.
+ .10, *p < .05, **p < .01, ***p < .001.
Given that a statistically significant effect for resident type by gender interaction was not found (p = .10), hypotheses 1a and 1b were not confirmed. While no predictions were made for gender differences, analyses show females reported lower anxiety and uprooting stress compared to males. No significant differences were reported in coping styles by gender or type of residence.
Testing the Interaction Effect on Anxiety Over Time
To test the main effect for uprooting stress (H2) and interaction effects (H3, H3a, H3b), a hierarchical linear regression model was conducted (as in Szabo et al., Reference Szabo, Ward and Jose2016). In step 1, anxietytime1 was residualised on the anxietytime2, thus predicting which variables from time 1 (demographic variables, uprooting stress, and coping strategies) lead to a change in anxiety (Jose & Huntsinger, Reference Jose and Huntsinger2005). The entire model revealed in step 5 explained 40% of the variance and the interactions yielded 3% of the change in anxietytime2. In step 2, control variables accounted for 3% of change in variance of the predictor. Gender was a positive and significant predictor (β = .15, p = .01), indicating that males experienced more anxiety over time. Uprooting stress did not exert a main effect on anxiety; instead, it interacted with primary coping (β = .20, p = .01). Secondary coping exerted no significant effect in the model. Table 4 shows the hierarchical linear regression model.
Table 4 Hierarchical Regression Analysis Predicting Anxiety Over Time
Note: Dependent variable was Anxiety time 2. Residence type dummy coded: 0 = rural, 1 = urban; Gender: 0 = female, 1 = male. + p < .10; *p ≤ .05; **p ≤ .001.
The significant moderator (uprooting stress × primary coping) was further subjected to simple slope analysis and was graphed for easier interpretation (as suggested by Cohen, Cohen, & Aiken, Reference Cohen, Cohen, West and Aiken1983). In Figure 1, we can see how primary coping functions as a moderator. According to simple slopes analyses, when students use low levels of primary coping, uprooting stress is negatively related to anxiety over time, t low (167) = -1.68 p = .09, Cohens d = .26. When students use high levels of primary coping, t high (167) = -2.34, p = .02, Cohen's d = .36, uprooting stress is positively related to anxiety over time. Uprooting is positively or negatively related with anxiety depending on the level of primary coping. In addition, this effect is only present when primary coping is taken into account. In other words, Chinese freshmen who utilise high levels of primary coping for managing an uncontrollable stressor, such as uprooting stress, also experience more anxiety over time.
Figure 1 Moderation effect of primary coping time 1 on uprooting stresstime1 and anxietytime2.
In reviewing the main hypotheses, although uprooting stress is correlated with anxietytim2, it failed to predict a change in anxiety in the regression model (H2). Hypothesis 3 was partially confirmed: primary coping interacted with uprooting stress (H3a), functioning as an amplifier, but secondary coping did not interact with uprooting stress, thus rejecting H3b.
Discussion
The main aim of the present study was to examine an emerging type of educational sojourner in a rapidly changing Chinese society, together with uprooting stress, as a potentially meaningful paradigm for understanding the adaptation process of leaving one's home behind. To further build upon the research framework, we included gender differences and residence type, categorised as urban versus rural. Our longitudinal study further supported the detrimental effects of employing primary coping strategies when dealing with uncontrollable stressors, such as uprooting stress. Below we will discuss the results in detail.
First, as expected, we found a positive correlation between uprooting stress and anxiety. In line with the goodness-of-fit hypothesis, we found primary coping was not successful in reducing maladaptation, but instead amplified the negative impact of stress on anxiety. This interaction effect supports Szabo et al. (Reference Szabo, Ward and Jose2016).
Second, our findings found no main effect or interaction effect in secondary coping control, which further contributes to the mixed findings regarding secondary coping (Morling & Evered, Reference Morling and Evered2007). In fact, some recent research has found that Asians may prefer to use a primary coping control as opposed to secondary coping (Sawaumi, Yamaguchi, Park, & Robinson, Reference Sawaumi, Yamaguchi, Park and Robinson2015). A recent study by English et al. (Reference English, Zeng and Ma2015), found that secondary coping was adaptive for both Westerners and (non-Chinese) Asians living in China. Further attention needs to be given to secondary coping regarding its benefits for situational stress and the impact of environment on coping styles. Because primary coping exacerbated the negative effects of uprooting stress on anxiety and secondary coping had no significant effect, it may be necessary to look outside of intrapersonal coping strategies for viable strategies for first-year college students in China. Brewin et al. (Reference Brewin, Furnham and Howes1989) found that students who discussed their feelings of homesickness were more likely to report depression, as there was a need for affiliation. Yet more recent studies have found certain types of social support (emotional and instrumental) to be beneficial. For example, Daugherty and Lane (Reference Daugherty and Lane1999) found that establishing social bonds and friendship support networks with college peers and academic advisors during the first-year transition improved academic performance and psychological wellbeing. Having a high perception of availability for social support friends and family have also been found to alleviate homesickness (Baqutayan, Reference Baqutayan2011) We believe that social support, especially within the collective society of China, might play an vital part in alleviating symptoms of stress particularly uprooting stress. Unfortunately, we did not examine the role of social support in dealing with uprooting stress, but it should be considered in future research.
Major gender differences were found. Females were better adapted, as they reported less anxiety and stress despite slightly more utilisation of primary coping. We attributed this finding to a shift in gender differences and social norms for Chinese males and females. Lee and Padilla (Reference Lee and Padilla2014) found that Asian (Korean) males reported greater stress, noting language and cultural difficulties during their study in the United States. While our findings are interesting, we are cautious about their interpretation and suggest further investigation, as generalisation of gender differences is often problematic. Future research is necessary to disentangle gender differences in a modern Chinese context.
We found one group difference based on whether a participant's hometown was rural or urban, that uprooting stress was greater for those from the countryside. Though previous researchers have found differences between rural and urban migration (Du et al., Reference Du, Li, Lin and Tam2014; Chen, Wang, Li, & Liu, Reference Chen, Wang, Li and Liu2014), our study failed to find differences in coping strategies or anxiety. Given that our sample was a homogenous group of college students, it could be argued that for Chinese youth today, the rural-urban gap is closing due to exposure to mass media. Rosen (Reference Rosen2009) observed that media was second only to the Chinese economy in terms of creating a significant cross-generational shift of values. Chinese people born after 1980 have been found to identify more with Western values and culture (Rosen, Reference Rosen2009). Furthermore, travel, both domestic and international, has become increasingly more common in China (Fish, Reference Fish2015), and the present study failed to account for that.
Cultural Differences Within a Chinese Context
China is a large, multi-ethnic (at least 90% Han Chinese) country, and it would be a mistake to overgeneralise any similar transition to college for all Chinese students. Talhelm et al. (Reference Talhelm, Zhang, Oishi, Shimin, Duan, Lan and Kitayama2014) argued that cognitive differences (i.e., holistic thinking) can be found among different regions in China based on the agricultural suitability for growing rice or wheat; while Van de Vliet, Yang, Wang, and Ren (Reference Van de Vliert, Yang, Wang and Ren2013) suggested that the interaction between climate and wealth foster collectivism. Climate and agriculture are obviously correlated, as are urban-rural and economic development. However, as applied to the present study, the key idea is that while our entire sample is leaving their hometown to attend college, we posited that a transition from a rural environment to an urban one would be more stressful than an urban-to-urban transition because of the implicit cultural distance and/or a more acute sense of uprooting. Unfortunately, the findings failed to confirm our expectation.
We suspect that the effect of secondary coping was confounded by regional differences, as the majority of the participants (85%) came from a northern province that can be considered a ‘wheat’ province (as defined by Talhelm et al., Reference Talhelm, Zhang, Oishi, Shimin, Duan, Lan and Kitayama2014). Wheat agriculture fosters a more independent (analytic thought) and individualistic orientation, which could be linked with a context where primary coping styles are more effective. Migrating from a wheat province to a rice province (more holistic and possibly more secondary) could result in coping styles that are (mal-)adaptive. Further research needs to be done to explore coping styles and adaptation trajectories based on these within-culture differences of collectivism versus individualism — for example, comparing the relationship among uprooting stress, anxiety, and coping strategies of northern Chinese moving to southern universities and vice versa. This is especially important, as China's internal migration is also largely interprovincial (Fan, Reference Fan2008).
Limitations and Conclusions
It is important to note several limitations of this study, and given the nature of our sample and methodology, it is of extreme importance to caution the interpretation of the reported findings. First, we relied on self-reported measures to assess uprooting stress, coping controls and anxiety, which may lead to measurement bias, though this has been a major method for acculturation research (Berry, Reference Berry, Wong and Wong2006). The present study also failed to examine the university site effect, as participants attended three different university campuses. In order to fully unpack the differences and understand the role of coping, it would be best for the researchers to examine regional differences more appropriately and systematically (north-south differences, east-west differences, and rural vs. urban differences) as suggested by Ruan, Xie, and Zhang (Reference Ruan, Xie and Zhang2015). Finally, by only studying educational migration, we cannot extend the results to build on the literature concerning the more prevalent economic migration in China.
Despite these limitations, the present longitudinal study replicated findings from Szabo et al. (Reference Szabo, Ward and Jose2016), which examined key demographic variables that undergird major aspects of Chinese society, and further advanced coping research to examine the understanding of coping control in both reducing uprooting stress and other types of stress during a college transition. In light of the growing trend of internal and international migration, we examined an under-researched group of educational migrants and answered the call for coping research outside of a Western context (Kuo, Reference Kuo2013).
The present study has clear implications for higher education in China as over 28 million students attend college each year. University students often seek out help due to homesickness or depression. As the present study suggests that primary coping leads to more maladaptation, practitioners at universities should pursue alternative tracks, such as cognitive behavioral therapy or directing students to support groups. Considering the prevalence of movement of people across international and intranational borders, the results of this study and other studies of coping with stress indicate that researchers and practitioners need to further understand coping and its relationship to transitional processes for successful and positive outcomes for migrants.
Conflict of Interest
None.
Ethical Approval
This research project was approved by the Human Ethics Committee of the Department of Psychology and Behavioral Sciences at Zhejiang University, Hangzhou, China. Informed consent was obtained from all individual participants in this study.
Acknowledgments
The authors are extremely grateful for the constructive comments on the earlier versions of this article from two anonymous reviewers. Authors would also like to thank Greg DuBois for his helpful comments and suggestions.
Open access
