Systematic Review ARTICLE
What Really Matters for Loneliness Among Left-Behind Children in Rural China: A Meta-Analytic Review
- 1Institute of Developmental Psychology, Beijing Normal University, Beijing, China
- 2Department of Psychology, Social and Health Psychology Research Center, Guangzhou University, Guangzhou, China
- 3Eliot-Pearson Department of Child Study and Human Development, Institute for Applied Research in Youth Development, Tufts University, Medford, MA, United States
In rural China, left-behind children are likely to suffer chronic loneliness. Research has identified a variety of factors that may be associated with loneliness among these children. A meta-analysis is needed to address the empirical inconsistencies and examine the strength of relations between different factors and loneliness. The current meta-analysis included 51 studies on predictors of loneliness published from 2008 to 2017. Results showed that one individual factor (social anxiety) is a key risk factor for loneliness, whereas eight individual (older age, self-esteem, resilience, extroversion) and contextual factors (family functioning, parent–child relationship, peer relationship, social support) serve as protective factors in predicting loneliness. In addition, boys were more likely to feel lonely than girls. Findings and implications of this study were discussed.
In some developing countries (e.g., China, Philippines, Mexico, Sri Lanka), millions of parents leave their children and migrate to other regions or countries for jobs (Wen and Lin, 2012; Givaudan and Pick, 2013; Siriwardhana et al., 2015). Leaving children behind has become a widespread phenomenon (Duan and Zhou, 2005; Dillon and Walsh, 2012) impacting children's development in both the short and long term (Lu and Treiman, 2011; Wen et al., 2015; Su et al., 2017). In China, left-behind children refer to those children under 18 years old who have been left-behind in their rural hometown when one or both parents migrate elsewhere to work (Duan and Zhou, 2005; Su et al., 2013). By the end of 2010, it was estimated that there were more than 61 million left-behind children, accounting for 37.70% of rural children and 21.88% of the child population in China (All China Women's Federation., 2013). Of these, 32.67% were in the care of their grandparents, 3.3% were cared for by other relatives, and 4% had no guardian at all (All China Women's Federation., 2013). Whereas parental migration brings economic benefits to left-behind children, it has deleterious impacts on the development of these children (Luo et al., 2009; Antón, 2010; Givaudan and Pick, 2013; Nguyen, 2016). Previous meta-analyses focusing on left-behind children in China suggested that parental migration had negative impacts on children's mental health, psychological well-being, and academic achievements (Wang and Mesman, 2015; Zhao and Yu, 2016), which is consistent with research on left-behind children in other countries (e.g., Philippines, Sri Lanka, Ecuador, Mexico) (Valtolina and Colombo, 2012).
Loneliness, which is conceptualized as an aversive state of discrepancy between desired and experienced social relationships (Peplau and Perlman, 1982), is identified as a typical developmental problem that left-children are likely to experience (Shen et al., 2015). The theory of loneliness and social connection posits that weak family connections are associated with emotional and social loneliness (Weiss, 1973; Cacioppo et al., 2015). According to this perspective, left-behind children who have experienced prolonged physical separation with their parent(s), are more vulnerable to loneliness. Indeed, a survey conducted in six provinces in China found that ~25% of left-behind children reported high levels of loneliness (Yin, 2014). In a cross-sectional study, left-behind children were 2.5 times more likely to suffer from loneliness, compared to children of non-migrant families (Jia and Tian, 2010); furthermore, compelling evidence from a meta-analysis by Chen et al. (2017) showed that left-behind children had a higher level of loneliness than their counterparts (d = 0.29).
A chronic and painful state of loneliness is harmful to mental and physical health (Heinrich and Gullone, 2006; Qualter et al., 2015; Lempinen et al., 2018); in fact, it may elevate the risk for depression (Cacioppo et al., 2015), mortality (Holt-Lunstad et al., 2015), social withdrawal, and suicidality (Schinka et al., 2013), even damaging the immune, cardiovascular, and nervous systems (Cacioppo et al., 2015; LeRoy et al., 2017). As for left-behind children, a sense of yearning for their parents and chronically high loneliness result in a constellation of mental problems, including conduct problems (Yu, 2017) and suicide attempts (Chang et al., 2017).
Recognizing the detrimental impact of loneliness in the development of left-behind children, an increasing number of studies have focused on individual and contextual factors that are related to loneliness, such as gender (Fan et al., 2016), age (Yue et al., 2014), self-esteem (Song et al., 2017), and family functioning (Zhao, 2013). However, a few gaps exist in the literature. First, mixed findings have been found in terms of predictors (e.g., gender). Second, most research showed a lack of theoretical framework about the pathways between these factors and loneliness. Third, based on these studies, it is difficult to tell what factors are more important for left-behind children. To address these gaps, a theory-based meta-analytic approach can be used to review the literature and to examine the influence of multiple factors on loneliness. Such an approach may contribute to interventions and policies that aim to reduce the risk of loneliness among left-behind children in China. Although a previous meta-analysis has identified some individual and contextual variables (e.g., age, self-esteem, social support) that are associated with loneliness in adolescence (Mahon et al., 2006), it is unclear whether these findings could account for loneliness among left-behind children.
Accordingly, we conducted a meta-analysis on individual and contextual factors associated with loneliness among left-behind children in China. The present research is grounded in the ecological systems framework, which emphasizes the connections between individual and environmental systems in understanding human development (Bronfenbrenner, 1979). According to the ideas of ecological systems framework, we should integrate multiple processes of individual functioning and multiple developmental contexts to better understand the risk or protective factors for loneliness among left-behind children. To be more specific, demographic, and intrapersonal psychological variables can be organized as individual level factors, and family-, school-, or community-related variables can be organized as contextual level factors. With respect to left-behind children, many individual and contextual factors have been found to be associated with loneliness (Shen et al., 2015). Next, we will give an overview of individual and contextual factors that have been identified to be associated with loneliness among left-behind children.
Factors Associated with Loneliness in Left-Behind Children
Loneliness has been found to be associated with a variety of individual factors of left-behind children, including demographic characteristics (e.g., gender, age) (Liu et al., 2008; Zhao and Shen, 2011), intrapersonal psychological factors (e.g., self-esteem) (Fan et al., 2014), and emotion-related problems (e.g., social anxiety) (Liao et al., 2014). To be specific, age may alter children's vulnerability to loneliness. Existing studies suggested that older left-behind children may experience lower levels of loneliness (Zhao and Shen, 2011; Yue et al., 2014). Gender also has been regarded to play a role in the development of loneliness among left-behind children. Previous findings on the association between them were, however, mixed: some researchers found that boys report more loneliness that of girls (Xu, 2008; Sun et al., 2010; Fan et al., 2016), whereas others studies found no gender difference or opposite results (Liu et al., 2007; Qi and Jia, 2010). Besides, some intrapersonal psychological characteristics were found to correlate with lower levels of loneliness among left-behind children, including high self-esteem (Fan et al., 2014), resilience (Ai and Hu, 2016), psychological capital (Fan et al., 2017), positive appraisals of adversity (Zhao et al., 2013), positive coping styles (Liao et al., 2014), extroversion (Fan et al., 2014), and hope (Fan et al., 2016). Thus, these inherent factors may play important roles in preventing loneliness. In addition, experiencing social anxiety and feeling lonely are common and interrelated internalizing problems in child and adolescence (Jones et al., 1990). Existing studies have noted that social anxiety was positively associated with loneliness among left-behind children (Yuan et al., 2014; Ren et al., 2017). These findings may imply that social anxiety is a risk factor for experiencing loneliness or, vice versa, that feeling lonely aggravates children's social anxiety.
In addition, a growing concern has arisen that many factors within their ecological contexts (e.g., family, school) can have a substantial influence in loneliness among left-behind children (Shen et al., 2015). Family is one of the key contexts that may provide resources and or challenges that may influence children's perception of loneliness (Sharabi et al., 2012). Research has indicated that the levels of loneliness perceived by left-behind children are significantly related to how well their families function (Xie, 2008; He, 2010; Yue et al., 2014). Positive family functioning may protect left-behind children from the impacts of loneliness (Zhong et al., 2010) whereas a dysfunctional family atmosphere is associated with high levels of loneliness among these children (Fan et al., 2014). Moreover, high parental support and better parent-child relationships were also associated with low levels of loneliness among left-behind children (Liu et al., 2008; Zhao et al., 2015). In addition, parental migration status, which is often classified into two groups: both-parent migration and one-parent migration, may also be related to children's loneliness. Existing literature contains mixed findings on the role of parental migration status. Some studies found that children with both-parent migrating reported higher levels of loneliness than children with one-parent migrating (Duan, 2014a; Yue and Lu, 2015), whereas other studies found no difference in loneliness between these two groups (Qi and Jia, 2010; Su et al., 2013).
With regard to school context, the roles of teachers and peers are important in predicting these children's loneliness (Asher and Paquette, 2003; Galanaki, 2004). For example, multiple studies have shown that peer relationship is negatively associated with loneliness (Asher and Paquette, 2003; Chen et al., 2004; Vanhalst et al., 2014; Spithoven et al., 2017). Peer acceptance and high-quality friendships are associated with low levels of loneliness among left-behind children (Sun et al., 2010; Wang et al., 2011). In addition, some other school level factors, such as perceived support from peers and teachers (Liu et al., 2008; Zhang, 2011a), teacher-student relationships (Xu, 2008), and sense of belonging at school (Yang et al., 2016), are also related to loneliness among left-behind children.
The Present Study
The aim of this study is to address the empirical inconsistencies and examine the strength of relations between different factors and loneliness among left-behind children using a meta-analytic approach based on the ecological systems framework. Although many studies have explored the influence of individual and contextual factors on loneliness among left-behind children, there is a need to review the literature and to evaluate the effects of key factors on loneliness based on numerous studies accumulated in this field. Understanding the influence of these factors in the experience of loneliness may inform intervention programs and social policies that focus on reducing the perception of loneliness among these children. In the present study, we used a meta-analytic approach to examine the influences of these factors.
Data Sources and Search Strategy
We conducted a systematic search of the literature in both Chinese and English using several electronic databases, including China National Knowledge Infrastructure (CNKI), PubMed, Web of Science, and PsycInfo. We also conducted a literature search by using Google Scholar and searched master's theses and doctoral dissertations through the China Dissertation Database and ProQuest Dissertations & Theses. The wide variety of key words we used included left-behind child, left-behind adolescent, loneliness, predictor, protective, and risk factors (A detailed description appears in Appendix A).
Inclusion and Exclusion Criteria
We set the following criteria for articles to be included in this study: (a) the articles had to be empirical investigations of Chinese left-behind children's loneliness; (b) the study design had to be quantitative; (c) the articles had to be published or reported from 2000 to 2017 and available in Chinese or English. The year 2000 was chosen because the Chinese government and researchers initiated their focus on left-behind children issues at that time (Tan, 2011); and (d) the articles had to provide sufficient statistic information for the calculation or estimation of effect sizes (e.g., correlation, t-value, F-value, p-value). Articles were excluded on any of the following grounds: the studies (a) took the form of a review, a case study, a qualitative report, or a comment; and (b) reported only loneliness prevalence and did not examined individual or contextual predictors of loneliness.
Coding of Studies and Quality Assessment
A coding protocol was designed to guide coding and information retrieval. The following information was extracted from each eligible study: author name, study design, sample size, gender, location, age range, age group (elementary school students: Grades 1–6 or age 6–12 years; junior high school students: Grades 7–9 or age 13–17 years), year of publication, publication type, measure of loneliness, and estimated effect size. The eligible studies were subjected to a methodological quality assessment by two coders (the first author and third author), using a 14-item instrument, a modified quality index based on prior literature (Downs and Black, 1998; Ferro and Speechley, 2009). One item (i.e., “Were the staff, places, and facilities where the patients were studied representative of the treatment the majority of patients receive?”) was deleted from Ferro and Speechley (2009) revised quality checklist because it was inappropriate in the context of left-behind children's loneliness. The quality checklist assessed four aspects of methodological quality: reporting (e.g., “Is the hypothesis/objective of the study clearly described?”), external validity (e.g., “Were the participants asked to participate in the study representative of the entire population from which they were recruited?”), internal validity (e.g., “Were the main outcome measures used valid and reliable?”), and power (“Did the study provide a sample size or power calculation to detect important effects where the probability value for a difference being due to chance is < 0.05?”). A detailed description of the modified quality index appears in Appendix B. Each item was scored 0 (no / unable to determine) or 1 (yes). The maximum score achievable was 14. Studies with higher scores indicated higher methodological quality. All eligible studies were reviewed by two coders to settle on the most appropriate coding. Differences in interpretation were resolved through discussion with a correspondence author to reach an agreement.
Effect Size of Calculation
In the current review, we used Pearson's correlation coefficient r as the effect-size index for this meta-analysis. For studies that presented data as means and standard deviations, or inferential statistics, such as t, F, or p-values, results were converted to Pearson's correlation coefficient r using the ES calculator provided by Wilson (2001). For the effect size of a longitudinal study at several different time points, we chose the effect size of the time point with the largest sample size. Furthermore, according to the shifting unit of analysis approach (Cooper, 2010), the effect sizes of support from different sources (e.g., father, mother, peer, and teacher) (Liu et al., 2008) were combined into an effect size of social support; the effect sizes of father-child relationship and mother-child relationship (Zhang, 2011b) were combined into an effect size of parent-child relationship. We used Cohen's guidelines to interpret the effect size, where r of at least 0.10 = small, 0.30 = medium, and 0.50 = large (Cohen, 1992).
Method of Meta-Analysis
A meta-analysis was conducted for each predictor where at least two independent studies reported a measure of effect size. Other predictors were excluded if only one study was available, including cognitive appraisals of struggles associated with being left-behind (Zhao and Shen, 2011), teacher-student relationship (Xu, 2008), sense of belonging at school (Yang et al., 2016), core self-evaluation (Zhao, 2015), dysfunctional family atmosphere (Fan et al., 2014), beliefs about adversity (Zhao et al., 2013), hope (Fan et al., 2016), psychological capital and stress (Fan et al., 2017), family abuse and neglect (Duan and Zhang, 2014), and coping styles (Liao et al., 2014). We performed this meta-analysis using comprehensive meta-analysis software (Borenstein et al., 2006). A separate meta-analysis was performed for each factor. In the meta-analysis we used random effects models. The assumption underlying fixed effects models is that one true effect size exists in all eligible studies, but random effects models allow that true effect could vary across studies (Borenstein et al., 2009). Random-effect meta-analyses were, therefore, generally more appropriate for review in this meta-analysis.
To examine the presence of heterogeneity, we computed the Q statistic (a measure of weighted squared deviations), I2 (the ratio of true heterogeneity to total observed variation), and τ2 (between-studies variance) (Borenstein et al., 2009). The following guidelines were used to interpret I2: low heterogeneity, I2 = 25%; moderate heterogeneity, I2 = 50%; high heterogeneity, I2 = 75% (Higgins et al., 2003).
Subgroup analysis was undertaken to explore whether potential moderator variables could account for significant variability among effect sizes. Four significant predictors of loneliness (gender, self-esteem, peer relationship, and social support) were tested for moderating effects. Other predictors (age, resilience, extroversion, social anxiety, family functioning, and parent–child relationship) were not considered in the subgroup analysis because of the small number of studies. Two potential moderator variables (i.e., age group and study quality) in each factor were tested. First, we tested age group difference in effect sizes because prior literature has shown that older age children experienced less loneliness (Zhao and Shen, 2011; Yue et al., 2014). Second, given that study quality may vary across studies and may affect the findings, we also tested its moderating effects on effect sizes.
Characteristics of Included Studies
In total, 51 studies published from 2008 to 2017 were included in the current review, with 96 effect sizes. A summary of the studies appears in Table 1.
Among the 51 studies, except for one longitudinal design (Fan et al., 2014), the other studies were all cross-sectional. Study sample sizes ranged from 94 to 985. As for type of publication, 43 were journal articles, 7 were master's theses, and one was a conference article. With respect to measures of children's loneliness, the most frequently used scale (40 studies, 78.34% of studies eligible in this meta-analysis) was the Chinese version of Children's Loneliness and Social Dissatisfaction Scale (Asher et al., 1984). Other measures included the Loneliness Scale of Adolescents (Zou, 2003), revised version of UCLA Loneliness Scale (Russell, 1996), and Left-behind children's Loneliness Questionnaire (Yue et al., 2014). Figure 1 shows the complete selection process.
The Outcome of Meta-Analysis
The factors associated with loneliness in left-behind children appear in Table 2, illustrating the number of studies, effect size, and 95% confidence intervals. Table 2 also provides information on heterogeneity and publication bias. The forest plots diagrams for each meta-analysis are presented in Appendix C.
Table 2. Meta-analyses of individual and contextual factors for loneliness among left-behind children.
Meta-analyses on the associations between several individual factors (demographic variables, self-esteem, resilience, personality traits, and social anxiety) and loneliness among left-behind children were conducted. First, we tested the relationships between demographic variables and loneliness. We found that some demographic variables (i.e., gender, age) were associated with loneliness among left-behind children. Gender showed a minimal effect size (k = 20, r = 0.07; 95% CI: 0.03–0.11, p < 0.01). Specifically, boys were more likely to be lonely than girls. Moreover, there was moderate heterogeneity in effect sizes between studies (Q = 55.63, p < 0.0001, I2 = 68.84%, τ2 = 0.01). Subgroup analysis showed that effect sizes did not vary by age groups (Qb = 3.09, pb = 0.21). Meta-regression showed that the study quality rating score could not account for heterogeneity (slope = 0.01, p = 0.63). With respect to age, older left-behind children experienced less loneliness with a small effect size (k = 2, r = −0.14; 95% CI: −0.22 to −0.05, p < 0.01). The overall effect size was not heterogeneous (Q = 0.18, p = 0.68, I2 = 0, τ2 < 0.0001).
Second, we conducted meta-analysis to examine the associations between self-esteem and loneliness. Results showed that higher self-esteem was associated with less loneliness with a medium effect size (k = 9, r = −0.42; 95% CI: −0.51 to −0.33, p < 0.0001). Moderate heterogeneity was found in the studies (Q = 71.681, p < 0.0001, I2 = 88.84%, τ2 = 0.02). Subgroup analysis showed that there was no significant difference in effect sizes across different age groups (Qb = 0.28, pb = 0.59). Meta-regression showed that the study quality rating score could not account for the heterogeneity (slope < 0.01, p = 0.84).
Third, children with higher levels of resilience experienced less loneliness with a medium effect size (k = 4, r = – 0.37; 95% CI: −0.48 to −0.24, p < 0.0001). There was high heterogeneity in effect sizes across studies (Q = 30.84, p < 0.001, I2 = 90.27%, τ2 = 0.02).
Fourth, extroverted children experienced less loneliness with a medium effect size (k = 2, r = −0.40; 95% CI: −0.48 to 0.32, p < 0.0001). No heterogeneity was found across studies (Q = 0.25, p = 0.62, I2 = 0, τ2 < 0.0001).
Finally, children with higher levels of social anxiety experienced more loneliness with a medium effect size (k = 3, r = 0.49; 95% CI: 0.43 to 0.55, p < 0.0001). Moderate heterogeneity was found across studies (Q = 3.72, p = 0.16, I2 = 46.21%, τ2 <0.01).
We also examined the predicting effects of several contextual factors, including family environment (e.g., family functioning, parent-child relationship, parental migration status), school environment (e.g., peer relationship), and social support on loneliness. As for family environment, positive family functioning was found to correlate with low levels of loneliness among left-behind children with a small mean effect size (k = 6, r = −0.27; 95% CI: −0.33 to −0.21, p < 0.0001). Moderate heterogeneity (Q = 11.57, p = 0.04, I2 = 56.80%, τ2 < 0.01) occurred. Meanwhile, positive parent-child relationship was associated with lower levels of loneliness with a medium mean effect size (k = 4, r = −0.31; 95% CI: −0.36 to −0.25, p < 0.0001). No heterogeneity was found across studies (Q = 0.33, p = 0.95, I2 = 0, τ2 < 0.0001). Finally, results showed that no differences in average correlation of loneliness in the two parental migration status (both-parent migration vs. one-parent migration) (k = 19, r = 0.004; 95% CI: −0.02 to 0.03, p = 0.74). No heterogeneity in effect size estimates (Q = 16.21, p = 0.58, I2 = 0, τ2 < 0.01) occurred. In addition, for mother-only migration (vs. father-only migration), there is no clear evidence of an association with higher levels of loneliness (k = 5, r = 0.02; 95% CI: −0.11 to 0.14, p = 0.81), and high heterogeneity (Q = 18.85, p < 0.01, I2 = 78.78%, τ2 = 0.02) occurred.
With respect to school environment, positive peer relationship was related to less loneliness with a medium mean effect size (k = 9, r = −0.45; 95% CI: −0.56 to −0.31, p < 0.0001). High heterogeneity occurred (Q = 137.63, p < 0.0001, I2 = 94.19%, τ2 = 0.06). Subgroup analysis showed no evidence that effect size differed by age group (Qb = 0.36, pb = 0.55). Meta-regression revealed that the study quality rating score could account for the heterogeneity (slope = 0.07, p < 0.0001).
In addition, social support was related to children' loneliness with a medium mean effect size (k = 8, r = −0.40; 95% CI: −0.50 to −0.28, p < 0.0001). High heterogeneity was found in the studies (Q = 91.87, p < 0.000, I2 = 92.38%, τ2 = 0.03). Subgroup analysis showed no evidence that effect size differed by age group (Qb = 1.07, pb = 0.30). Meta-regression showed that the study quality rating score could account for the heterogeneity (slope = 0.07, p < 0.0001).
No publication bias was found in the meta-analysis for all outcomes according to the Eggers test or Rosenthal's failsafe number (see Table 2). In addition, the funnel plot of each predictor was generally symmetrical (see Appendix D). In short, the risk for publication bias in this meta-analysis can be considered low.
Grounded in an ecological systems perspective, this meta-analysis review examined what factors are associated with loneliness among left-behind children in rural China. Specifically, some key individual factors and contextual factors for loneliness among these children were identified across 51 studies published between 2008 and 2017.
With respect to individual factors, demographic factors (i.e., gender, age) and intrapersonal psychological characteristics (i.e., self-esteem, resilience, extroversion, and social anxiety), were explored in this review. We found a minimal but significant gender differences in the feeling of loneliness based on 20 studies: boys experienced higher levels of loneliness as compared to girls. One possible explanation for this finding is that, suppression may benefit interpersonal adaptation in Chinese culture (Butler et al., 2007), and Chinese boys are likely to be encouraged to suppress rather to express negative emotions (Sun et al., 2010). According to gender schema theory, children may develop a schema to fit gender roles in a unique cultural environment (Martin and Halverson, 1981): “I am a boy, so I can't cry and I am tough.” What's more, suppression may increase memory for negative emotions during tense social interactions (Richards et al., 2003). Another reason may be that boys are more likely to show social and school adjustment problems than girls, which may relate to loneliness (Chen et al., 2001). Researchers, however, remain split on gender difference in the severity of loneliness in children and adolescents (Koenig and Abrams, 1999; Weeks and Asher, 2012). Therefore, researcher should further explore gender differences in loneliness in different socio-cultural contexts.
In terms of age, the findings show a small but significant effect on loneliness. Consistent with previous studies (Harris et al., 2013; Ladd and Ettekal, 2013), normative (mean-level, or average) changes in loneliness tend to decline during childhood and adolescence. The age differences in loneliness may result from age differences in children's social and cognitive competence (Laursen and Hartl, 2013). Older children may be able to reappraise change in family structure caused by parental migration and regulate the negative emotional impact. Evidence from a qualitative study indicated that the ability to construct positive meaning from parental migration increased with age (Fu and Law, 2018). Therefore, older age may be a protective factor for loneliness among left-behind children.
In addition, intrapersonal psychological characteristics are also important predictors of loneliness among left-behind children. First, the strongest effect on loneliness were related to self-esteem with a medium effect size in the current review. This finding, to some extent, supports the cognitive discrepancy model (Peplau and Perlman, 1982), which posits that children with low levels of self-esteem are likely to engage in certain irrational cognition and behaviors and will not establish and maintain satisfactory social relationships, which contribute to loneliness. Also, according to the social exclusion theory (Baumeister and Tice, 1990) and sociometer theory (Leary et al., 1995), self-esteem is an interpersonal monitor, which reflects the individual historical experience of social exclusion and inclusion, and there may be a reciprocal relationship between self-esteem and loneliness. People with low self-esteem are more likely to feel real and imagined threats related inclusion in social context, providing an explanation for why they are more likely to be lonely; in turn, lonely people may blame themselves for being isolated and thus damage the self-esteem system (Leary, 1990; Leary et al., 1995). Indeed, the emerging empirical literature have attested to this reciprocal relationship between them (e.g., Vanhalst et al., 2013; Du et al., 2019).
Second, we found that extroverted left-behind children appeared to be less lonely than introverts perhaps because extroverts have large social networks and high-quality social relationships (Levin and Stokes, 1986; Lopes et al., 2003). Third, we also found that resilience is a protective factor for children's loneliness. Resilience has been found to act as a buffer between adversity (e.g., absence of parental care, poor social support) and loneliness among left-behind children (Liu et al., 2014; Ai and Hu, 2016). Specifically, developing resilience can help children build self-confidence and optimism against the risk of loneliness (Fergus and Zimmerman, 2005; Ai and Hu, 2016).
Finally, the results of this review coincide with a prior meta-analysis (Mahon et al., 2006) in which high social anxiety was found to be a risk factor for children's loneliness. Higher levels of social anxiety have been linked with lower levels of peer relations and friendships (Greca and Lopez, 1998), and negative peer consequences related to social anxiety may increase feeling of loneliness among left-behind children. We cannot, however, elucidate the direction of the association because of most studies in this meta-analysis are cross-sectional. A reverse interpretation (that is, loneliness leads to social anxiety) and a bidirectional relationship may exist. Further longitudinal studies are needed to explore the relationships between social anxiety and loneliness among left-behind children.
As far as contextual factors are concerned, we found some significant protective factors within family and school contexts that had small to medium effect sizes through this meta-analysis. Within family context, family functioning is a key predictor of loneliness in left-behind children. Consistent with previous studies, children who perceived positive family functioning always reported lower levels of loneliness (Sturge-Apple et al., 2010; Sharabi et al., 2012). Conversely, family dysfunction (e.g., improper parenting style) can increase the likelihood of loneliness in children through insecure attachment (Rotenberg, 1999). In addition, positive parent-child relationship plays a protective role for loneliness among left-behind children. Moreover, one may suspect that both-parent migration is a risk factor and children in that situation may experience higher levels of loneliness than children with one-parent migration; however, no evidence was found for an association between parental migration status and loneliness in this meta-analysis. Further efforts are therefore required to explore the reason why the parental migration status was not related to children's loneliness.
In addition to family, school is also an important ecological context for children' loneliness. Consistent existing studies (Asher and Paquette, 2003), our results indicated that peer relationship was significantly associated with children's loneliness. According to the theory of social loneliness (Weiss, 1973), the key marker of feeling loneliness is a lack of close and satisfying relationships. As such, peer relationship is a well-established predictor of loneliness among left-behind children.
Finally, consistent with the results of a previous meta-analysis (Masi et al., 2011), multiple sources of social support play an important role in buffering the risk of children's loneliness. Higher levels of support across sources (e.g., parents, peers, teachers) may provide children with a strong sense of belonging and companionship and then alleviate the feeling of loneliness (Cavanaugh and Buehler, 2016); therefore, enhancing social support may be an important intervention strategy to reduce loneliness among left-behind children.
Moderator Analysis of Factors Associated With Loneliness
Moderator analysis for factors with moderate to high heterogeneity (resilience, social anxiety, and family functioning were excluded because of the small number of studies) showed that the study quality was a significant moderator of the relationships between peer relationship, social support, and loneliness. Specifically, higher-quality studies showed smaller effect sizes than lower-quality studies. This finding means that the methodological quality of studies should be considered when understanding these effect sizes.
Limitations and Implications
Several limitations should be noted in this review. First, most of the eligible studies involved cross-sectional data, limiting our capability to evaluate the appropriateness of causal inferences to evaluate the appropriateness of causal inferences. Future meta-analysis may be needed when more longitudinal studies emerge. Second, given that there are not many studies focusing on the predictors of loneliness, in the analyses we only examined a relatively small number of studies for each factor that was included. Third, we were not able to control for other factors when examining the effects of each factor. Further meta-analysis is needed to address these issues when more studies are filled into this field. Fourthly, meta-analysis showed some factors in high heterogeneity, suggesting that a potential moderator can explain the differences in effect sizes. Unfortunately, because of the small number of studies, we only tested two moderators, and their moderating effects were mostly not significant. Finally, based on the location characteristics of included studies (see Table 1) and previous research (e.g., He, 2008), some parents of children may work abroad, not in China, especially in the coastal areas of southeastern China (e.g., Zhejiang, Guangdong). The studies included in this review, unfortunately, didn't clearly report relevant information. Therefore, it remains unclear whether the two groups differ in these associations. Further research should focus on this issue by adopting a comparative approach.
Despite the limitations, the present study has several important implications for future research and practice focusing on the reduction of loneliness and the promotion of well-being among left-behind children. First, a developmental and ecological systems perspective can be used in examining the effects of individual and contextual variables on the development of loneliness in left-behind children in future research. Second, parenting training can be provided to help parents improve the frequency and quality of communication with children and enhance family warmth and emotional connections among family members. In addition, school personnel are encouraged to create a positive school climate (e.g., support, caring, respect, equality, positive expectations) by providing effective left-behind children development programs. For example, out-of-school time programs emphasizing on improving peer relationships, skills of seeking social support, and other intrapersonal strengths (e.g., resilience and self-esteem) can be offered to left-behind children to reduce their loneliness and promote their positive development. Moreover, social security and protection systems associated with the development of left-behind children should be established through the collaboration of families, schools, and communities and can, for example, focus on increasing investment in rural communities and education, caring for migration families, and improving caregivers' parenting skills by offering parenting lessons.
Framed within the ecological systems model, this meta-analytic review identified small to medium effects for a variety of factors that predict left-behind children's loneliness. Overall, this review identified one individual risk factor (i.e., social anxiety) and several individual (i.e., older age, self-esteem, resilience, extroversion) and contextual (i.e., family functioning, parent-child relationship, peer relationship, and social support) protective factors for loneliness among left-behind children. Despite the methodological limitations of the studies included in this meta-analysis, these findings are important for understanding how to reduce loneliness and promote well-being among left-behind children.
XC, XL, and DL collaboratively designed this meta-analysis. XC and XL conducted the literature search and coded studies. XC and DL did data analysis and wrote the first draft of the manuscript. HD provided valuable ideas and revised the manuscript. SS helped revise the manuscript.
This study was supported by the National Social Science Foundation of China (15ZDB138). Our institutions have received for open access publication fees.
Conflict of Interest Statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpsyg.2019.00774/full#supplementary-material
*Ai, H., and Hu, J. (2016). Psychological resilience moderates the impact of social support on loneliness of “left-behind” children. J. Health Psychol. 21, 1066–1073. doi: 10.1177/1359105314544992
Butler, E. A., Lee, T. L., and Gross, J. J. (2007). Emotion regulation and culture: are the social consequences of emotion suppression culture-specific? Emotion. 7, 30–48. doi: 10.1037/1528-35126.96.36.199
Chang, H., Yan, Q., Tang, L., Huang, J., Ma, Y., Ye, X., et al. (2017). A comparative analysis of suicide attempts in left-behind children and non-left-behind children in rural China. PLoS ONE. 12:e0178743. doi: 10.1371/journal.pone.0178743
Chen, M., Sun, X., Chen, Q., and Chan, K. L. (2017). Parental migration, children's safety and psychological adjustment in rural China: a meta-analysis. Trauma Viol. Abuse. doi: 10.1177/1524838017744768. [Epub ahead of print].
Chen, X., Chen, H., and Kaspar, V. (2001). Group social functioning and individual socioemotional and school adjustment in Chinese children. Merrill. Palmer. Q. 47, 264–299. doi: 10.1353/mpq.2001.0008
Chen, X., He, Y., De Oliveira, A. M., Coco, A. L., Zappulla, C., Kaspar, V., et al. (2004). Loneliness and social adaptation in Brazilian, Canadian, Chinese and Italian children: a multi-national comparative study. J. Child Psychol. Psychiatry 45, 1373–1384. doi: 10.1111/j.1469-7610.2004.00329.x
*Dong, Z., and Zhang, D. (2013). The relationship between the resilience and loneliness among left-behind children in minority regions. Chin. J. School Health. 34, 827–829. [in Chinese].
Downs, S. H., and Black, N. (1998). The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J. Epidemiol. Commun. Health. 52, 377–384. doi: 10.1136/jech.52.6.377
Du, H., Li, X., Chi, P., Zhao, S., and Zhao, J. (2019). Loneliness and self-esteem in children and adolescents affected by parental HIV: a 3-year longitudinal study. Appl. Psychol. 11, 3–19. doi: 10.1111/aphw.12139
*Duan, B. (2014a). The correlations between loneliness and social anxiety of grades 3-6 left-behind children. J. Ningbo Univ. Educ. Sci. Edn. 36, 15–19. [in Chinese].
*Duan, B. (2014b). Loneliness and social anxiety in left-behind children: The mediating role of self-concept. Mod. Prim. Second. Educ. 30, 62–65. doi: 10.3969/j.issn.1002-1477.2014.06.017 [in Chinese].
*Duan, B., and Zhang, Y. (2014). Abuse, neglect and loneliness in left-behind children: the mediating role of self-consciousness. Chin. J. School Health. 35, 1551–1553. [in Chinese].
*Fan, X. (2011). On the comparison of emotional adaptation between left-behind children of diffirent types and normal children. Chin. J. Special Educ. 71–77. doi: 10.3969/j.issn.1007-3728.2011.02.014 [in Chinese].
*Fan, X., Fang, X., Zhang, S., Chen, F., and Huang, Y. (2014). Mediation of extroversion and self-esteem on relationship between family atmosphere and loneliness in rural parent-absent children. Chin. J. Clin. Psychol. 22, 680–687. [in Chinese].
*Fan, X., He, M., and Chen, F. (2016). Effect of hope on the relationship between parental care and loneliness in left-behind children in rural China. Chin. J. Clin. Psychol. 24, 702–705. doi: 10.16128/j.cnki.1005-3611.2016.04.028 [in Chinese].
Fan, X., Yu, S., Peng, J., and Fang, X. (2017). The relationship between perceived life stress, loneliness and general well-being among the left-behind rural children: Psychological capital as a mediator and moderator. J. Psychol. Sci. 40, 388–394. doi: 10.16719/j.cnki.1671-6981.20170221 [in Chinese].
Fergus, S., and Zimmerman, M. A. (2005). Adolescent resilience: a framework for understanding healthy development in the face of risk. Annu. Rev. Public Health. 26, 399–419. doi: 10.1146/annurev.publhealth.26.021304.144357
Ferro, M. A., and Speechley, K. N. (2009). Depressive symptoms among mothers of children with epilepsy: a review of prevalence, associated factors, and impact on children. Epilepsia. 50, 2344–2354. doi: 10.1111/j.1528-1167.2009.02276.x
Givaudan, M., and Pick, S. (2013). Children left behind: how to mitigate the effects and facilitate emotional and psychosocial development. Child Abuse Neglect. 37, 1080–1090. doi: 10.1016/j.chiabu.2013.10.019
Harris, R. A., Qualter, P., and Robinson, S. J. (2013). Loneliness trajectories from middle childhood to pre-adolescence: impact on perceived health and sleep disturbance. J. Adolesc. 36, 1295–1304. doi: 10.1016/j.adolescence.2012.12.009
*He, W. (2010). The Characteristics of Two Types of Family Relationship Disadvantaged Teenager's Family Functioning and Its Relation to Social Behavior and Loneliness. Chongqing: Xinan University.
He, Y. (2008). Report on left-children's development in oversea Chinese hometown: a case study on Qingtian, a county in Zhejiang Province. China Youth Study. 10, 53–57. doi: 10.3969/j.issn.1002-9931.2008.10.012
Holt-Lunstad, J., Smith, T. B., Baker, M., Harris, T., and Stephenson, D. (2015). Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspect. Psychol. Sci. 10, 227–237. doi: 10.1177/1745691614568352
Jones, W. H., Rose, J., and Russell, D. (1990). “Loneliness and social anxiety,” in Handbook of Social and Evaluation Anxiety., ed H. Leitenberg (Boston, MA: Springer), 247–266. doi: 10.1007/978-1-4899-2504-6_9
Koenig, L. J., and Abrams, R. F. (1999). “Adolescent loneliness and adjustment: A focus on gender differences,” in Loneliness in Childhood and Adolescence. K. eds J. Rotenberg and S. Hymel (Cambridge: Cambridge University Press), 296–324. doi: 10.1017/CBO9780511551888.015
*Kong, X., Liu, Y., and Zhang, J. (2016). Left-behind children's loneliness and self-injury: the role of social support. Chin. J. School Doctor. 30, 641–646. [in Chinese].
Ladd, G. W., and Ettekal, I. (2013). Peer-related loneliness across early to late adolescence: normative trends, intra-individual trajectories, and links with depressive symptoms. J. Adolesc. 36, 1269–1282. doi: 10.1016/j.adolescence.2013.05.004
Laursen, B., and Hartl, A. C. (2013). Understanding loneliness during adolescence: developmental changes that increase the risk of perceived social isolation. J. Adolesc. 36, 1261–1268. doi: 10.1016/j.adolescence.2013.06.003
Leary, M. R., Tambor, E. S., Terdal, S. K., and Downs, D. L. (1995). Self-esteem as an inter-personal monitor: the sociometer hypothesis. J. Pers. Soc. Psychol. 68, 518–530. doi: 10.1037/0022-35188.8.131.528
Lempinen, L., Junttila, N., and Sourander, A. (2018). Loneliness and friendships among eight-year-old children: time-trends over a 24-year period. J. Child Psychol. Psychiatry. 59, 171–179. doi: 10.1111/jcpp.12807
LeRoy, A. S., Murdock, K. W., Jaremka, L. M., Loya, A., and Fagundes, C. P. (2017). Loneliness predicts self-reported cold symptoms after a viral challenge. Health Psychol. 36, 512–520. doi: 10.1037/hea0000467
*Liao, C., Liu, Q., and Zhang, J. (2014). The correlation between social anxiety and loneliness of left-behind children in rural China: effect of coping style. Health. 6, 1714–1723. doi: 10.4236/health.2014.614204
*Liu, L. J., Sun, X., Zhang, C. L., Wang, Y., and Guo, Q. (2010). A survey in rural China of parent-absence through migrant working: the impact on their children's self-concept and loneliness. BMC Public Health. 10, 32–40. doi: 10.1186/1471-2458-10-32
Liu, X., Fan, X., and Shen, J. (2007). Relationship between social support and problem behaviors of the left-home-kids in junior middle school. Psychol. Dev. Educ. 23, 98–102. doi: 10.3969/j.issn.1001-4918.2007.03.017 [in Chinese].
*Liu, X., Hu, X., and Shen, J. (2008). The influence of different social support on the loneliness of left-behind children in rural areas. J. Henan Univ. 48, 18–22. doi: 10.3969/j.issn.1000-5242.2008.01.004 [in Chinese].
*Liu, X. Q. (2014). Self-Consciousness, Loneliness and Aggression of the Left-Behind Children. Shijiazhuang: Hebei Normal University.
*Liu, Y., Zumu, L., and Ba, L. (2014). The relationship between resilience and loneliness of left-behind children of Uygur in Kashi area. Chin. J. School Health. 35, 1397–1399. [in Chinese].
Lopes, P. N., Salovey, P., and Straus, R. (2003). Emotional intelligence, personality, and the perceived quality of social relationships. Pers. Individ. Dif. 35, 641–658. doi: 10.1016/S0191-8869(02)00242-8
Lu, Y., and Treiman, D. J. (2011). Migration, remittances and educational stratification among blacks in apartheid and post-apartheid South Africa. Soc. Forces. 89, 1119–1143. doi: 10.1093/sf/89.4.1119
Mahon, N. E., Yarcheski, A., Yarcheski, T. J., Cannella, B. L., and Hanks, M. M. (2006). A meta-analytic study of predictors for loneliness during adolescence. Nurs. Res. 55, 308–315. doi: 10.1097/00006199-200609000-00003
Nguyen, C. V. (2016). Does parental migration really benefit left-behind children? Comparative evidence from Ethiopia, India, Peru and Vietnam. Soc. Sci. Med. 153, 230–239. doi: 10.1016/j.socscimed.2016.02.021
*Qi, H., and Jia, Y. (2010). Loneliness of adolescents in rural areas and its influencing factors. Soft Sci. Health 24, 450–453. doi: 10.3969/j.issn.1003-2800.2010.05.022 [in Chinese].
*Ren, N., and Shen, L. (2008). A survey on loneliness status of children left in primary school in rural areas. China J. Health Psychol. 16, 754–756. [in Chinese].
*Ren, Y., Yang, J., and Liu, L. (2017). Social anxiety and internet addiction among rural left-behind children: the mediating effect of loneliness. Iran. J. Public Health. 46, 1659–1668.
Richards, J. M., Butler, E. A., and Gross, J. J. (2003). Emotion regulation in romantic relationships: the cognitive consequences of concealing feelings. J. Soc. Pers. Relat. 20, 599–620. doi: 10.1177/02654075030205002
*Rong, D. (2008). Loneliness and problem behavior of left-behind children in rural areas. Literat. Sci. Guid. 10, 25–28. [in Chinese].
Rotenberg, K. J. (1999). “Parental antecedents of children's loneliness,” in Loneliness in Childhood and Adolescence, eds K. J. Rotenberg and S. Hymel (Cambridge: Cambridge University Press), 176–200.
Schinka, K. C., van Dulmen, M. H., Mata, A. D., Bossarte, R., and Swahn, M. (2013). Psychosocial predictors and outcomes of loneliness trajectories from childhood to early adolescence. J. Adolesc. 36, 1251–1260. doi: 10.1016/j.adolescence.2013.08.002
Sharabi, A., Levi, U., and Margalit, M. (2012). Children's loneliness, sense of coherence, family climate, and hope: developmental risk and protective factors. J. Psychol. 146, 61–83. doi: 10.1080/00223980.2011.568987
Siriwardhana, C., Wickramage, K., Siribaddana, S., Vidanapathirana, P., Jayasekara, B., Weerawarna, S., et al. (2015). Common mental disorders among adult members of ‘left-behind' international migrant worker families in Sri Lanka. BMC Public Health. 15, 299. doi: 10.1186/s12889-015-1632-6
*Song, J. J., Zuo, B., Tian, X., and Dai, Y. E. (2017). Mediating effect of self-esteem in relationship betweenparental cohesion, peer acceptance and loneliness in left-behind children. Chin. Ment. Health J. 31, 376–381. doi: 10.3969/j.issn.1000-6729.2017.05.008 [in Chinese].
Spithoven, A. W. M., Lodder, G. M. A., Goossens, L., Bijttebier, P., Bastin, M., Verhagen, M., et al. (2017). Adolescents' loneliness and depression associated with friendship experiences and well-being: a person-centered approach. J. Youth Adolesc. 46, 429–441. doi: 10.1007/s10964-016-0478-2
Sturge-Apple, M. L., Davies, P. T., and Cummings, E. M. (2010). Typologies of family functioning and children's adjustment during the early school years. Child Dev. 81, 1320–1335. doi: 10.1111/j.1467-8624.2010.01471.x
*Su, S., Li, X., Lin, D., Xu, X., and Zhu, M. (2013). Psychological adjustment among left-behind children in rural China: the role of parental migration and parent–child communication. Child Care Health Dev. 39, 162–170. doi: 10.1111/j.1365-2214.2012.01400.x
Su, S., Li, X., Lin, D., and Zhu, M. (2017). Future orientation, social support, and psychological adjustment among left-behind children in rural China: a longitudinal study. Front. Psychol. 8:1309. doi: 10.3389/fpsyg.2017.01309
*Sun, X., Zheng, X., and Zhou, Z. (2013). The relationship between friendship quality and loneliness of left-behind middle school students: the role of resilience. Educ. Res. Exp. 4, 75–79. [in Chinese].
*Sun, X., Zhou, Z., Wang, Y., and Fan, C. (2010). Loneliness of children left in rural areas and its relation to peer relationship. J. Psychol. Sci. 33, 337–340. [in Chinese].
*Sun, X. M. (2015). A Research on the Relationship among Peer Relationships, Self-Concept and Loneliness on the Left-Behind Children in the Suburbs Overlap. Wuhu: Anhui Normal University.
Valtolina, G. G., and Colombo, C. (2012). Psychological well-being, family relations, and developmental issues of children left behind. Psychol. Rep. 111, 905–928. doi: 10.2466/21.10.17.PR0.111.6.905-928
Vanhalst, J., Goossens, L., Luyckx, K., Scholte, R. H., and Engels, R. C. (2013). The development of loneliness from mid to late adolescence: trajectory classes, personality traits, and psychosocial functioning. J. Adolesc. 36, 1305–1312. doi: 10.1016/j.adolescence.2012.04.002
Vanhalst, J., Luyckx, K., and Goossens, L. (2014). Experiencing loneliness in adolescence: a matter of individual characteristics, negative peer experiences, or both? Soc. Dev. 23, 100–118. doi: 10.1111/sode.12019
*Wang, H. (2008). A comparative study of left-behind children and non-left-behind children in rural areas. J. Educ. Sci. Hunan Normal Univ. 6, 99–103. doi: 10.3969/j.issn.1671-6124.2007.04.026 [in Chinese].
*Wang, L. (2013). The Loneliness of Left-Behind and Non Left-Behind Children in Rural Areas of Shandong Province. Jinan: Shangdong University.
*Wang, S. (2011). A Study on the Relationship Between Loneliness and Personality Traits and Social Support in Left-Behind Children. Xining: Qinghai Normal University.
Weeks, M. S., and Asher, S. R. (2012). “Loneliness in childhood: Toward the next generation of assessment and research,” in Advances in Child Development and Behavior Vol. 42, ed J. B. Benson (San Diego, CA: Academic Press), 1–39.
Wen, M., and Lin, D. (2012). Child development in rural China: children left behind by their migrant parents and children of nonmigrant families. Child Dev. 83, 120–136. doi: 10.1111/j.1467-8624.2011.01698.x
Wilson, D. B. (2001). Effect Size Determination Program [Computer software]. Retrieved from: http://mason.gmu.edu/~dwilsonb/ma.html (retrieved January 30, 2019).
*Wu, W., Chen, X., Wen, G., and Luo, L. (2010). Relationship between loneliness and social support of the rural left-behind children in chaoshan area. China J. Health Psychol. 18, 1220–1122. doi: 10.13342/j.cnki.cjph.2010.10.028.1220 [in Chinese].
*Xiao, F., and Zhang, Q. (2015). The relationship between social support and loneliness among left-behind children in Jiangxi province. J. Pingxiang Univ. 32, 110–113. [in Chinese].
*Xie, H. (2008). Study on Loneliness of the Parent-Absent Middle School Students in the Conuntryside and Its Family Factors. Chongqing: Xinan University.
*Xu, J. (2016). An investigation on the influencing factors of loneliness in left-behind children in rural areas. J. Shangqiu Vocat. Tech. College 15, 108–111. [in Chinese].
*Xu, S. (2008). Loneliness and School Environment in Left-Behind Children. Wuhan: China University of Geosciences.
*Xu, X. (2014). The Relationship of Left-Behind Middle School Student's Self-Awarenessm, Peer Relationship and Loneliness. Hefei: Anhui Normal University.
*Yang, Q., Yi, L., and Song, W. (2016). Loneliness, family cohesion and school belonging in left-behind children. Chinese Mental Health J. 30, 197–201. doi: 10.3969/j.issn.1000-6729.2016.03.008 [in Chinese].
*Yang, Q. F., Luo, H. W., Hou, Y. T., and Zhang, J. F. (2013). The relationship between loneliness and personality in left-behind middle school students. Health Med. Res. Pract. 10, 63–65. [in Chinese].
Yin, P. (2014). Charity Fund Launched to Help Needy People in Zhejiang. Women of China. Retrieved from: http://m.womenofchina.cn/womenofchina/xhtml1/news/charity/1410/1055-1.htm (retrieved January 30, 2019).
*Yu, L. (2017). School climate, loneliness, and problem behaviors among left-behind children in Zhaoqing. Chin. J. School Health. 38, 942–945. doi: 10.16835/j.cnki.1000-9817.2017.06.047 [in Chinese].
*Yuan, B., Jin, C., and Yang, S. (2014). Loneliness and social anxiety of different types of rural left-behind children. China J. Health Psychol. 22, 1564–1566. doi: 10.13342/j.cnki.cjhp.2014.10.051 [in Chinese].
*Yue, L., Sun, F., and Liu, Y. (2014). A survey on loneliness of left-behind children in rural areas of Hunan province. Matern. Child Health Care China. 29, 3320–3323. doi: 10.7620/zgfybj.j.issn.1001-4411.2014.20.43 [in Chinese].
*Yue, S., and Lu, X. (2015). Comparison of social support and loneliness among left-behind children in eastern and western China. Chin. J. School Health. 36, 1662–1664. [in Chinese].
*Zhang, L. (2011a). The relationship between social support and loneliness of left-behind children in rural areas. Chin. J. Spec. Educ. 80–84. doi: 10.3969/j.issn.1007-3728.2011.05.014 [in Chinese].
*Zhang, L. (2011b). A study of social relationships and loneliness of the rural left-behind children. Chin. J. Clin. Psychol. 19, 123–125. [in Chinese].
*Zhang, Q., and Hu, J. (2015). Loneliness and self-concept of left-behind children in Jianxi province. J. Nanchang Normal Univ. 36, 85–89. [in Chinese].
Zhao, F., and Yu, G. (2016). Parental migration and rural left-behind children's mental health in China: A meta-analysis based on mental health Test. J. Child Fam. Stud. 25, 3462–3472. doi: 10.1007/s10826-016-0517-3
*Zhao, J. (2013). Parental monitoring and left-behind children antisocial behavior and loneliness. Chin. J. Clin. Psychol. 21, 500–504. [in Chinese].
*Zhao, J., Liu, X., and Shen, J. (2008). Left-home adolescents'perception of social support networks and their associations with individual depression and loneliness: variable-centered and person-centered perspectives. Psychol. Dev. Educ. 24, 36–42.
*Zhao, J., Liu, X., and Wang, M. (2015). Parent–child cohesion, friend companionship and left-behind children's emotional adaptation in rural China. Child Abuse Neglect. 48, 190–199. doi: 10.1016/j.chiabu.2015.07.005
*Zhao, J., Liu, X., and Zhang, W. (2013). Peer rejection, peer acceptance and psychological adjustment of left-behind children: the roles of parental cohesion and children's cultural beliefs about adversity. Acta Psychol. Sin. 45, 797–810. doi: 10.3724/SP.J.1041.2013.00797
Zhao, J., and Shen, J. (2011). Relationship between cognitive appraisals for left-home hassles, depression and loneliness in rural left-home-children. Chin. J. Clin. Psychol. 19, 515–517. [in Chinese].
*Zhao, Y. (2015). Core self-evaluations as a mediator of the relationship between social support and mental health among left-behind children. J. Xingyi Normal Univ. National. 3, 86–90.
*Zhong, X., Chen, X., and Xiong, X. (2010). A study on the relationship between family functioning and loneliness of left-behind middle school students. J. Inner Mongol. Normal Univ. 23, 42–44. doi: 10.3969/j.issn.1671-0916.2010.04.012 [in Chinese].
*^References marked with an asterisk indicate studies included in the meta-analysis.
Keywords: loneliness, left-behind children, predictors, meta-analysis, systematic review
Citation: Chai X, Du H, Li X, Su S and Lin D (2019) What Really Matters for Loneliness Among Left-Behind Children in Rural China: A Meta-Analytic Review. Front. Psychol. 10:774. doi: 10.3389/fpsyg.2019.00774
Received: 19 October 2018; Accepted: 21 March 2019;
Published: 18 April 2019.
Edited by:Yvette Renee Harris, Miami University, United States
Copyright © 2019 Chai, Du, Li, Su and Lin. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Danhua Lin, email@example.com