The Effect of Left-Behind Experience and Self-Esteem on Aggressive Behavior in Young Adults in China

Abstract

There is little investigation on the independent effects of left-behind experience (LBE) on self-esteem and aggressive behavior in Chinese young adult populations, or the interaction effects of LBE and self-esteem on aggressive behavior. Thus, a school-based health survey was conducted in Anhui province in China in 2017. A total of 4,154 college students completed standard questionnaires which contain details of left-behind–related characters, self-esteem, aggressive behavior, and sociodemographic profile. Of included students, 55.3% were those with LBE (LBEs). Compared to students without left-behind experiences (NLBEs), LBEs had significantly increased scores of aggressive behavior and decreased score of self-esteem. The increased aggression in LBEs was highly related to longer left-behind duration, younger age of left-behind for the first time, and decreased self-esteem. On the other side, the aggressive behavior was demonstrated negatively correlated with self-esteem in both LBEs and NLBEs. There was an interaction effect of left-behind duration and self-esteem on physical aggression and of frequency of contacting with parents and self-esteem on verbal aggression. Besides, the interaction of primary caregiver and self-esteem on hostility and aggression toward self were also observed, respectively. Our results indicated LBEs and low self-esteem are associated with increased risk of aggressive behavior in Chinese young adults. The increased aggressive behavior in LBEs were highly related to longer left-behind duration, younger age of left-behind for the first time and decreased self-esteem. In those LBEs with some certain left-behind–related characters, aggressive behavior decreased more prominently with the increase of self-esteem. Strategies to improve self-esteem, particularly among young adults who have certain characters of LBE, should be a significant component of prevention and interventions of aggressive behavior.

Keywords

aggressive behavior left-behind experiences self-esteem

Introduction

Young Adults With Left-Behind Experiences

China’s rapid development and urbanization process has resulted in a large number of residents of underdeveloped areas (e.g., rural areas) moving from their hometowns to cities in search of better employment opportunities. Parents generally leave their children behind with caregivers, creating a new, potentially vulnerable subgroup of left-behind children (LBC) in rural areas. LBC refer to “children who live in their original domicile, but do not live together with their parents, as either one parent or both parents have migrated” (National Bureau of Statistics of China et al., 2017). LBC in China appeared along with reform and opening up and rapid social and economic development since the early 1990s. The number of LBC abruptly increased around the year of 2000, approximately 30 million which was 8.7% of the total number of children in the country (Lv et al., 2018). Fifteen years later, it increased to 68.77 million which accounting for 25.4% of the total child population, based on data from the 2015 1.0% National Population Sample Survey (National Bureau of Statistics of China et al., 2017). As for LBC, with one or both parents working away from home, it is difficult for the children to receive emotional support and help on studies from their families, which can be detrimental to their physical and mental health (Fellmeth et al., 2018). There are some studies which indicated that LBC had increased the risk of anxiety, depression, low self-esteem, aggression, disruptive behavior, suicidal ideation, conduct disorder, substance use, wasting and stunting compared with children of non-left-behind (NLBC; Aryal et al., 2019; Fellmeth et al., 2018; Naughton et al., 2013; Tang et al., 2018).

The LBC who have grown up and admitted to university were called university students with left-behind experiences (LBEs). Although admitted to university is one of the best outcomes for LBC in some sense, LBEs are demonstrated to have more depression emotion and other mental health problems, lower scores in self-esteem and resilience than the college students without left-behind experiences (NLBEs; Han et al., 2018; Shi et al., 2016). Compared to NLBEs, the LBEs were also rated significantly lower on the self-efficacy, optimism, hope, and overall psychological capital (Li et al., 2018). But until now, most studies in LBEs focused on psychological health rather than behavioral problems.

Aggressive Behavior

Aggression, in socio-psychological perspective, was defined as a psychological phenomenon which depicts a large category of intentional behavior of bringing harm to other or self by means of physical or verbal attacks (Comer & Gould, 2011). The behavior of aggression can lead to violence such as school bulling, murder, self-injury, and suicide (Anderson & Bushman, 2002; Delva-Tauili’ili, 1995). Being a bully in childhood was demonstrated to be correlated with higher aggression in adulthood (Matthews et al., 2017). Furthermore, there are researchers who thought that bullying may be a predictor of offending, delinquency, violence, and other antisocial behavior in adulthood (Bender & Lösel, 2011). Meanwhile, some researches revealed that the increasing of aggression may increase the risk for suicide ideation and suicide attempt (Huz et al., 2016). Correlated with high level of testosterone and low levels of serotonin, aggression is believed to be affected by both genetics and social learning (Ferguson & Beaver, 2009; Lansford et al., 2012; Terburg et al., 2009). Parenting environment was a crucial factor for aggression (Waller et al., 2018). Children who were neglected manifest a series of serious aggressive behavior (such as disruptive behavior, angry, hostility, and oppositional behavior) and low self-esteem (Naughton et al., 2013). A systematic review and meta-analysis in adolescents aged 6 to 11 years found that the LBC scored higher in aggressive behavior subscales than NLBC (Qu et al., 2018). Another study in students of Grades 4 to 6 of primary school found that the scores of total aggression and five subscales (physical aggression, verbal aggression, anger, hostility, and indirect aggression) of LBC were higher than NLBC (Zhang et al., 2018). However, it is not clear if aggressive behavior is increased in LBEs and what kind of left-behind characters will moderate the effect.

The Association Between LBEs and Self-Esteem and Aggressive Behavior

The general aggression model theory thinks that “person factors and situational factors lead to the occurrence of final aggressive behavior through influencing personal present internal state (e.g., self-esteem) and subsequent appraisal and decision processes” (Anderson & Bushman, 2002). Previous researches had demonstrated that low self-esteem was more prevalent among children, adolescents, and young adults who had LBEs (Han et al., 2018; Naughton et al., 2013; Tang et al., 2018). Self-esteem, a significant personality variable, affects our emotion, cognition, motivation, and behavior, has been shown negatively correlated to aggression. Garofalo et al. (2016) found that low self-esteem was related to physical aggression, verbal aggression, anger, and hostility in community sample and violent offenders. Study in the United States and New Zealand reveals that adolescents and college students who were in low self-esteem were more aggressive (Donnellan et al., 2005). What’s more, low self-esteem was a predictor of total aggression, physical aggression, verbal aggression, anger, and hostility (Donnellan et al., 2005). Trzesniewski et al. (2006) also found that adolescents with low level of self-esteem exhibited more criminal behavior (e.g., assault, robbery, manslaughter, and rape) during adulthood. However, there also exist studies showing the positive association between self-esteem and aggression (Baumeister et al., 1996). Further large population-based studies need to be carried out to study the relationship between self-esteem and aggression. On the other side, since self-esteem was demonstrated to be affected by LBE and also had effect on aggression, it would be natural to suggest that self-esteem may moderate the relationship between left-behind–related factors and aggressive behavior. However, no study was conducted on the interaction of LBE and self-esteem on aggressive behavior in college students who had LBE. What’s more, the detailed interaction of self-esteem and certain characters of LBE underlies aggression were still unknown.

Based above, our study first investigated the independent effects of LBE, left-behind–related characters and self-esteem on aggression in Chinese college students. Then we examine the interaction effects of left-behind–related characters and self-esteem on aggression. Meanwhile, we also want to see the effect of LBE on self-esteem on the college students.

Method

Sample and Procedures

Anhui Province, one of the largest immigration provinces in China, was chosen as our study fields for data collection. A survey indicated that Anhui ranked second in terms of the number of LBC, with a total of 5.71 million, which represented more than 40.0% of the rural children in the province (Duan &Yang, 2008). A three-stage random cluster-sampling approach was used to select participants. At the first stage, three regions, namely Suzhou, Hefei, and Ma’anshan, were elected, which can well demonstrate the north, middle, and south of the province. In the second stage, we chose one university or college from each region. Then, two schools within each university or college were randomly chosen in the third stage according to simple randomization. The investigation was conducted in classroom. First, teacher gave a brief introduction of the survey, then, we gave a detailed introduction of the content of the questionnaire to students who would like to take part in the survey. An informed consent was obtained from each participates. To protect the privacy of the participates, anonymous paper questionnaires were administered by trained investigators in the absence of teachers, and each participates filled in the questionnaire separately. The participants were required to complete the paper questionnaires during class time. In total, 4,350 students, from Grades 1 to 4, were contacted for this survey. One hundred ninety-six (4.5%) were excluded from the study due to either absence from school on the day of the survey or unwillingness to reply to the questionnaire, or high levels of missing data or discrepant responses or obviously fictitious. Thus, a total sample of 4,154 (95.5%) participants was included in the analyses. The design and data collection procedures were approved by the Ethics Committee of Anhui Medical University. The survey was conducted from September to November in 2017.

Measures

Aggressive behavior

We used Buss-Perry Aggression Questionnaire (AQ), to assess the occurrence of aggression. The Chinese version of Buss–Perry Aggression Questionnaire (AQ-CV) is a 30-item questionnaire with five subscales: physical aggression (7 items), verbal aggression (5 items), anger (6 items), hostility (7 items), and aggression toward self (5 items; An et al., 2010). All items are 5-point Likert-type scales (none, seldom, sometimes, often, and always) scored 1 to 5, hence the total score ranges from 30 to 150 with higher scores indicating higher aggression. The Chinese version of the scale has been formerly used in Chinese college students (An et al., 2010; Guo & Chen, 2016; Li et al., 2011), the Cronbach’s alpha of the total AQ-CV score was .89 and that of the physical aggression, verbal aggression, anger, hostility, and aggression toward self subscale scores were .77, .66, .81, .77, and .62, respectively. The intraclass correlation coefficient (ICC) of the total score was .74, and those for the physical aggression, verbal aggression, anger, hostility, and aggression toward self subscales scores were .74, .65, .74, .73, and .65, respectively (Li et al., 2011). In this study, the Cronbach’s alpha of the total AQ-CV score was .92 and that of the physical aggression, verbal aggression, anger, hostility, and aggression toward self subscales scores were .81, .79, .84, .81, and .75, respectively. The scale was found to be dependable.

Self-esteem

Rosenberg Self-Esteem Scale (RSES) was used to assess self-esteem of the participants. The Scale includes 10 items, consisting of five positive and five negative items, which participants need to complete according to how they usually felt about themselves. All responses of the items were made on scales ranging from 1 (strongly disagree) to 4 (strongly agree). Accordingly, the total score ranges from 10 to 40 with higher scores representing greater self-esteem. Previous research has demonstrated the RSES, and its Chinese translation has adequate psychometric properties (Wang et al., 2010). The Chinese version of the scale has been formerly used in Chinese college students (Guo & Chen, 2016; Han et al., 2018; Shi et al., 2017), the Cronbach’s alpha was .87, and the ICC was .76 (Chen et al., 2015). Cronbach’s alpha in this study was .80.

LBE

The grouping of LBE was based on previous studies and the socioeconomic characteristics of the local population. For example, before 6 years old, children accept preschool education, and then they would finish the primary school education before 12 years old generally. Then, they would be admitted to junior and senior school for 3 years, respectively. Most students have 2 days off study a week (Saturday and Sunday). Some students are resident students and they usually go home once a month. Most parents working outside would come home for family reunion in Chinese Spring Festival (January and February) and lived at home for a relatively long time. Most LBC have another chance of meeting and living with their parents for a period of time in summer holidays (during July and August). Besides, according to China’s 1% population survey data, more than half of the LBC lived with their grandparents, more than one third of them lived with their father or mother, and others lived with their other relatives. In addition, 5 years is an important time quantum for Chinese (e.g., National Development Plan is named as Five-Year Plan). Thus, LBE in our study includes left-behind duration (0.5–1 year, 1–5 years, more than 5 years), age of left-behind for the first time (6 years old or less, 7–12 years old, and 13–17 years old), frequency of contacting with parents (once less than 1 week, once every 1–4 weeks, and once more than 4 weeks), frequency of meeting with parents (once less than 6 months and once more than 6 months) and primary caregiver (father or mother, grandparents, and others).

Sociodemographic profile

Include gender (boys or girls), urban/rural residency, only one child (yes or no), parental marital status (married, divorced, and others), self-perceived economic status of the family (poor, moderate, or good), parents’ education level (primary School or less, junior middle school, senior middle school or more), type of school (university or college), grades (freshmen, sophomores, junior students, and senior students), disciplines (Medicine, Science, or Engineering), perceived school performance (poor, moderate, and good), smoking (yes or no), and drinking (yes or no).

Definition

In this study, the “left-behind children (LBC)” refers to juveniles, aged 0 to 17, whose parents (one or both) cannot live with them over a long period of time (more than 6 months) because of leaving the original domicile as migrant workers or others. The “university students with LBE (LBEs)” refers to the LBC who have grown up and been admitted to university.

Statistical Analysis

Sociodemographic data were described in participants. The differences of aggression (total aggression, physical aggression, verbal aggression, anger, hostility, and aggression toward self) and self-esteem between LBEs and NLBEs were examined using t-test. The differences of aggression and self-esteem in different LBE were evaluated using t-test and analysis of variance (ANOVA). Linear correlation analysis was conducted to describe the relationship between aggression and self-esteem. Multiple linear regression model was performed to investigate the associations of related characters of LBE and self-esteem with aggression. Factorial ANOVA was applied to explore the interaction effect of LBE and self-esteem on aggression. In all models, adjustment was made for gender, urban/rurality, only child, parental marital status, economic status of the family, parents’ education level, education level of participants, grades, discipline, perceived school performance, smoking, and drinking. All analyses were conducted with SPSS software, Windows version 23.0 (SPSS Inc., Chicago, IL).

Results

Table 1 shows the sociodemographic profile of the LBEs and NLBEs groups. Of the 4,154 participants, 2,296 (55.3%) were LBEs, and 1,858 (44.7%) were NLBEs. Among LBEs, 37.8% were girls, 15.2% lived in urban area, 25.5% were the only child, 93.6% with parents who were married, more lived in a family with moderate or poor economic status (81.0%), more than half of father graduated from junior high school (57.8%) and only 18.5% graduated from senior high school or received higher education, while more mother were only educated in primary school (54.6%), 65.7% were undergraduates, and approximately half (50.5%) were science and engineering students, with more freshmen (45.1%) than sophomores (31.3%), seniors (12.1%), and juniors (11.5%); 7.4% and 30.4% indicated that they had smoked or drank in the past 30 days, respectively. The detailed are listed in Table 1.

Table 1.

The Sociodemographic Profile of the LBEs and NLBEs groups, Data Shown as n (%).

Variables	Total	NLBEs	LBEs	χ²	p
Gender				1.98	.159
Boy	2,542 (61.2)	1,115 (60.0)	1,427 (62.2)
Girl	1,612 (38.8)	743 (40.0)	869 (37.8)
Home place				469.71	<.001
Urban	1,199 (28.9)	851 (45.8)	348 (15.2)
Rural	2,955 (71.1)	1,007 (54.2)	1,948 (84.8)
Only child				61.12	<.001
Yes	1,269 (30.5)	683 (36.8)	586 (25.5)
No	2,885 (69.5)	1,175 (63.2)	1,710 (74.5)
Parental marital status				0.39	.532
Married	3,895 (93.8)	1,747 (94.0)	2,148 (93.6)
Divorced or others	259 (6.2)	111 (6.0)	148 (6.4)
Economic status of family				61.75	<.001
Poor	1,206 (29.0)	456 (24.5)	750 (32.7)
Moderate	1,985 (47.8)	876 (47.2)	1,109 (48.3)
Good	963 (23.2)	526 (28.3)	437 (19.0)
Father’s education level				266.80	<.001
Primary school or less	814 (19.6)	269 (14.5)	545 (23.7)
Junior middle school	2,150 (51.8)	823 (44.3)	1,327 (57.8)
Senior middle school or more	1,190 (28.6)	766 (41.2)	424 (18.5)
Mother’s education level				393.42	<.001
Primary school or less	1,838 (44.2)	584 (31.4)	1,254 (54.6)
Junior middle school	1,565 (37.7)	715 (38.5)	850 (37.0)
Senior middle school or more	751 (18.1)	559 (30.1)	192 (8.4)
Type of school				17.20	<.001
University	2,614 (62.9)	1,105 (59.5)	1,509 (65.7)
Junior college	1,540 (37.1)	753 (40.5)	787 (34.3)
Grade
1	1,917 (46.1)	881 (47.4)	1,036 (45.1)	2.25	.522
2	1,277 (30.7)	558 (30.0)	719 (31.3)
3	463 (11.1)	200 (10.8)	263 (11.5)
4	497 (12.1)	219 (11.8)	278 (12.1)
Discipline				18.59	<.001
Medicine	2,050 (49.4)	986 (53.1)	1,064 (46.3)
Science or engineering	2,104 (50.6)	872 (46.9)	1,232 (53.7)
School performance				2.52	.284
Good	1,181 (28.4)	537 (28.9)	644 (28.1)
Moderate	2,095 (50.4)	949 (51.1)	1,146 (49.9)
Poor	878 (21.2)	372 (20.0)	506 (22.0)
Smoking				0.12	.731
No	3,843 (92.5)	1,716 (92.4)	2,127 (92.6)
Yes	311 (7.5)	142 (7.6)	169 (7.4)
Drinking				0.18	.671
No	2,878 (69.3)	1,281 (68.9)	1,597 (69.6)
Yes	1,276 (30.7)	577 (31.1)	699 (30.4)

Note. LBEs = college students with left-behind experiences; NLBEs = college students without left-behind experiences. The bold values in the table are used to highlight statistically significant values.

Table 2 presents the left-behind characteristics of the 2,296 LBEs. Most of the students’ both parents migrated (58.6%), followed by only father migrated (35.8%), and only mother migrated (5.6%). 22.2% of the parents, either one or both, had been migrants for 1 to 5 years, 23.4% had been migrants for more than 5 years.More than a third of parents migrated before their children were 6 years old. About 55.3% of the parents contacted their LBC once a week or more, 21.0% contacted them once every 1 to 4 weeks, and 23.7% contacted them once a month or less. More than half of the LBEs met with their parents once more than 6 months. In terms of the children’s primary caregivers, 34.2% were cared by their father or mother, 54.7% by grandparents, and 11.1% by relatives or others.

Table 2.

Comparison of Aggressive Behavior and Self-Esteem in Left-Behind Characteristics of LBEs, Data Shown as M (SD).

Variable	Number (%)	Total Aggression	Physical Aggression	Verbal Aggression	Anger	Hostility	Aggression Toward Self	Self-Esteem
Left-behind duration (year)
0.5–1	1,249 (54.4)	59.0 (18.5)	13.7 (5.5)	10.2 (4.1)	12.3 (5.1)	13.7 (5.3)	9.0 (3.7)	27.8 (4.5)
1–5	509 (22.2)	61.0 (16.2)	14.0 (5.0)	10.6 (3.7)	12.7 (4.6)	14.4 (4.5)	9.3 (3.6)	27.4 (4.7)
>5	538 (23.4)	62.8 (17.5)	14.3 (5.4)	11.2 (4.1)	13.1 (5.2)	14.6 (4.8)	9.6 (3.8)	27.0 (4.9)
F		8.87***	2.04	10.13***	4.76**	7.00**	5.62**	6.01**
Age of left-behind for the first time (year)
<6	930 (40.5)	61.7 (18.2)	14.2 (5.7)	10.8 (4.1)	12.8 (5.0)	14.3 (5.0)	9.6 (3.8)	27.2 (4.9)
7–12	911 (39.7)	60.1 (17.4)	13.9 (5.1)	10.4 (4.0)	12.5 (5.1)	14.1 (4.9)	9.2 (3.6)	27.8 (4.5)
13–17	455 (19.8)	58.0 (17.8)	13.5 (5.3)	10.1 (3.9)	12.1 (4.8)	13.5 (5.1)	8.7 (3.7)	27.9 (4.5)
F		6.92**	2.62	5.20**	3.09*	3.76*	9.21***	5.35**
Frequency of contacting with parents
Once less than 1 week	1,270 (55.3)	59.5 (17.5)	13.7 (5.2)	10.4 (4.0)	12.5 (5.0)	13.9 (5.0)	9.1 (3.6)	27.8 (4.6)
Once every 1–4 weeks	483 (21.0)	60.3 (17.5）	14.2 (5.3)	10.5 (3.9)	12.5 (4.8)	13.9 (4.9)	9.2 (3.6)	27.4 (4.6)
Once more than 4 weeks	543 (23.7)	62.3 (18.9)	14.4 (5.8)	11.0 (4.2)	12.8 (5.3)	14.6 (5.2)	9.6 (4.0)	27.0 (4.8)
F		4.99**	3.62*	4.74**	0.87	4.50*	3.15*	6.36**
Frequency of meeting with parents
Once less than 6 months	1,117 (48.6)	59.4 (17.9)	13.7 (5.1)	10.4 (3.9)	12.4 (4.9)	13.8 (5.1)	9.2 (3.7)	27.8 (4.5)
Once more than 6 months	1,179 (51.4)	61.2 (17.7)	14.2 (5.6)	10.7 (4.1)	12.8 (5.1)	14.3 (4.9)	9.3 (3.7)	27.3 (4.8)
T		−2.34*	−2.01*	−2.05*	−1.87	−2.07*	−0.79	2.36*
Primary caregiver
Father/mother	785 (34.2)	58.7 (17.2)	13.5 (5.1)	10.3 (4.0)	12.2 (4.9)	13.7 (4.8)	9.0 (3.4)	28.1 (4.7)
Grandparents	1,255 (54.7)	60.9 (17.9)	14.1 (5.4)	10.6 (4.1)	12.7 (5.1)	14.2 (5.0)	9.3 (3.8)	27.4 (4.5)
Others	256 (11.1)	62.5 (19.1)	14.5 (5.9)	11.1 (4.0)	13.0 (5.2)	14.4 (5.4)	9.5 (3.8)	26.5 (5.2)
F		5.94**	4.00*	3.63*	3.63*	3.76*	2.86	11.74***

Note. LBEs = college students with left-behind experiences.

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

The scores of total aggression (60.3 ± 17.8), physical aggression (13.9 ± 5.4), verbal aggression (10.5 ± 4.0), anger (12.6 ± 5.0), hostility (14.1 ± 5.0), and aggression toward self (9.2 ± 3.7) of LBEs were significantly higher than that (in proper order, 57.7 ± 17.1, 13.6 ± 4.9, 10.0 ± 3.3, 12.1 ± 4.7 13.4 ± 4.8, and 8.5 ± 3.3) of NLBEs (p < .05; Figure 1). Meanwhile, the scores of self-esteem of LBEs (27.5 ± 4.7) was significantly lower than that (28.7 ± 4.0) of NLBEs (p < .001; Figure 1).

Figure 1.

Comparison of aggressive behavior and self-esteem between NLBEs and LBEs.

Linear correlation analysis showed that self-esteem was significantly associated with total aggression (r = −.242, p < .001), verbal aggression (r = −.118, p < .001), anger (r = −.234, p < .001), hostility (r = −.307, p < .001), and aggression toward self (r = −.287, p < .001), respectively, while we did not find the association between self-esteem and physical aggression (r = −.027, p > .05).

A series of analyses of variance (ANOVAs) were run to examine various left-behind characteristics on total aggression, physical aggression, verbal aggression, anger, hostility, aggression toward self, and self-esteem (Table 2). The differences of total aggression, verbal aggression, hostility, and self-esteem were found significant in all left-behind–related characters we included in this study, while the difference of physical aggression, anger, and aggression toward self of the LBEs were only significant in some of the left-behind–related factors. The LBEs with longer duration of left-behind, younger age of first time left-behind, lower contact frequency with parents and cared by grandparents or others scored higher in total aggression and lower in self-esteem. Compared with LBEs who met with parents once more than 6 months, the LBEs who met with parents once less than 6 months had lower level of total aggression, physical aggression, verbal aggression, and hostility, while having higher level of self-esteem (p < .05, Table 2). Multiple comparisons results showed that LBEs who were left-behind for more than 5 years scored higher in total aggression, verbal aggression, anger, hostility, and aggression toward self than the LBEs who were left-behind less than 1 year (p < .05; appendix, Table A1). The LBEs who were left-behind under 6 years old for the first time were found more on total aggression, verbal aggression, anger, hostility, and aggression toward self than the LBEs who were first time left behind over 13 years old (p < .05; appendix, Table A2). Meanwhile, LBEs who were left-behind for more than 5 years or left-behind under 6 years old for the first time had lower self-esteem. Compared with LBEs who contacted with parents weekly, the LBEs who contacted with parents once more than 4 weeks had higher level of total aggression, physical aggression, verbal aggression, hostility, and aggression toward self, while having lower level of self-esteem (p < .05; appendix, Table A3). Regarding primary caregiver, the LBEs who were cared by one of parent scored lower in total aggression and hostility than those who were cared by grandparents. They also scored lower in total aggression, physical aggression, and verbal aggression than those cared by other relatives (p < .05; appendix, Table A4). The LBEs who were cared by one of parent had higher self-esteem, followed by those who were cared by grandparents and other relatives, the differences were statistically significant (p < .05; appendix, Table A4).

To identify the risk and protective factors of aggressive behavior in LBEs, we further conducted multiple regression analyses (Table 3). After controlling for gender, urban/rural, only child, parental marital status, economic status of family, parents’ education level, type of school, grade, discipline, school performance, smoking, and drinking, the results showed that LBEs with more than 5 years of left-behind duration showed significantly higher total aggression (β = 2.626), verbal aggression (β = .763), anger (β = .564), and aggression toward self (β = .398) than the LBEs with 0.5 to 1 year left-behind duration. Compared with the LBEs who were left-behind after 12 years old, those who were left-behind before 6 years old reported significantly higher total aggression (β = 2.523), verbal aggression (β = .552), and aggression toward self (β = .606). The results of the relationship between self-esteem and aggression were similar with that of linear correlation analysis. Neither left-behind–related characters nor self-esteem was significantly associated with physical aggression (Table 3).

Table 3.

Results of Multiple Linear Regression Analyses Predicting Aggressive Behavior of LBEs Groups.

Variable	Total Aggression	Physical Aggression	Verbal Aggression	Anger	Hostility	Aggression Toward Self
Variable	β	β	β	β	β	β
Left-behind duration (year)
0.5–1 (ref)
1–5	1.252	0.027	0.276	0.373	0.412	0.164
>5	2.626**	0.413	0.763***	0.564 *	0.488	0.398 *
Age of left-behind for the first time (year)
12–17 (ref)
6–12	1.789	0.298	0.223	0.353	0.463	0.451 *
<6	2.523 *	0.557	0.522 *	0.505	0.333	0.606**
Frequency of contacting with parents
Once less than 1 week (ref)
Once every 1–4 weeks	−0.634	0.016	−0.134	−0.110	−0.255	−0.150
Once more than 4 weeks	1.021	0.261	0.313	0.032	0.323	0.092
Frequency of meeting with parents
Once less than 6 month (ref)
Once more than 6 month	0.142	0.082	0.013	0.039	0.033	−0.025
Primary caregiver
Father/Mother (ref)
Grandparents	0.992	0.450	0.032	0.186	0.198	0.125
Others	2.281	0.737	0.495	0.476	0.277	0.296
Self-esteem	−0.804***	−0.011	−0.087***	−0.206***	−0.298***	−0.202***
F	11.405***	10.351***	6.072***	8.115***	12.195***	12.688***
R ²	0.104	0.095	0.058	0.076	0.110	0.114
Adjusted R²	0.094	0.086	0.048	0.067	0.101	0.105

Note. All models controlled for gender, urban/rural, only child, parental marital status, economic status of family, parents’ education level, type of school, grade, discipline, school performance, smoking, and drinking. LBEs = college students with left-behind experiences. β = regression coefficient.The bold values in the table are used to highlight statistically significant values.

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

To verify if there existed interaction effects of LBE and self-esteem on aggressive behavior, we conducted factorial ANOVA to explore the interaction effects of left-behind–related characters and self-esteem on aggressive behavior of LBEs (Table 4). The result showed that there was an interaction effect of left-behind duration and self-esteem on physical aggression (p < .05) and of frequency of contacting with parents and self-esteem on verbal aggression (p < .05) and of primary caregiver and self-esteem on hostility (p < .01) and of primary caregiver and aggression toward self (p < .05). There were no other significant interaction effects of left-behind–related characters and self-esteem on aggression (Table 4).

Table 4.

The Interaction Effect of Left-Behind–Related Factors and Self-Esteem on Aggressive Behavior.

Variable	Total Aggression	Physical Aggression	Verbal Aggression	Anger	Hostility	Aggression Toward Self
Left-behind duration × SE	2.42	3.24*	0.15	1.79	1.45	1.22
Age of left-behind for the first time × SE	0.40	0.14	0.16	0.89	0.34	1.17
Frequency of contacting with parents × SE	1.18	1.10	3.54*	0.03	1.34	2.24
Frequency of meeting with parents × SE	0.36	0.09	1.52	0.44	0.43	1.29
Primary caregiver × SE	2.50	0.70	0.03	0.94	5.31**	3.46*

Note. The statistics were F value from the results of factorial ANOVA. SE = self-esteem; ANOVA = analysis of variance.

*p < .05. **p < .01.

In Figure 2, the simple effect analysis showed that although all the scores of physical aggression decreased with the increase of self-esteem, those LBEs who have longer left-behind duration changed much more obviously compared with those having shorter left-behind duration. In terms of verbal aggression, those LBEs who contacted with their parents with lower frequency showed more obvious change with the alteration of self-esteem. As long as hostility and aggression toward self were concerned, with the increase of self-esteem, those LBEs cared by grandparents or others showed more prominent changes compared with those cared by parents (Figure 2).

Figure 2.

Associations between aggressive behavior and self-esteem at different levels of left-behind–related characters.

Discussion

A school-based survey was conducted to examine the effects of LBEs on aggressive behavior (physical aggression, verbal aggression, anger, hostility, and aggression toward self) and self-esteem and the interaction effects of LBE and self-esteem on aggression. Our data revealed that the students with LBE were more likely to have higher level of aggression and lower level of self-esteem. The increased aggressive behavior was highly related to longer left-behind duration, younger age of left-behind for the first time and decreased self-esteem. Moreover, aggressive behavior was demonstrated negatively associated with self-esteem. Interaction effects of some left-behind–related characters and self-esteem on aggressive behavior were also found.

Our study is consistent with previous findings in children which have also indicated that individuals with LBE have a significantly higher level of aggression and lower self-esteem (Tang et al., 2018; Zhang et al., 2018). We also found that in young adults, the higher level of total aggression and its subscales were correlated with lower level of self-esteem except for physical aggression, which was consistent with previous studies focused on the correlations of antisocial behavior and self-esteem (Donnellan et al., 2005; Trzesniewski et al., 2006). The possible explanation may be that low level of self-esteem weakens communication with society, and accordingly, promote social maladaptation and delinquency (Rosenberg et al., 1978). A cohort study in New Zealand also confirmed that low self-esteem blossoms into not only delinquency but also multiple problematical behavior and health issues during and after puberty (Trzesniewski et al., 2006). Different from previous research (Garofalo et al., 2016), our result showed that physical aggression was not associated with lower level of self-esteem. One reason may be associated with different sample in different research. Moreover, studies had indicated that physical aggression decreased gradually for most children by school age except for a small group of children who remain physical aggression throughout adolescence and adulthood (Côté et al., 2006, (2007; Underwood et al., 2009). The participants of our study were undergraduates, most of them were more than 18 years old and well educated. They are more inclined to express angry and hostile by indirect ways instead of direct physical aggression (Luo et al., 2011). Thus, the overall level of the scores of physical aggression was relatively low, which may not show an obvious trend with the change of self-esteem. Further researches are needed to verify the association between physical aggression and self-esteem.

Our further multiple linear analysis found that left-behind–related characters and self-esteem had dose-dependent effects on aggression, even after adjusting for sociodemographic factors. The longer left-behind duration, younger age of left-behind for the first time and lower self-esteem they had, the higher aggression they possessed, which is also similar with the results of previous studies (Zhang et al., 2018). Fan et al.’s (2010) research showed that LBC are more likely to develop emotional/behavior problems, particularly if they were left behind earlier in life, for longer time. Some researches confirmed that children who were emotionally deprived could develop insecure attachments to their caregivers, which correspondingly leads to children’s negative emotions and their stress cannot be reasonably vented as well (Dai et al., 2017; O’Connor & Zeanah, 2003; Zhao et al., 2011). Ultimately, it may bring an increase of aggressive behavior.

The relationship among left-behind–related characters, self-esteem, and aggression is complicated and the results are not unified. Arslan (2016) suggested that the association between psychological maltreatment and emotional and behavioral problems was mediated by self-esteem. A study of junior high school students found that self-esteem is a partial mediator between family function and aggressive behavior (Zheng, 2018). Hence, it would be reasonable to suggest that self-esteem may also moderate the relationship between left-behind–related factors and aggressive behavior. In our research, we found that the scores of physical aggression decreased with the increase of self-esteem in LBEs. Those LBEs who have longer left-behind duration change much more obviously compared to those with shorter left-behind duration. In terms of verbal aggression, those LBEs who contacted with their parents with lower frequency showed more obvious changes with the alteration of self-esteem. As long as hostility and aggression toward self were concerned, with the increase of self-esteem, those LBEs cared by grandparents or others showed more obvious change compared with those cared by parents. These interaction results indicated that for those left-behind young adults who have long left-behind duration, less contact frequency with their parents and cared by grandparents or others, improving their self-esteem level may be an effective way to decrease aggressive behavior. However, only some interact effects were found, further studies are needed to explore the interact effects between left-behind–related characters and self-esteem on aggression.

Strengths and Limitations

Our study extends existing knowledge by examining independent and interactive effects between LBE and self-esteem on aggressive behavior within Chinese university students. The study sample is representative, covering junior college and university in Anhui province, China. The response rate of the participants was high. The large sample from college and university provided enough statistical power to examine the relationship among LBE, self-esteem, and aggression, with multivariate adjustment analysis. However, the study has limitations. First, the study was cross-sectional, thus it is hard to establish a causal relationship. Hence, future researches should adopt cohort designs. Second, owing to the use of self-reported questionnaires for data collection, it is impossible to avoid recall bias or detection bias. This may finally influence the strength of the observed relationships, although we conducted a rigorous screening of the questionnaires. Third, only some left-behind–related characters were focused in this study, it may be significant to include an individual’s subjective experience as the factors. Finally, the study focused on LBEs in traditional school environments; thus, the findings did not represent adolescents who dropped out, which is important because the study showed that the LBC had higher dropout rates (Valtolina & Colombo, 2012). Those young adults who dropped out may have poorer left-behind conditions and more unfavorable living conditions. Hence, caution should be exercised in the application of the findings to the whole population of students with LBE in China. Besides, all participants were students of Medicine, Science, or Engineering majors, caution should be exercised in the application of the findings to the other students of disciplines.

Conclusion and Implications

The findings indicate that LBE is independently associated with an increased risk of aggressive behavior and the decrease of self-esteem, and the low self-esteem is correlated with high aggressive behavior. Furthermore, there were interact effects between some of left-behind–related characters and self-esteem on aggression, which indicated that improving self-esteem especially in those LBEs with long duration of left-behind, low contact frequency with parents and cared by grandparents or others may be an intervention and prevention strategy of reducing aggressive behaviors. Moreover, more targeted policies are encouraged to help the LBC adapt to changeable environment and to promote the mental health of them, ultimately reducing the occurrence of behavior problems. For example, the education department should provide schools with the necessary support and assistance to encourage them to establish psychological counseling centers, provide basic mental health courses, and conduct group psychological guidance on a regular basis. The labor department needs to appropriately increase paid holidays for LBC’s parents so that they have more time to reunite with their children, increase outdoor sport, and strengthen their emotional communication. The result also reminds us that the effects of LBE could last until young adulthood or whole life cycle, which indicated young adults or adults with LBE deserve more attention. The government should proceed with designing the policy framework for young adults with LBE.

Educational environment is likely to represent an important conduit through which to improve mental health to LBC. On a targeted way, a training program specifically focusing on the mental health education of teachers and other school staff need be put into force. This should center around both improving their knowledge of the signs and symptoms of mental problems, particularly among individuals with LBE, and on implementing basic education in mental health (Zalsman et al., 2016). Interpersonal skills training for students may also help to improve social and peer interactions, which could enhance self-esteem, ultimately, reduce the occurrence of aggressive behavior. Moreover, as part of a systems project, strengthening ties among schools, families, and communities, encouraging them to positively engage in extracurricular activities, could improve the social support of students, and eventually conducive to the development of mental health (e.g., self-esteem) for them (Ougrin et al., 2012; Wan et al., 2019), and decrease the incidence of aggressive behavior. At last, a nationwide sample study or even a cohort study is needed, which contributes to discover long-term trends of aggressive behavior and self-esteem and explore the underlying mechanisms of the relationship between LBE and aggressive behavior and, ultimately, provides stronger sufficient theoretical evidences for the formulation of relevant public health policies and the implementation of specific interventions.

Appendix

Table A1.

Multiple Comparisons of Aggressive Behavior and Self-Esteem in Left-Behind Duration of LBEs.

Variable	Left-Behind Duration (Year)
Variable	Difference in 0.5–1 and 1–5	Difference in 0.5–1 and >5	Difference in 1–5 and >5
Total aggression	.102	<.001	.317
Physical aggression	.999	.143	.999
Verbal aggression	.221	<.001	.081
Anger	.425	.008	.609
Hostility	.036	.002	.999
Aggression toward self	.420	.003	.396
Self-esteem	.200	.002	.652

Note. Data shown as p value of Bonferroni. LBEs = left-behind experiences.The bold values in the table are used to highlight statistically significant values.

Table A2.

Multiple Comparisons of Aggressive Behavior and Self-Esteem in Age of First Time Left-Behind of LBEs.

Variable	Age of First Time Left-Behind (Year)
Variable	Difference in <6 and 7–12	Difference in <6 and 13–17	Difference in 6–12 and 13–17
Total aggression	.148	.001	.115
Physical aggression	.448	.082	.913
Verbal aggression	.104	.006	.528
Anger	.418	.046	.673
Hostility	.999	.022	.082
Aggression toward self	.068	<.001	.052
Self-esteem	.021	.017	.999

Note. Data shown as p value of Bonferroni. LBEs = left-behind experiences.The bold values in the table are used to highlight statistically significant values.

Table A3.

Multiple Comparisons of Aggressive Behavior and Self-Esteem in Frequency of Contacting With Parents of LBEs.

Variable	Frequency of Contacting With Parents
Variable	Difference in Once Les Than 1 Week and Once Every 1–4 Weeks	Difference in Once Less Than 1 Week and Once More Than 4 Weeks	Difference in Once Every 1–4 Weeks and Once More Than 4 Weeks
Total aggression	.999	.005	.201
Physical aggression	.282	.037	.999
Verbal aggression	.999	.007	.143
Anger	.999	.577	.999
Hostility	.999	.010	.087
Aggression toward self	.999	.037	.335
Self-esteem	.223	.002	.562

Note. Data shown as p value of Bonferroni. LBEs = left-behind experiences.The bold values in the table are used to highlight statistically significant values.

Table A4.

Multiple Comparisons of Aggressive Behavior and Self-Esteem in Primary Caregiver of LBEs.

Variable	Primary Caregiver
Variable	Difference in Father/Mother and Grandparents	Difference in Father/Mother and Others	Difference in Grandparents and Others
Total aggression	.018	.009	.582
Physical aggression	.065	.048	.948
Verbal aggression	.362	.026	.250
Anger	.093	.060	.934
Hostility	.044	.110	.999
Aggression toward self	.137	.140	.999
Self-esteem	.007	<.001	.013

Note. Data shown as p value of Bonferroni. LBEs = left-behind experiences.The bold values in the table are used to highlight statistically significant values.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by grants from Major Program of Provincial University Natural Science Research Project of Anhui Province (grant nos. KJ2014ZD18 and KJ2019ZD23), the Education Quality Engineering of Anhui Province (grant no. 2016jyxm0510), and the National Natural Science Foundation of China (grant nos. 30970906 and 81974306).

Author Biographies

Bang-Lin Yu is an undergraduate student at School of Public Health at Anhui Medical University, China. He is interested in the psychology and behavior science in adolescents.

Juan Li is a postgraduate student at School of Public Health at Anhui Medical University, China. She is interested in adverse childhood experiences and development in children.

Wei Liu is a postgraduate student at School of Public Health at Anhui Medical University, China. He is interested in the prevention and control of injury and violence in adolescents.

Sheng-Hai Huang, PhD, is a professor in School of Basic Medical Sciences at Anhui Medical University, China. His research focuses on the pathogenesis and interventions of severe lower respiratory tract infectious disease in infants, with particular interest in the pathogenesis of bronchiolitis and pneumonia caused by respiratory syncytial virus, and the interaction between the virus and the host.

Xiu-Jing Cao, PhD, is a professor in the School of Public Health at Anhui Medical University, China. Her research focuses on the effects of adverse childhood experience such as left behind of children on cognition and behavior development in adolescents and adults. She is interested in answering these questions by population and experiment studies.

ORCID iD

Xiu-Jing Cao

References

Phillips

Conner

(2010). Validity of proxy-based reports of impulsivity and aggression in Chinese research on suicidal behavior. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 31, 137–142. https://doi.org/10.1027/0227-5910/a000019

Anderson

Bushman

(2002). Human aggression. Annual Review of Psychology, 53, 27–51. https://doi.org/10.1146/annurev.psych.53.100901.135231

Arslan

(2016). Psychological maltreatment, emotional and behavioral problems in adolescents: The mediating role of resilience and self-esteem. Child Abuse & Neglect, 52, 200–209. https://doi.org/10.1016/j.chiabu.2015.09.010

Aryal

Regmi

P. R.

Van Teijlingen

Simkhada

Mahato

(2019). Adolescents left behind by migrant workers: A call for community-based mental health interventions in Nepal. WHO South-East Asia Journal of Public Health, 8, 38–41. https://doi.org/10.4103/2224-3151.255348

Baumeister

Smart

Boden

(1996). Relation of threatened egotism to violence and aggression: The dark side of high self-esteem. Psychological Review, 103, 5–33.

Bender

Lösel

(2011). Bullying at school as a predictor of delinquency, violence and other anti-social behaviour in adulthood. Criminal Behaviour and Mental Health, 21, 99–106. https://doi.org/10.1002/cbm.799

Chen

F. F.

C. Z.

Han

M. F.

(2015). The reliability and validity of the Chinese version of the revised-positive version of Rosenberg Self-Esteem Scale. Advances in Psychology, 5, 531–535.

Comer

Gould

(2011). Psychology around us. John Wiley.

Côté

S. M.

Vaillancourt

Barker

E. D.

Nagin

Tremblay

R. E.

(2007). The joint development of physical and indirect aggression: Predictors of continuity and change during childhood. Development and Psychopathology, 19, 37–55. https://doi.org/10.1017/S0954579407070034

10.

Côté

S. M.

Vaillancourt

LeBlanc

J. C.

Nagin

D. S.

Tremblay

R. E.

(2006). The development of physical aggression from toddlerhood to pre-adolescence: A nation wide longitudinal study of Canadian children. Journal of Abnormal Child Psychology, 34, 68–82. https://doi.org/10.1007/s10802-005-9001-z

11.

Dai

Yang

Wang

Liu

Guang

, … Feng

(2017). The alienation of affection toward parents and influential factors in Chinese left-behind children. European Psychiatry, 39, 114–122. https://doi.org/10.1016/j.eurpsy.2016.07.008

12.

Delva-Tauili’ili

(1995). Assessment and prevention of aggressive behavior among youths of color: Integrating cultural and social factors. Children & Schools, 17, 83–91. https://doi.org/10.1093/cs/17.2.83

13.

Donnellan

M. B.

Trzesniewski

K. H.

Robins

R. W.

Moffitt

T. E.

Caspi

(2005). Low self-esteem is related to aggression, antisocial behavior, and delinquency. Psychological Science, 16, 328–335. https://doi.org/10.1111/j.0956-7976.2005.01535.x

14.

Duan

C. R.

Yang

(2008). The left-behind children in rural China. Population Research, 32, 15–25.

15.

Fan

Gill

M. K.

Birmaher

(2010). Emotional and behavioral problems of Chinese left-behind children: A preliminary study. Social Psychiatry and Psychiatric Epidemiology, 45, 655–664. https://doi.org/10.1007/s00127-009-0107-4

16.

Fellmeth

Rose-Clarke

Zhao

Busert

L. K.

Zheng

Massazza

, … Devakumar

(2018). Health impacts of parental migration on left-behind children and adolescents: A systematic review and meta-analysis. The Lancet, 392, 2567–2582. https://doi.org/10.1016/S0140-6736(18)32558-3

17.

Ferguson

C. J.

Beaver

K. M.

(2009). Natural born killers: The genetic origins of extreme violence. Aggression and Violent Behavior, 14, 286–294. https://doi.org/10.1016/j.avb.2009.03.005

18.

Garofalo

Holden

C. J.

Zeigler-Hill

Velotti

(2016). Understanding the connection between self-esteem and aggression: The mediating role of emotion dysregulation. Aggressive Behavior, 42, 3–15. https://doi.org/10.1002/ab.21601

19.

Guo

Chen

(2016). Relationship between childhood physical abuse and aggressivity among undergraduate students: Mediating role of self-esteem. Chinese Journal of School Health, 37, 529–531. https://doi.org/10.16835/j.cnki.1000-9817.2016.04.016

20.

Han

Zhao

Pan

Liao

(2018). The impact of students with left-behind experiences on childhood: The relationship between negative life events and depression among college students in China. International Journal of Social Psychiatry, 64, 56–62. https://doi.org/10.1177/0020764017739332

21.

Huz

Nyer

Dickson

Farabaugh

Alpert

Fava

Baer

(2016). Obsessive-compulsive symptoms as a risk factor for suicidality in U.S. college students. Journal of Adolescent Health, 58, 481–484. https://doi.org/10.1016/j.jadohealth.2015.11.011

22.

Lansford

J. E.

Skinner

A. T.

Sorbring

Giunta

L. D.

Deater-Deckard

Dodge

K. A.

, … Chang

(2012). Boys’ and girls’ relational and physical aggression in nine countries. Aggressive Behavior, 38, 298–308. https://doi.org/10.1002/ab.21433

23.

Yang

Xiao

(2018). Young people with left-behind experiences in childhood have higher levels of psychological resilience. Journal of Health Psychology. Advance online publication. https://doi.org/10.1177/1359105318801056

24.

X. Y.

Phillips

M. R.

Zhang

Y. L.

Niu

Y. J.

Tong

Y. S.

Yang

S. J.

(2011). Development, reliability and validity of the Chinese version of Buss & Perry Aggression Questionnaire. Chinese Journal of Nervous and Mental Diseases, 37, 607–613. https://doi.org/10.3969/j.issn.1002-0152.2011.10.010

25.

Luo

Y. X.

Y. Z.

Yang

Sun

(2011). Correlation study of aggressive behavior and explicit self-esteem among middle school students. Chinese Journal of School Health, 32, 914–915, 918. https://doi.org/34-1092/R.20110816.1046.007

26.

L. D.

Yan

Duan

C. R.

Cheng

M. Y.

(2018). Changing patterns and development challenges of child population in China. Chinese Journal of Population Research, 42, 65–78.

27.

Matthews

Jennings

Lee

Pardini

(2017). Bullying and being bullied in childhood are associated with different psychosocial risk factors for poor physical health in men. Psychological Science, 28, 808–821. https://doi.org/10.1177/0956797617697700

28.

National Bureau of Statistics of China, United Nations Children’s Fund China, & United Nations Population Fund China. (2017). Population status of children in China in 2015: Facts and figures. https://www.unicef.cn/media/9901/file/Population%20Status%20of%20Children%20in%20China%20in%202015%20Facts%20and%20Figures.pdf

29.

Naughton

A. M.

Maguire

S. A.

Mann

M. K.

Lumb

R. C.

Tempest

Gracias

Kemp

A. M.

(2013). Emotional, behavioral, and developmental features indicative of neglect or emotional abuse in preschool children: A systematic review. JAMA Pediatrics, 167, 769–775. https://doi.org/10.1001/jamapediatrics.2013.192

30.

O’Connor

T. G.

Zeanah

C. H.

(2003). Attachment disorders: Assessment strategies and treatment approaches. Attachment & Human Development, 5, 223–244. https://doi.org/10.1080/14616730310001593974

31.

Ougrin

Tranah

Leigh

Taylor

Asarnow

J. R.

(2012). Practitioner review: Self-harm in adolescents. Journal of Child Psychology and Psychiatry, 53, 337–350. https://doi.org/10.1111/j.1469-7610.2012.02525.x

32.

G. B.

Wang

Tang

Sun

Y. H.

Wang

(2018). Systematic review and meta-analysis found higher levels of behavioural problems in male left-behind children aged 6–11 years. Acta Paediatrica, 107, 1327–1334. https://doi.org/10.1111/apa.14199

33.

Rosenberg

F. R.

Rosenberg

McCord

(1978). Self-esteem and delinquency. Journal of Youth and Adolescence, 7, 279–294. https://doi.org/10.1007/BF01537978

34.

Shi

Chen

Yin

Zhao

Yao

(2016). Resilience as moderator of the relationship between left-behind experience and mental health of Chinese adolescents. International Journal of Social Psychiatry, 62, 386–393. https://doi.org/10.1177/0020764016636910

35.

Shi

Wang

Yao

Zhao

Chen

(2017). Family impacts on self-esteem in Chinese college freshmen. Frontiers in Psychiatry, 8, 279–285. https://doi.org/10.3389/fpsyt.2017.00279

36.

Tang

Wang

Dai

Yang

(2018). Mental health and psychosocial problems among Chinese left-behind children: A cross-sectional comparative study. Journal of Affective Disorders, 241, 133–141. https://doi.org/10.1016/j.jad.2018.08.017

37.

Terburg

Morgan

van Honk

(2009). The testosterone–cortisol ratio: A hormonal marker for proneness to social aggression. International Journal of Law and Psychiatry, 32, 216–223. https://doi.org/10.1016/j.ijlp.2009.04.008

38.

Trzesniewski

K. H.

Donnellan

Moffitt

Robins

Poulton

Caspi

(2006). Low self-esteem during adolescence predicts poor health, criminal behavior, and limited economic prospects during adulthood. Developmental Psychology, 42, 381–390. https://doi.org/10.1037/0012-1649.42.2.381

39.

Underwood

M. K.

Beron

K. J.

Rosen

L. H.

(2009). Continuity and change in social and physical aggression from middle childhood through early adolescence. Aggressive Behavior, 35, 357–375. https://doi.org/10.1002/ab.20313

40.

Valtolina

Colombo

(2012). Psychological well-being, family relations, and developmental issues of children left behind. Psychological Reports, 111, 905–928. https://doi.org/10.2466/21.10.17.PR0.111.6.905-928

41.

Waller

Hyde

Klump

Burt

(2018). Parenting is an environmental predictor of callous unemotional traits and aggression: A monozygotic twin differences study. Journal of the American Academy of Child and Adolescent Psychiatry, 57, 955–963. https://doi.org/10.1016/j.jaac.2018.07.882

42.

Wan

Chen

McFeeters

Sun

Hao

Tao

(2019). Associations of adverse childhood experiences and social support with self-injurious behaviour and suicidality in adolescents. The British Journal of Psychiatry, 214, 146–152. https://doi.org/10.1192/bjp.2018.263

43.

Wang

M. C.

Cai

B. G.

Dai

X. Y.

(2010). The factor structure of Chinese Rosenberg’ Self-Esteem Scale affected by item statement method. Chinese Journal of Psychological Exploration, 30, 63–68.

44.

Zalsman

Hawton

Wasserman

van Heeringen

Arensman

Sarchiapone

, … Zohar

(2016). Suicide prevention strategies revisited: 10-year systematic review. The Lancet Psychiatry, 3, 646–659. https://doi.org/10.1016/S2215-0366(16)30030-X

45.

Zhang

Meng

Yao

(2018). Association of aggressive behavior with separation from parents and social anxiety in grade four to six of rural senior primary school students in Anhui Province in 2014. Chinese Journal of Wei Sheng Yan Jiu, 47, 206–217.

46.

Zhao

Tang

K. S.

(2011). A study of the relationship of aggressive, social support, family adaptability and cohesion of the left-behind delinquents. Chinese Journal of Clinical Psychology, 19, 790–791.

47.

Zheng

W. Q.

(2018). The influence of family function on junior high school students’ aggressive behavior: The mediating effect of self-esteem [Master’s thesis]. Tianjin Normal University.