The effect of life events, resilience, self-esteem, and coping styles on aggressive behavior among left-behind adolescents: Structural equation modeling

Jie Zhang, Yifei Li, Juan Li, Mengmeng Lyu, Yiping Chen, Silan Yang, Chuxia Tan, Yingxiang Tao, Biyun Ye, Jingping Zhang, Fangqun Cheng, Ting MaoFrontiers in Psychiatry 2023 · DOI

TL;DR

This study uses structural equation modeling (SEM) to reveal how negative life events drive aggressive behavior in China’s "left-behind" adolescents. The findings highlight that while life events trigger aggression, psychological "buffers"—specifically resilience, self-esteem, and positive coping styles—can significantly mitigate these effects, providing a roadmap for targeted mental health interventions.

What Problem Does This Paper Address?

In the wake of rapid urbanization, China has seen a massive internal migration of workers from rural to urban areas. A side effect of this economic shift is the "left-behind" phenomenon: millions of children remain in rural villages while one or both parents move away for work. These adolescents often lack consistent parental support, making them more susceptible to psychological distress and aggressive behavior.

While previous research established that left-behind adolescents face higher risks, the mechanisms—the "how" and "why"—remained fuzzy. This paper seeks to move beyond simple correlations to understand the complex interplay between external stressors (life events) and internal psychological traits (resilience, self-esteem, and coping styles) in predicting aggression.

Methods

The researchers conducted a cross-sectional survey of 751 left-behind adolescents (aged 10–16) in Hunan Province, China. To move beyond basic regression, they employed Structural Equation Modeling (SEM). SEM is a powerful statistical technique that allows researchers to test a "web" of relationships simultaneously, identifying both direct effects and indirect "mediating" pathways.

The study utilized five validated psychometric instruments:

  • ASLEC: Adolescent Self-Rating Life Events Checklist (stressors).
  • Resilience Scale for Chinese Adolescents: Measuring adaptability.
  • RSES: Rosenberg Self-Esteem Scale.
  • CSQ: Coping Style Questionnaire (categorized into positive and negative coping).
  • BWAQ: Buss–Warren Aggression Questionnaire (the primary outcome).

Datasets

The study relied on a primary dataset collected through a multi-stage stratified random sampling of four rural junior high schools in Hunan.

  • Sample Size: 751 valid responses (95.18% response rate).
  • Demographics: Roughly equal split between male (50.5%) and female (49.5%) participants.
  • Context: All participants met the criteria of having one or both parents away for work for at least six months.

Key Insights and Ideas

  • The "Buffer" Effect: The most critical insight is that negative life events don't lead to aggression in a vacuum. Resilience and self-esteem act as significant mediators. High resilience doesn't just make an adolescent "tougher"; it fundamentally changes how they perceive and react to stress.
  • Resilience is the Heavy Hitter: Of all the factors studied, resilience had the strongest total effect (-0.626) on reducing aggression. It acts as a primary defense mechanism, suggesting that interventions focusing on "grit" and adaptability might be more effective than those focusing solely on self-esteem.
  • The Gender Nuance: The study found that girls reported lower levels of negative coping than boys. This suggests that adolescent girls in this demographic may be more socially inclined to seek help or process stress through communication, whereas boys may be more prone to internalizing or acting out.
  • Economic Paradox: Interestingly, household income had only a very weak indirect effect on aggression. This suggests that for left-behind children, psychological support and internal traits are far more influential than the family's financial status.

Key Results

The SEM model showed a "good fit" (GFI = 0.910, CFI = 0.922, RMSEA = 0.069), indicating the theoretical model aligned well with the real-world data.

  • Direct Impact: Negative life events had a direct positive effect on aggression (0.147).
  • Indirect Impact: The indirect effect of life events (mediated through resilience, self-esteem, and coping) was much higher (0.361), bringing the total effect to 0.508.
  • Protective Factors: Resilience (-0.626) and Self-esteem (-0.253) were both negatively correlated with aggression.
  • Coping Styles: Negative coping styles (like avoidance or rumination) significantly increased aggression (total effect of 0.313), while positive coping had a negligible direct effect but helped build resilience.

Limitations and Future Work

The authors acknowledge several constraints:

  1. Cross-Sectional Design: Because the data was collected at a single point in time, the study can show relationships but cannot definitively prove causality. A longitudinal study (tracking the same kids over years) would be the logical next step.
  2. Self-Reporting Bias: Adolescents might under-report aggressive behavior or over-report resilience due to social desirability bias.
  3. External Factors: The model did not account for external social support (teachers, grandparents, or peers), which likely plays a massive role in rural Chinese communities.

Future research could explore how digital communication with migrant parents (e.g., video calls) influences these psychological mediators.

Summary updated

Abstract

Left-behind adolescents can reduce their aggressive behavior by assuaging the adverse effects of life events via increased resilience and self-esteem as well as the adoption of positive coping strategies.
BibTeX
@inproceedings{zhang2023effect,
  author = {Jie Zhang and Yifei Li and Juan Li and Mengmeng Lyu and Yiping Chen and Silan Yang and Chuxia Tan and Yingxiang Tao and Biyun Ye and Jingping Zhang and Fangqun Cheng and Ting Mao},
  title = {{The effect of life events, resilience, self-esteem, and coping styles on aggressive behavior among left-behind adolescents: Structural equation modeling}},
  booktitle = {Frontiers in Psychiatry},
  year = {2023},
  doi = {10.3389/fpsyt.2023.991608},
}
Figure