A Longitudinal Investigation of the Associations Among Parenting, Deviant Peer Affiliation, and Externalizing Behaviors: A Monozygotic Twin Differences Design

Jinqin Hou, Zhiyan Chen, Misaki N. Natsuaki, Xinying Li, Xiaodong Yang, Jie Zhang, Jianxin ZhangTwin 2013 · DOI

Abstract

Non-shared parenting and deviant peer affiliation are linked to differences in externalizing behaviors between twins. However, few studies have examined these two non-shared environments simultaneously. The present study examined the transactional roles of differential parenting (i.e., warmth and hostility) and deviant peer affiliation on monozygotic (MZ) twin differences in externalizing behaviors using a two-wave longitudinal study of twins and their parents. The sample consisted of 520 pairs of MZ twins (46.5% males, 53.5% females), with a mean age of 13.86 years (SD = 2.10) at the T1 assessment, residing in Beijing, China. The association between non-shared hostility in parenting and adolescent externalizing behaviors was mainly explained by a child-driven effect whereby the twin with a higher level of externalizing behaviors than his or her co-twin was more likely to receive more hostility from the parents. Similarly, the relationship between deviant peer affiliation and adolescent externalizing behaviors supported the selection effect whereby the twin with a higher level of externalizing behaviors than his or her co-twin was more likely to affiliate with deviant peers.

Introduction

Why do siblings—even monozygotic (MZ) twins with identical DNA—turn out so differently? Developmental psychologists categorize environmental influences into two buckets: shared (factors that make siblings similar, such as family income) and non-shared (unique experiences that make siblings different). While non-shared environments like "differential parenting" (one child receiving more warmth or hostility than the other) and "deviant peer affiliation" are linked to externalizing behaviors (aggression, rule-breaking), the direction of causality is often unclear. Does harsh parenting cause a child to act out, or does a difficult child elicit harsh parenting? This study employs a longitudinal design to untangle these transactional relationships.

Methods

The researchers utilized a Monozygotic Twin Differences Design. Because MZ twins share 100% of their genes and their shared family environment, any difference in behavior between Twin A and Twin B must be due to non-shared environmental factors.

The study used data from the Beijing Twin Study (BeTwiSt), involving 520 pairs of MZ twins (average age 13.86 at T1). Data was multi-informant, including self-reports from twins and reports from both mothers and fathers. By calculating "difference scores" (Twin A minus Twin B) for parenting, peers, and behavior at two time points (T1 and T2, roughly 1.5 years apart), the team used cross-lagged path analysis. This model, illustrated in Figure 1, allowed researchers to determine if a difference in parenting at T1 predicted a change in behavioral differences at T2 (a parent-driven effect), or if the behavioral difference at T1 predicted a change in parenting at T2 (a child-driven effect).

Path diagram of the cross-lagged model for twin differences.

Key Results

The analysis revealed significant child-driven and selection effects. As shown in Figure 4, externalizing behaviors at T1 significantly predicted increases in parental hostility at T2 ($b_{31}$ = 0.12 for mothers, 0.14 for fathers). Conversely, differential parenting at T1 did not significantly predict later behavioral differences.

Regarding peer groups, behavioral differences at T1 predicted differences in deviant peer affiliation at T2 ($b_{32}$ = 0.15–0.17). There was no evidence for a "socialization effect" (where peers cause the behavior) in this specific developmental window. Furthermore, twin differences in behavior and environment showed moderate stability over the 1.5-year period ($r$ = .17 to .37), suggesting that once a "gap" opens between twins, it tends to persist.

Standardized path coefficients and model fit indices for the four tested models.

Discussion

The most striking takeaway is that adolescents are active agents who shape their social worlds. The results suggest that twins create their own "environmental niches." A twin who is slightly more aggressive than their co-twin at the start of the study is more likely to gravitate toward "troublemaker" friends later on.

Interestingly, the data showed that twins who hung out with more deviant peers actually received less hostility from parents over time. The authors suggest this might be a "withdrawal" strategy—parents may back off to avoid further damaging a strained relationship when they see their child's social circle deteriorating.

Limitations and Future Work

While the study shows how differences are maintained and amplified, it doesn't explain what caused the twins to differ at T1 initially. Future research should investigate early childhood precursors. Additionally, with only two time points, mapping the full dynamic cycle of these transactions is difficult; a three-wave study would allow for more complex modeling. Finally, the cultural context of Beijing may influence how these child-driven effects manifest compared to Western contexts.

Summary updated

Abstract

Non-shared parenting and deviant peer affiliation are linked to differences in externalizing behaviors between twins. However, few studies have examined these two non-shared environments simultaneously. The present study examined the transactional roles of differential parenting (i.e., warmth and hostility) and deviant peer affiliation on monozygotic (MZ) twin differences in externalizing behaviors using a two-wave longitudinal study of twins and their parents. The sample consisted of 520 pairs of MZ twins (46.5% males, 53.5% females), with a mean age of 13.86 years (SD = 2.10) at the T1 assessment, residing in Beijing, China. The association between non-shared hostility in parenting and adolescent externalizing behaviors was mainly explained by a child-driven effect whereby the twin with a higher level of externalizing behaviors than his or her co-twin was more likely to receive more hostility from the parents. Similarly, the relationship between deviant peer affiliation and adolescent externalizing behaviors supported the selection effect whereby the twin with a higher level of externalizing behaviors than his or her co-twin was more likely to affiliate with deviant peers. The theoretical and practical implications of these findings are discussed.
BibTeX
@inproceedings{hou2013longitudinal,
  author = {Jinqin Hou and Zhiyan Chen and Misaki N. Natsuaki and Xinying Li and Xiaodong Yang and Jie Zhang and Jianxin Zhang and Jianxin Zhang and Jianxin Zhang},
  title = {{A Longitudinal Investigation of the Associations Among Parenting, Deviant Peer Affiliation, and Externalizing Behaviors: A Monozygotic Twin Differences Design}},
  booktitle = {Twin},
  year = {2013},
  doi = {10.1017/thg.2013.24},
}
Figure