Questionable mediated effect of peer victimization on prosocial behavior in adolescents: A reanalysis of Guo et al. (2026)
Abstract
Using the covariance matrix reported by Guo et al., we reassessed the evidence supporting their mediation model relating peer victimization to lower levels of prosocial behavior among adolescents. We demonstrate that the empirical associations can be reproduced with high accuracy by an alternative latent-variable model that does not depend on the order of influence proposed in the original article. In this model, self-reported peer victimization, psychological capital, empathy, and prosocial behavior are conceptualized as indicators of a broader latent construct, namely core self-evaluation. The findings illustrate a fundamental issue in observational research: different theoretical models may account equally well for the same covariance structure. Consequently, correlational data alone provide limited grounds for favoring one specific explanatory mechanism over competing alternatives. Consequently, researchers should consider alternative specifications before drawing strong conclusions about underlying processes.
Related
- Questionable mediated effect of physical activity on music performance anxiety: A reanalysis of Liu et al. (2026)
- Questionable mediated effect of short-video addiction on social connection among older adults: A reanalysis of Guo et al. (2026)
- Inconclusive mediated effect of adverse childhood experiences on suicide risk among depressed adolescents: A reanalysis of Ye et al. (2026)
- Questionable prospective effects between shame, loneliness, and non-suicidal self-injury among Chinese adolescents: A reassessment of Chen et al. (2026)
- Questionable mediated effect of self-management on athletic performance: A comment on Fan et al. (2026)
BibTeX
@misc{sorjonen2026questionable,
title = {{Questionable mediated effect of peer victimization on prosocial behavior in adolescents: A reanalysis of Guo et al. (2026)}},
author = {Kimmo Sorjonen and Bo Melin and Marika Melin},
howpublished = {PsyArXiv},
year = {2026},
month = sep,
doi = {10.31234/osf.io/cvtmj_v1},
url = {https://osf.io/preprints/psyarxiv/cvtmj_v1/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Questionable mediated effect of peer victimization on prosocial behavior in adolescents: A reanalysis of Guo et al. (2026)
AU - Kimmo Sorjonen
AU - Bo Melin
AU - Marika Melin
PY - 2026
DA - 2026/09/11
PB - PsyArXiv
DO - 10.31234/osf.io/cvtmj_v1
UR - https://osf.io/preprints/psyarxiv/cvtmj_v1/
AB - Using the covariance matrix reported by Guo et al., we reassessed the evidence supporting their mediation model relating peer victimization to lower levels of prosocial behavior among adolescents. We demonstrate that the empirical associations can be reproduced with high accuracy by an alternative latent-variable model that does not depend on the order of influence proposed in the original article. In this model, self-reported peer victimization, psychological capital, empathy, and prosocial behavior are conceptualized as indicators of a broader latent construct, namely core self-evaluation. The findings illustrate a fundamental issue in observational research: different theoretical models may account equally well for the same covariance structure. Consequently, correlational data alone provide limited grounds for favoring one specific explanatory mechanism over competing alternatives. Consequently, researchers should consider alternative specifications before drawing strong conclusions about underlying processes.
ER -