Inconclusive prospective effects between social participation and mental health: A reassessment of Peng et al. (2026)
Abstract
Peng et al. interpreted their longitudinal findings as indicating that social participation promotes later improvements in mental health and that mental health, in turn, enhances subsequent social participation. However, the presence of statistically significant cross-lagged coefficients does not necessarily demonstrate true prospective effects, as comparable estimates may emerge from alternative statistical mechanisms. Using simulated data that reproduced both the sample size and correlation matrix reported by Peng et al., we evaluated whether the original conclusions would remain consistent across several defensible analytic approaches. The findings varied considerably across specifications. Some models yielded positive prospective associations, whereas others produced negative or non-significant estimates. When results from all specifications were combined using meta-analytic techniques, no overall significant effect was detected. These results suggest that the evidence reported by Peng et al. is insufficient to support firm conclusions regarding reciprocal prospective influences between social participation and mental health. Consistent with recommendations from the multiverse-analysis literature, inferences should be based on evidence synthesized across plausible analytic alternatives rather than on a single modeling strategy.
Related
- Inconclusive prospective effects between technology acceptance, classroom distraction, and academic engagement among Chinese high school students: A reassessment of Li and Chen (2026)
- Inconclusive prospective effects between fear of failure, academic buoyancy, and academic procrastination among Chinese primary school students: A reassessment of Fang et al. (2026)
- Questionable prospective effects between shame, loneliness, and non-suicidal self-injury among Chinese adolescents: A reassessment of Chen et al. (2026)
- Inconclusive evidence for a mediated association between childhood adversity and resilience: A reexamination of Suriano and Valentino (2026)
- Inconclusive mediated effect of adverse childhood experiences on suicide risk among depressed adolescents: A reanalysis of Ye et al. (2026)
BibTeX
@misc{sorjonen2026inconclusive,
title = {{Inconclusive prospective effects between social participation and mental health: A reassessment of Peng et al. (2026)}},
author = {Kimmo Sorjonen and Bo Melin and Marika Melin},
howpublished = {PsyArXiv},
year = {2026},
month = sep,
doi = {10.31234/osf.io/zgkxp_v1},
url = {https://osf.io/preprints/psyarxiv/zgkxp_v1/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Inconclusive prospective effects between social participation and mental health: A reassessment of Peng et al. (2026)
AU - Kimmo Sorjonen
AU - Bo Melin
AU - Marika Melin
PY - 2026
DA - 2026/09/07
PB - PsyArXiv
DO - 10.31234/osf.io/zgkxp_v1
UR - https://osf.io/preprints/psyarxiv/zgkxp_v1/
AB - Peng et al. interpreted their longitudinal findings as indicating that social participation promotes later improvements in mental health and that mental health, in turn, enhances subsequent social participation. However, the presence of statistically significant cross-lagged coefficients does not necessarily demonstrate true prospective effects, as comparable estimates may emerge from alternative statistical mechanisms. Using simulated data that reproduced both the sample size and correlation matrix reported by Peng et al., we evaluated whether the original conclusions would remain consistent across several defensible analytic approaches. The findings varied considerably across specifications. Some models yielded positive prospective associations, whereas others produced negative or non-significant estimates. When results from all specifications were combined using meta-analytic techniques, no overall significant effect was detected. These results suggest that the evidence reported by Peng et al. is insufficient to support firm conclusions regarding reciprocal prospective influences between social participation and mental health. Consistent with recommendations from the multiverse-analysis literature, inferences should be based on evidence synthesized across plausible analytic alternatives rather than on a single modeling strategy.
ER -