Inconclusive prospective associations between self-esteem and internalizing symptoms: A comment on Solberg et al. (2025)

Based on findings from the random intercept cross-lagged panel model (RI-CLPM), a recent study concluded decreasing prospective associations between self-esteem and internalizing symptoms. However, the RI-CLPM is susceptible to spurious findings. Here, we reanalyzed the data with alternative models and found discrepant decreasing and increasing associations depending on the analyzed model. Hence, conclusions in the previous study may be challenged. It is important for researchers to bear in mind that correlations in observational data, including associations in the RI-CLPM, may be spurious in order not to overinterpret findings. We recommend researchers to analyze data with alternative models and to base conclusions on a juxtaposition of the findings.