Experimentally established expected effects in random-intercept cross-lagged panel models when causality holds DOI Open Access
Kimmo Sorjonen, Bo Melin

Опубликована: Май 28, 2024

There are indications that findings from random-intercept cross-lagged panel models (RI-CLPM) may be biased and spurious, similarly to traditional models. The objective of the present study was establish what effects expect in analyses with RI-CLPM when causality holds. We generated experimental data truly causal by estimating weight a container after adding or removing stones to/from container. Data were analyzed three complementary RI-CLPMs as well corresponding multilevel model (MLM) person-mean centered scores. results suggested if prior level on predictor X has within-individual increasing effect an outcome Y, we should show has: (1) A positive subsequent Y adjusting for Y; (2) negative (3) - difference. Corresponding scores MLM yield similar effects. In case decreasing effects, opposite signs (i.e., negative, positive, respectively). These experimentally established expected holds can help researchers triangulate scrutinize order observed might they appear spurious.

Язык: Английский

Effects in random-intercept cross-lagged panel models may be spurious: A simulated reanalysis and comment on Specker et al.’s (2024) study on post-migration stressors, emotion dysregulation, and symptoms of posttraumatic stress disorder DOI Open Access
Kimmo Sorjonen, Bo Melin

Опубликована: Март 30, 2024

Specker et al. (2024) conducted analyses with the random-intercept cross-lagged panel model (RI-CLPM) and concluded, for example, that emotion dysregulation post-migration stressors have increasing effects on posttraumatic stress disorder (PTSD) symptom hyperarousal among refugees. Here, we evaluated conclusion by through triangulation. We fitted several models data simulated to resemble analyzed used as empirical was not available us. The present results suggested simultaneous contradictory decreasing of hyperarousal. Due these findings, concluded longitudinal associations between constructs probably were spurious rather than truly increasing. This corroborated good fit a (the associations, MoSLA) where / due confounding common trait auto-correlated state factors. It is important researchers be aware correlations, including in RI-CLPM, do prove causality. Researchers are recommended scrutinize their findings from correlational With congruent models, conclusions genuine or corroborated. If, other hand case, different diverge, seem premature.

Язык: Английский

Процитировано

1

Inconclusive effects between executive functions and symptoms of psychiatric disorders in random-intercept cross-lagged panel models: A simulated reanalysis and comment on Halse et al. (2022) DOI Open Access
Kimmo Sorjonen, Bo Melin

Опубликована: Апрель 23, 2024

In a recent study of Norwegian children (N = 874), Halse et al. (2022) used random-intercept cross-lagged panel models (RI-CLPM) and concluded that their findings supported the assumption deficiencies in executive functions psychopathology are both cause consequence other. Here, we reanalyzed data simulated to resemble by with several complementary models. Our contradictory simultaneous increasing decreasing effects suggested prospective between deficits symptoms were spurious rather than truly increasing. Consequently, conclusions not own data. It is important for researchers bear mind correlations, including RI-CLPM, do prove causality. We recommend use, as did here, triangulation scrutinize from analyses observational

Язык: Английский

Процитировано

0

Experimentally established expected effects in random-intercept cross-lagged panel models when causality holds DOI Open Access
Kimmo Sorjonen, Bo Melin

Опубликована: Май 28, 2024

There are indications that findings from random-intercept cross-lagged panel models (RI-CLPM) may be biased and spurious, similarly to traditional models. The objective of the present study was establish what effects expect in analyses with RI-CLPM when causality holds. We generated experimental data truly causal by estimating weight a container after adding or removing stones to/from container. Data were analyzed three complementary RI-CLPMs as well corresponding multilevel model (MLM) person-mean centered scores. results suggested if prior level on predictor X has within-individual increasing effect an outcome Y, we should show has: (1) A positive subsequent Y adjusting for Y; (2) negative (3) - difference. Corresponding scores MLM yield similar effects. In case decreasing effects, opposite signs (i.e., negative, positive, respectively). These experimentally established expected holds can help researchers triangulate scrutinize order observed might they appear spurious.

Язык: Английский

Процитировано

0