Detecting spurious effects in cross-lagged panel models: Triangulation is not a valid test DOI
Richard E. Lucas, Rebekka Weidmann, Mark J. Brandt

et al.

European Journal of Personality, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 1, 2024

The cross-lagged panel model (CLPM) is an analytic technique used to examine the reciprocal causal effects of two or more variables assessed on occasions. Although widely used, CLPM has been criticized for relying implausible assumptions, violation which can often lead biased estimates effects. Recently, a triangulation method proposed identify spurious in simple analyses (e.g., Sorjonen, Melin, & 2024). We use simulations and discussion formulas underlying regression coefficients show that this does not provide valid indicator spuriousness. This identifies true as realistic situations should be diagnose whether effect estimated from not. There are clear reasons doubt CLPM, but results do add information about such spurious.

Language: Английский

Prospective effects of mindfulness on anxiety and depressive symptoms may be spurious: Simulated reanalysis of a meta-analytic cross-lagged panel analysis DOI Creative Commons
Kimmo Sorjonen, Bo Melin

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(5), P. e0302141 - e0302141

Published: May 13, 2024

A recent meta-analysis claimed decreasing prospective effects of acting with awareness and non-reacting, two facets dispositional mindfulness, on subsequent anxiety depressive symptoms. However, the meta-analytic cross-lagged were estimated while adjusting for a prior measurement outcome variable it is known that such adjusted may be spurious due to correlations residuals regression mean. We fitted competing models simulations same data found mindfulness symptoms probably spurious. It important researchers aware limitations effects, meta-analytically or not, in order not overinterpret findings.

Language: Английский

Citations

9

Spurious prospective effects between general and domain-specific self-esteem: A reanalysis of a meta-analysis of longitudinal studies DOI Creative Commons
Kimmo Sorjonen, Bo Melin

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(2), P. e0298158 - e0298158

Published: Feb. 13, 2024

A recent meta-analysis, of 38 studies with data from 43 independent samples (total N = 24,668), claimed evidence for positive reciprocal prospective effects, and hence both top-down bottom-up processes, between general domain-specific self-esteem. However, the meta-analytic cross-lagged effects were estimated while adjusting a prior measurement outcome variable it is known that such adjusted may be spurious due to correlations residuals regression mean. In present reanalyses, we found all spurious. Consequently, claims about increasing processes self-esteem can questioned. It important researchers aware limitations panel analyses, analyses correlational in general, order not overinterpret findings.

Language: Английский

Citations

7

Questioning the Reciprocal Effects Model of Academic Self-Concept and Achievement: A Reanalysis of a Meta-Analysis of Longitudinal Studies and a Simulation DOI Creative Commons
Kimmo Sorjonen, Michael Ingre, Bo Melin

et al.

SAGE Open, Journal Year: 2024, Volume and Issue: 14(4)

Published: Oct. 1, 2024

A recent meta-analysis claimed to provide evidence that academic self-concept and achievement have reciprocal prospective effects on each other (reciprocal model). However, were estimated while adjusting for a prior measurement of the outcome, this method is susceptible spurious findings due correlations with residuals regression mean. Here we re-analyze meta-analytic show different plausible models can support opposing claims: either had an increasing or decreasing effect achievement, vice versa. Consequently, claims beyond positive cross-sectional correlation between including model, be questioned. The validated by analyses simulated data, which indicated true not necessary observed associations. We further propose extended skill development model (ESDM) as more parsimonious alternative model.

Language: Английский

Citations

4

Meta-analytic prospective associations between self-esteem and eating disorders appear to be spurious: a reanalysis and comment on Krauss et al. (2023) DOI Creative Commons
Kimmo Sorjonen, Ata Ghaderi, Bo Melin

et al.

Frontiers in Psychology, Journal Year: 2025, Volume and Issue: 16

Published: Feb. 7, 2025

In a recent meta-analysis, Krauss et al. found support for reciprocal model of low self-esteem and eating disorders where, in vicious circle, makes people more vulnerable to developing disorders, turn, scars individuals' self-esteem. However, the present reanalyses same meta-analytic data, we that prospective effects between are likely spurious, meaning they do not reflect true causal effect, but rather correlations with residuals regression mean. Consequently, claims by can be challenged. To avoid statistical artifacts, recommend researchers fit, as did study, complementary models their data order evaluate if may genuinely increasing or decreasing appear spurious.

Language: Английский

Citations

0

Distorted meta-analytic findings on peer influence: A reanalysis DOI Creative Commons
Kimmo Sorjonen, Gustav Nilsonne, Bo Melin

et al.

Heliyon, Journal Year: 2023, Volume and Issue: 9(11), P. e21458 - e21458

Published: Oct. 24, 2023

In a recent meta-analysis, Giletta et al. (2021) [1] found positive effect of peers' behavior at time 1 on target youths' 2 while adjusting for and claimed to have quantified peer influence. However, it is established that controlled cross-lagged effects could be due correlations with measurement errors reversion in the direction mean rather than true decreasing or increasing effects. Here, reanalysis same meta-analytic data as used by al., we influence, operationalized may been distorted (i.e. spurious). We do not claim influence does exist, but hard, maybe even possible, prove analyses observational exist. Difficulties causal common all areas research specific

Language: Английский

Citations

8

Triangulation does not support reported prospective effects on work engagement: Simulated reanalyses of twelve cross-lagged panel models DOI Open Access
Kimmo Sorjonen, Filippa Folke, Bo Melin

et al.

Published: May 17, 2024

Studies using cross-lagged panel models (CLPM) have shown prospective effects of initial scores on various predictors, e.g., job control and depressive symptoms, subsequent work engagement when adjusting for engagement. However, it is well established that may be spurious due to correlations with residuals regression the mean. Here, we reanalyzed data simulated resemble in twelve studies CLPM, suggesting increasing/decreasing Through triangulation, show all these suggested appear been rather than truly increasing/decreasing. Researchers need aware correlations, including models, do not prove causality. For improved causal inference, recommend researchers triangulate scrutinize by fitting complementary their data.

Language: Английский

Citations

1

Detecting spurious effects in cross-lagged panel models: Triangulation is not a valid test DOI
Richard E. Lucas, Rebekka Weidmann, Mark J. Brandt

et al.

European Journal of Personality, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 1, 2024

The cross-lagged panel model (CLPM) is an analytic technique used to examine the reciprocal causal effects of two or more variables assessed on occasions. Although widely used, CLPM has been criticized for relying implausible assumptions, violation which can often lead biased estimates effects. Recently, a triangulation method proposed identify spurious in simple analyses (e.g., Sorjonen, Melin, & 2024). We use simulations and discussion formulas underlying regression coefficients show that this does not provide valid indicator spuriousness. This identifies true as realistic situations should be diagnose whether effect estimated from not. There are clear reasons doubt CLPM, but results do add information about such spurious.

Language: Английский

Citations

0