New Insights in Power and Power Analysis in Mediation Models

Abstract

Mediation analysis is commonly used in social-personality psychology to evaluate potential mechanisms of effects. With the recent replicability crisis, researchers are turning to power analysis to help plan studies; however, power analysis for mediation is not implemented in popular software (e.g., G*Power). Our symposium includes two presentations focusing on implementation of power analysis for mediation: (1) describing easy-to-use tools for implementing power analysis (e.g., pwr2ppl R package), and (2) evaluating whether different inferential methods result in similar recommended sample sizes and the role of assumption violations in these differences. Two presenters focus on study characteristics which can affect power: (1) use of the bias-corrected confidence interval and alternatives which better balance power and type I error, and (2) how measurement error on the mediator can impact power and how to correct this issue with latent variable models. Presentations will include applied examples, aimed at a social-personality audience, and provide concrete steps for increasing the validity and replicability of mediation analyses conducted in social-personality research. (Symposium Presented at SPSP 2021)

Power Analysis for Conditional Indirect Effects: A Tutorial for Conducting Monte Carlo Simulations with Categorical Exogenous Variables

“…Given the increasing popularity of moderated mediation models, as represented both in theory and in primary empirical studies that test such models, considering ways of optimizing tests of such effects is of paramount importance. This is particularly so, given increasing criticism of these models in the literature (e.g., Rohrer et al, 2021) and with respect to their statistical power, specifically (Montoya et al, 2021). Our hope is that the "toolkit" presented here will inspire researchers to more carefully consider the sample size requirements necessary to test moderated POWER ANALYSIS FOR MODERATED MEDATION 38 mediation models and adopt a more critical perspective on the strengths and limitations of such 841 models.…”

Power analysis for conditional indirect effects: A tutorial for conducting Monte Carlo simulations with categorical exogenous variables

“…Given the increasing popularity of moderated mediation models-as represented both in theory and in primary empirical studies that test such models-considering ways of optimizing tests of such effects is critical. This is particularly important, given increasing criticism of these models in the literature (e.g., Rohrer et al, 2021), specifically with respect to their statistical power (Montoya et al, 2021). Although we have mentioned other ways to model moderated effects, the methods discussed here can also be extended to model multiple (parallel or serial) mediators, as discussed in the tutorial for Monte Carlo power analysis by Schoemann et al (2017).…”

Responses of the public towards the government in times of crisis

“…To compute statistical power, we were unable to find a procedure to test a complex moderated moderated mediation like PROCESS Model 21 (see Montoya et al, 2021 ). Thus, we employeded a pragmatic approach and tested power analysis for the most relevant parts of the decomposed model.…”

“… Thankful – .78 – .85 Grateful – .81 Inspired – .71 Touched – .64 Uplifted – .51 Awe – – Admiration – .73 Attitude towards supporting government (5‐ point semantic differential) Source: Ajzen & Fishbein ( 1980 ). Bad‐good .94 .91 .95 .92 Unfavourable‐favourable .92 .87 Negative–positive .92 .90 Trust in the government (5‐ point strongly disagree‐strongly agree) Source: Muthusamy & White ( 2005...…”