Power analysis for loglinear growth models

Abstract

Multilevel analysis is often used to analyse longitudinal growth data. However, sample size calculations in an a priori power analysis are often based on simpler models. As a result, sample sizes may be inadequate or unnecessarily high.This article shows how to calculate sample sizes for a log-linear growth model using a user-friendly online app. The methodology is illustrated using a psychotherapy trial.

Authors 3

1. Mirjam Moerbeek

   Utrecht University

   Affiliation as printed

   Department of Methodology and Statistics , Utrecht University , Utrecht , The Netherlands

2. Kim de Jong Aachen Institute of Psychology, the Netherlands

   Leiden University

   Affiliation as printed

   Leiden University , Institute of Psychology , The Netherlands

3. Bram Bovendeerd

   University of Groningen · Dimence

   Affiliation as printed

   Department of Clinical Psychology and Experimental Psychopathology , Faculty of Behavioural and Social Sciences , University of Groningen , The Netherlands

   GGZ center for mental health care , Dimence , The Netherlands