Minimum Required Sample Size for Modelling Daily Cyclic Patterns in Ecological Momentary Assessment Data

Authors

  • Robin van de Maat Orcid
    Faculty of Psychology, Open Universiteit, Heerlen, the Netherlands
  • Johan Lataster Orcid
    Faculty of Psychology, Open Universiteit, Heerlen, the Netherlands; Department of Psychiatry and Psychology, School for Mental Health and Neuroscience, Maastricht University Medical Centre, Maastricht, the Netherlands
  • Peter Verboon Orcid
    Faculty of Psychology, Open Universiteit, Heerlen, the Netherlands

Abstract

Cyclical patterns in ecological momentary assessment (EMA) data on emotions have remained relatively underresearched. Addressing such patterns can help to better understand emotion dynamics across time and contexts. However, no general rules of thumb are readily available for psychological researchers to determine the required sample size for measuring cyclical patterns in emotions. This study, therefore, estimates the minimum required sample sizes—in terms of the number of measurements per time period and subjects—to obtain a power of 80% given a certain underlying cyclical pattern based on input parameter values derived from an empirical EMA dataset. Estimated minimum required sample sizes varied between 50 subjects and 10 measurements per subject for accurately detecting cyclical patterns with a large magnitude, to 60 subjects and 30 measurements per subject for cyclical patterns of small magnitude. The resulting rules of thumb for sample sizes are discussed with a number of considerations in mind.