Whenever we measure a psychological variable for multiple persons we typically capture both stable differences between persons, resulting in ``between-person variance'', and fluctuations within a person over time, resulting in ``within-person variance''. Both types of variance are of key interest to psychology: Stable differences between persons are the main interest in psychological research on, for example, traits and risk factors. Variation in variables over time are the main interest in studies that focus on psychological processes, change, and development. Failing to distinguishing between these sources of variance in our data can quickly result in wrong conclusions about both types of variation. This is an essential problem for all studies with data from multiple persons – cross-sectional and longitudinal. However, for many psychologists the problem with failing to distinguish within- and between-person variance remains obscure. My aim is to provide an entry-level review of the problem: To elucidate related jargon, the potential consequences of the problem, and various practical solutions to it. To illustrate the problem, I provide an online app in which researchers can change various settings to see how they influence the conclusions of analyses.
No takes yet. Share an insight, caveat, or question.
Noémi Katalin Schuurman (2023) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: