A Procedure for Evaluating Sensitivity to Within-Person Change: Can Mood Measures in Diary Studies Detect Change Reliably?

The recent growth in diary and experience sampling research has increased research attention on how people change over time in natural settings. Often however, the measures in these studies were originally developed for studying between-person differences, and their sensitivity to within-person changes is usually unknown. Using a Generalizability Theory framework, the authors illustrate a procedure for developing reliable measures of change using a version of the Profile of Mood States (POMS; McNair, Lorr, & Droppleman, 1992) shortened for diary studies. Analyzing two data sets, one composed of 35 daily reports from 68 persons experiencing a stressful examination and another composed of daily reports from 164 persons over a typical 28-day period, we demonstrate that three-item measures of anxious mood, depressed mood, anger, fatigue, and vigor have appropriate reliability to detect within-person change processes.