The perils and pitfalls of comparing UK and US samples of people enrolled in an Arthritis Self-Management Program: The case of the Center for Epidemiological Studies-Depression (CES-D) Scale

The aim of this contribution from the field is to raise awareness and debate about the potential perils and pitfalls of attempting cross-cultural comparisons. Comparisons between samples are meaningful only if the measures employed have consistent psychometric properties. Arthritis-specific measures, modified for use in UK samples, were found to be reliable and valid. This article focuses on examination of dimensionality of the Center for Epidemiological Studies–Depression Scale (CES-D), a generic measure of psychological distress. Data were collected through self-administered questionnaires that were mailed to participants in an arthritis self-management program in the UK and the US. A 4-dimensional structure was identified in both samples: self-worth, depressed affect, positive affect, and somatic disturbance. However, with the exception of positive affect, the item composition of the subscales was subject to variation. This might be influenced by age, type of arthritis, cultural factors, or criterion contamination. In both the samples, positive affect might be an artifact of measurement. Although disappointing for our initial aim of comparing 2 apparently similar US and UK samples, this brief contribution has raised our awareness of the potential perils of assuming that a scale developed in one setting performs consistently when transposed to another setting. The limitations of this contribution from the field will be discussed.