ABSTRACT

Researchers often describe the questionnaires they have used as having “established reliability and validity”. They may then provide a reference to a paper reporting the psychometric properties of the scale developed for an entirely different population from the one to which the scales have subsequently been applied. For example, the Beck Depression Inventory (BDI) was developed as a measure of depression in the general population and was not developed specifically for people with diabetes. However valid and reliable a measure of depression the BDI may be with a general population sample, it cannot be assumed that it will be equally valid or reliable if used with respondents who have diabetes. Indeed, there are good reasons to suggest that the BDI is not a valid measure of depression in people with diabetes as several symptoms of depression in non-diabetic populations are symptoms that in diabetic populations may well arise from poor diabetes control rather than from depression (Bradley and Lewis, 1990). These issues are considered in more detail in Chapter 6 on the Well-being Questionnaire included in this volume.