In theory construct validation secures good measurement by balancing all available evidence. But does it do so in practice? This chapter argues that validation of well-being scales as currently practiced is unduly selective about what evidence counts. It is not enough to check whether a measure correlates with other measures and indicators that background theory deems relevant if this background theory does not include normative and conceptual considerations about the nature of well-being. In this sense psychometrics commits theory avoidance. Its root is a disciplinary convention of evidential subjectivism, which reduces the big philosophical questions to technical exercises in statistics and factor analysis or reformulates them as reports of subjects’ opinions. To overcome these problems philosophical considerations need to be given a more central place in validation.
Keywords: construct validity; validation; psychometrics; validity; theory avoidance; evidential subjectivism; factor analysis
Chapter. 8297 words.
Subjects: Moral Philosophy
Full text: subscription required