Instructional set and the structure of responses to rating scales

The extent to which psychosocial phenomena are categorical or dimensional has been a topic of interest in recent years, in part spurred by the development of taxometric analysis as a statistical procedure for evaluating underlying structure (Meehl & Yonce, 1994, 1996; Waller & Meehl, 1998). Beauchaine & Waters (2003) suggested that the structure of scores on rating scales may be susceptible to instructional manipulation. However, they only investigated circumstances in which the ratings targets were unknown to the raters. The current study examined ratings of a small set of familiar targets. A sample of 608 undergraduate students completed five rating scales describing themselves and a significant other. Students were randomly assigned to instructional sets encouraging either dimensional or categorical ratings. Results consistently indicated a dimensional structure across both instructional sets and targets. The findings suggest that under normal conditions for the use of rating scales, instructional set does not affect the data structure.