Scale Construction and Psychometric Considerations
Generate an AI Snapshot to get a quick, structured summary of this paper.
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
Abstract
This chapter discusses the construction of scales. Scales are common features of social science research, reflecting the belief that data in the social sciences should be quantitative in form. This chapter focuses on the construction of self-report scales, which include personality scales, interest inventories, measures of attitudes, opinions, needs and values, and many of the ad hoc measures constructed or used by scholars in the social sciences for various purposes other than the testing of ability and achievement. This chapter concludes with two observations and a prescription. First, the list of published self-report scales is too long and is growing longer at an alarming rate. Second, the .50 barrier on validity coefficients appears to be alive and well. Furthermore, to understand better the correlation between construct scale and criterion scale, it is recommended that validity be examined at the item level. Finally, in scale construction—to paraphrase a saying from another field of endeavor—validity is not the only thing, it is everything.
