Scale Development on a Limited Budget: Guidelines for Statisticians

• Activity Number: 589
• Type: Topic Contributed
• Date/Time: Thursday, August 6, 2009 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Consulting
• Author(s): Richard F. Ittenbach*+ and Amy E. Cassedy and Todd G. Nick and Justine Shults
• Companies: Cincinnati Children's Hospital Medical Center and Cincinnati Children's Hospital Medical Center and Arkansas Children's Hospital and University of Pennsylvania
• Keywords: scale development ; psychometrics ; consulting ; item-response theory

Abstract

Increasingly, clients are turning to statisticians with questions about measurement of unobservable events and constructs. Although standards exist for the development of large, standardized instruments and surveys, guidelines for locally derived measures are far more elusive. Helping clients navigate the scale development process with limited time and resources but without a loss of statistical rigor and precision remains challenging at best. The purpose of this paper will be to identify and describe key steps in the scale development process with recommendations as to how statisticians can help their clients in this rapidly evolving area of statistics. Attention will be devoted to the three major theoretical frameworks underlying the scale development process, namely, classical true-score theory, item-response theory, and rasch measurement theory.