An Introduction to Rasch Measurement for Scale Development and Person Assessment

Smith, Everett V. Jr.
Conrad, Karen M.
Chang, Karen
Piazza, Jo

¹University of Illinois at Chicago
²Purdue University, West Lafayette, IN
³Advocate Illinois Masonic Medical Center, Chicago

Offprints. Requests for offprints should be directed to Everett V. Smith, Jr., PhD, University of Illinois at Chicago, 1040 W. Harrision Street, M/C 147, Chicago, IL 60607. E-mail:[email protected]

Journal of Nursing Measurement 1(3):p 189-206, Winter 2002.

This article: a) outlines some of the limitations of True Score Theory, b) discusses some advantages of Rasch measurement, and c) compares the results of Rasch and factor analytic techniques applied to a common data set. Using a mail survey, 206 female members of a national nursing association completed Perceived Physical Ability Scale. Exploratory factor analysis identified one factor with a Cronbach alpha of .78. Rasch fit statistics also identified one dimension with a person reliability of .76. Rasch methods were also used to investigate the rating scale functioning, additional construct validity issues, and person diagnosis. Discrepancies between the two estimates of internal consistency are discussed and differences in the utility of the two methodologies are reviewed.