Abstract
Despite evidence that Allen and Meyer’s scales measure three-component commitment in a reliable and valid manner, the literature contains recurring criticism of several scale items. Criticisms refer to violations of content validity, poor reflection of the construct, and a measurement artifact in the form of a reverse-scoring method factor. This study surveyed five published sets of factor pattern coefficients from exploratory and confirmatory factor analyses, all of which corroborate the consistently weak performance of these items across study samples. Confirmatory factor analysis was employed with two samples, one new and one previously reported, to compare original and revised scales. Results showed that substantially improved construct measurement is achievable with relatively modest scale revisions.
Get full access to this article
View all access options for this article.
