Congeneric Modeling of Reliability Using Censored Variables

Publisher Website
Google Scholar
Abstract
This paper explores the use of Jöreskog's (1970) congeneric modeling approach to reliability using cen sored quantitative variables, and discusses the com pound problem of non-normality and attenuation that occurs when estimating censored continuous variables. Two monte carlo studies were conducted. The first study demonstrated the inappropriateness of using nor mal theory generalized least-squares (NTGLS) for esti mating reliability on censored variables. The second study compared three different estimation procedures —NTGLS, asymptotically distribution free (ADF) esti mators, and latent TOBIT estimators—as to their effi ciency in estimating individual and composite reliabil ity on censored variables. Results from the studies indicate that problems of non-normality and attenua tion must be addressed before accurate reliability esti mates may be obtained. Index terms: censored vari ables, congeneric model, covariance modeling, monte carlo study, reliability, TOBIT correlations.