The Effect of Using Item Parcels on Ad Hoc Goodness-of-Fit Indexes in Confirmatory Factor Analysis: An Example Using Sarason's Reactions to Tests

Abstract

The impact of using item parcels on ad hoc goodness-of-fit indexes in confirmatory factor analysis was examined using the Arabic version of Sarason's Reactions to Tests scale. Different numbers of parcels per factor and different numbers of items per parcel were used. Data were collected from a sample of 421 Arabic-speaking students from 2 similar high schools. Model fit in terms of the goodness-of-fit index, the comparative fit index, the Tucker-Lewis index, and the single-sample expected cross-validation index systematically improved as the number of items per parcel increased. However, the c2:degrees of freedom ratio and the root-mean-square error of approximation did not change systematically. The results were replicated with another sample of 374 students drawn from the same 2 schools.