Measurement equivalence: A comparison of methods based on confirmatory factor analysis and item response theory.

Abstract

Current interest in the assessment of measurement equivalence emphasizes 2 major methods of analysis. The authors offer a comparison of a linear method (confirmatory factor analysis) and a nonlinear method (differential item and test functioning using item response theory) with an emphasis on their methodological similarities and differences. The 2 approaches test for the equality of true scores (or expected raw scores) across 2 populations when the latent (or factor) score is held constant. Both approaches can provide information about when measurement nonequivalence exists and the extent to which it is a problem. An empirical example is used to illustrate the 2 approaches.

How Harmful are Survey Translations? A Test with Schwartz's Human Values Instrument

“…99-107) are not sufficiently adequate as they only deal with information on covariance structures but not with information on mean structures. However, MACS analysis (Sörbom, 1974(Sörbom, , 1978, as well as 'differential item functioning' approaches based on item response theory provide an adequate means to test for scalar equivalence of survey instruments across groups (Raju, Lafitte, & Byrne, 2002;Reise, Widaman, & Pugh, 1993;Stark, Chernyshenko, & Drasgow, 2006).…”

The measurement of guanxi: Introducing the GRX scale

“…These indices include the Chisquared statistics, the probability value, the comparative fit index (CFI), the goodness-of-fit index (GFI), the incremental fit index (IFI), the Tucker-Lewis index (TLI) and the root mean square error of approximation (RMSEA). The acceptable cut-off points of these indices were based on CFI, GFI, IFI, TLI > 0.90, and RMSEA < 0.08 Hoyle and Panter, 1995;Steenkamp and Baumgartner, 1998;Raju et al, 2002).…”

Measuring attitudes toward plagiarism: issues and psychometric solutions

“…That is, Rasch analysis is considered more appropriate for scale development and evaluation because it can provide detailed information about how particular items function relative to the level of the latent trait under investigation. In this study we use a combination of Traditional Test Theory and the Rasch Model in order to provide a more comprehensive evaluation of this measurement instrument (see Bartholomew et al, 2011;Raju et al, 2002).…”