Abstract

Two simulation studies were conducted to investigate the effects of the practice of item parceling. In Study 1, unidimensional sets of normally and nonnormally distrib-uted item-level data were categorized into 2-, 3-, and 4-item parcels. Analyses re-vealed that the use of item parcels resulted in better fitting solutions, as measured by the root mean squared error of approximation (RMSEA), comparative fit index (CFI), and chi-square test, when items had a unidimensional structure. Parceled solu-tions also resulted in less bias in estimates of structural parameters under these condi-tions than did solutions based on the individual items. In Study 2 the issue of whether the use of item parceling could mask a known multidimensional factor structure among a set of items was investigated. Results indicated that certain types of item parceling can obfuscate a multidimensional factor structure in such a way that ac-ceptable values of fit indexes are found for a misspecified solution. In addition, par-celing under these conditions was found to result in bias in the estimates of structural parameters. Although parceling can ameliorate the effects of coarsely categorized and nonnormally distributed item-level data when the items are unidimensional, the