Mixing the implicit: A Linear Mixed-Effects Models approach for a Rasch analysis of the Implicit Association Test and the Single Category Implicit Association Test

The indirect investigation of psychological constructs has become prominent in social sciences thanks to the so-called implicit measures. Different implicit measures can be administered concurrently to the same respondents for obtaining detailed and multifaceted information on the constructs of interest. In this study, a Rasch analysis of accuracy and time responses of two commonly used implicit measures is presented. The focus in on the concurrent administration of the Implicit Association Test (IAT; Greenwald et al., 1998) and the Single Category IAT (SC-IAT; Karpinski & Steinman, 2006). Linear Mixed-Effects Models are used to address the within– and between–measures sources of variability and to obtain a Rasch parametrization of the data. By disentangling the respondent’s contribution from the stimulus contribution to the observed responses, these models allow for delving deeper on the functioning of the IAT and the SC-IAT, as well as for grasping a better understanding of the processes driving a behavioral decision. Implications of the results for social sciences and future research directions are discussed.