Preparation and Analyses of Implicit Attitude Measures: Challenges, Pitfalls, and Recommendations

• In Event: Best of Information Systems

Abstract

Implicit measures such as the Implicit Association Test (IAT) and the Personalized-IAT can be useful tools for studying automatic processes and socially sensitive topics. But with them come issues with data preparation and analysis because they rely on reaction time data. Dealing with reaction time data, particularly with the IAT or PIAT, can be complex, with many steps and alternatives available. Greenwald et al. (2003) offer guidelines for handling these types of data. However, these guidelines are often cited with little accompanying information as to which steps and alternatives were chosen, and the effects they had on the data. In addition, many steps in the guide reduce variation in the data, which could decrease validity. This manuscript reports what happens to IAT/PIAT scores when steps in the Greenwald et al. (2003) guide are omitted, and also offers some new alternatives to some of the steps in the guide.

Authors

• Morgan E. Ellithorpe, University of Pennsylvania

• David R. Ewoldsen, Ohio State U

• John Velez, Texas Tech University