Reproducibility of Implicit Association Test (IAT) – Case Study of Meta-Analysis of Racial Bias Research Claims

The Implicit Association Test, IAT, is widely used to measure hidden (subconscious) human biases – implicit bias – of many topics of interest: race, gender, age, ethnicity, religion stereotypes. There is a need to understand the reliability of these measures as they are being used in many decisions in society today. A study was undertaken to independently test the reliability of (ability to reproduce) racial bias research claims of Black−White relations based on IAT (implicit bias) and explicit bias measurements using statistical p-value plots. These claims were for IAT−real-world behavior correlations and explicit bias−real-world behavior correlations of Black−White relations in a meta-analysis. The p-value plots were constructed using data sets from the meta-analysis and the plots exhibited considerable randomness for all correlations examined. This randomness supports a lack of correlation between IAT (implicit bias) and explicit bias measurements with real-world behaviors of Whites towards Blacks. These findings were observed for microbehaviors (measures of nonverbal and subtle verbal behavior) and person perception judgments (explicit judgments about others). Findings of the p-value plots were consistent with the meta-analysis research claim that the IAT provides little insight into who will discriminate against whom. It was also observed that the amount of real-world variance explained by the IAT and explicit bias measurements was small – less than 5%.