How to Set Focal Categories for Brief Implicit Association Test? “Good” Is Good, “Bad” Is Not So Good

Abstract

Three studies were conducted to examine the validity of the four versions of BIATs that are supposed to measure the same construct but differ in shared focal category. Study 1 investigated the criterion validity of four BIATs measuring attitudes toward flower versus insect. Study 2 examined the experimental sensitivity of four BIATs by considering attitudes toward induced ingroup versus outgroup. Study 3 examined the predictive power of the four BIATs by investigating attitudes toward the commercial beverages Coke versus Sprite. The findings suggested that for the two attributes “good” and “bad,” “good” rather than “bad” proved to be good as a shared focal category; for two targets, so long as they clearly differed in goodness or valence, the “good” rather than “bad” target emerged as good for a shared focal category. Beyond this case, either target worked well. These findings may facilitate the understanding of the BIAT and its future applications.