Previous research obtained inconsistent findings as to whether implicit and explicit measures of prejudice are related. According to our view, some of these inconsistencies are due to whether implicit measures assess exclusively the strength of association between the social category and the evaluation or the activation of the social category in addition. Derived from recent theorizing we suggest that it is especially associative strength between the social category and its evaluation that determines the endorsement of prejudice assessed in explicit measures. To test this assumption we used the Implicit Association Test (IAT) to assess the associative strength of German students' prejudice towards Turkish people and the Blatant-and-Subtle-Prejudice-Scale (BS scale) as an explicit measure of prejudice. We obtained evidence for prejudice at a relative level in that the category of Turks was more closely associated with a negative evaluation and the category of Germans more closely associated with a positive evaluation. In line with our assumptions the implicit measure was correlated with the explicit measure in that the stronger the representation of Turks was associated with a negative evaluation the more prejudiced individuals responded in the explicit prejudice measure. Using a trait assignment technique, we found that neither associative strength nor the endorsement of the prejudice in the explicit measure were related to the contents of the stereotype. Copyright © 2001 John Wiley & Sons, Ltd.