Lars Penke, Jan Eichstaedt, and Jens B Asendorpf (2006)

Single attribute implicit association tests (SA-IAT) for the assessment of unipolar constructs: The case of sociosexuality.

Experimental Psychology, 53:283-291.

A major problem with Implicit Association Tests (IATs) is that they require bipolar attributes (e.g., good-bad). Thus, IAT effects for an attribute category can be interpretated only relative to an opposite category. Problems arise if there is no clear opposite category; in this case a neutral category can be used, although it induces systematic error variance and thus reduces validity. The present study suggests that this problem can be solved using single-attribute IATs (SA-IATs). Sociosexuality (the tendancy to engage in uncommited sex) was expected to be related at the implicit level to stronger stranger-sex associations. An IAT was constructed that used conversation as a neutral attribute, it showed satisfactory reliability but only low correlations with explicit sociosexuality. An alternative SA-IAT with sex as the only attribute showed a similar reliability but higher correlations with explicit sociosexuality.