Organizations invest heavily in diversity training that aims to correct the biases of individual decision makers, yet such programs seldom change hiring outcomes. Most of these efforts are subtractive: they try to dismantle stereotypes, and people resist overt attempts to change their attitudes. This talk takes an additive approach: the arrangement of the choice options, rather than the mind of the chooser, offers the more tractable lever against hiring discrimination. Three projects make the case. The first shows that partitioning candidates by gender, nationality, or university leads decision makers, including experienced human resource professionals, to select more diverse candidates on the partitioned dimension without compromising the average competence of the selected candidates. Simply telling decision makers that the candidates belong to different categories does not increase diversity: the partitions themselves, not awareness of the categories, drive the effect. The second project shows that preselecting a demographically diverse set of candidates increases the diversity of the final selections even when decision makers abandon the preselected set. Mental inertia, the mechanism that accounts of the default bias typically invoke, therefore cannot explain this effect; mediation analyses instead trace it to perceived organizational expectations about diversity, and decision makers with stronger stereotypes respond more strongly to those expectations. The third project reconciles decades of inconsistent findings on serial position effects by comparing simultaneous choice, sequential evaluation, and end-of-sequence choice. A primacy effect emerges in all three formats, and attention allocation explains it: decision makers attend more to early candidates, and equalizing the time spent on each candidate attenuates the advantage. Placing qualified candidates from underrepresented groups early in the sequence therefore raises their odds of selection. Choice architecture tools like partitions, defaults, and sequence positions shift hiring outcomes without mentioning diversity, discrimination, gender, or race, and without sacrificing candidate quality, thereby providing an alternative approach to diversity training and implicit bias training.

wlucy2@ceibs.edu