Reconceptualizing Diversity Training Predictors of Training Success

Background

Consistent with the definition of diversity training in Pendry et al. (2007), Bezrukova et al. (2016) define diversity training as “a distinct set of instructional programs aimed at facilitating positive intergroup interactions, reducing prejudice and discrimination, and enhancing the skills, knowledge, and motivation of participants to interact with diverse others” (p. 6).

Kirkpatrick (2006) is a prominent training model that include levels for measuring results, such as the application of what was learned and impact of training. However, these levels involve multiple collaborators with reliance on administrators and other individuals beyond diversity trainers to collect the data and are far less likely to be implemented. Bezrukova et al. (2012) reviewed and examined one hundred and seventy-eight (178) articles whose authors have investigated numerous aspects of diversity training programs in Higher Education and in other workplace settings. Their review revealed only 11 “results-level” data for diversity training (Bezrukova et al., 2012). Kirkpatrick and Kirkpatrick (2006) labeled “results-level” measures as measures relating to grievances, turnover, promotions, and the recruitment of individuals from marginalized populations. Bezrukova et al. (2012) noted that it is rare to see “results-level” data for diversity training (only 11 of 178 in their study).

Kirkpatrick’s Training Model seeks to capture increased job performance and maximum results with a four-level approach. Levels one and two are within a trainer’s capacity. Level one assesses participants reaction to the training, whether they found it favorable, engaging, and relevant to their jobs. Level two assesses what was learned from the training and whether targeted outcomes were met. This level seeks to gauge the degree to which participants acquired the intended knowledge, skills, and confidence as a result of participating in a training. These two levels can be assessed soon after the training by asking survey questions along with a post-test examination of key points and aims of the training. However, levels three and four are beyond a trainer’s “reach” and would involve administrator approval and administrative support. Level three assesses behavior of participants after they have returned to their roles at work. Specifically, whether they are applying what they learned during the training. Level four is more involved than level three. Level four assesses the long-term results associated with the training, and if the targeted outcomes were met. Moreover, since level four is the finality of the model, support systems and accountability become more transparent. Considering the tasks associated with “results-level” data, it’s not surprising that only eleven (11) out of one hundred and seventy-eight (178) articles reviewed by Bezrukova et al. (2012) included data associated with Kirkpatrick’s level three and level four. Leadership support is necessary.