Tuesday, March 11, 2008

The diversity-validity dilemma (+ free articles!!)

The latest issue of Personnel Psychology has some great articles in it and right now they're free! So before you do anything else, get while the gettin's good, because normally each article will run ya $30.

So what's in there? The main attraction is a great series of articles on the "diversity-validity" dilemma, which Pyburn, Ployhart, and Kravitz in their article on the legal context, define as:

"The ability of organizations to simultaneously identify high-quality candidates and establish a diverse work force can be hindered by the fact that many of the more predictive selection procedures negatively influence the pass rates of racioethnic minority group members (non-Whites) and women."

This article is a great short read that goes over the major legal points, including adverse impact and the major court cases.

The next article, by Ployhart and Holtz, is a print-and-save type article (yes it's that good) that summarizes the various strategies employers can use to help resolve the dilemma. The article includes a couple of great tables, including one that summarizes most selection mechanisms with their corresponding criterion-related validity and d-values (pp. 155-156) and another that summarizes the various resolutions to the dilemma (pp. 158-163).

Bottom line from that article? I'll let the authors say it:

"Among the most effective strategies, the only strategy that does not also reduce validity is assessing the full range of KSAOs." (bold added)

Hallelujah. Yes, certain assessment methods tend to work better than others (e.g., structured interviews, job knowledge tests) but the best approach is plain old fashioned good practice: Start with job analysis and use the testing methods that best target the knowledge, skills, abilities, and other characteristics (KSAOs) that rise to the top. It really is pretty simple.

The third article in the series is another fabulous one, this time targeting the role that affirmative action (AA) plays in the dilemma.

In it, Kravitz provides a great overview of the basis of AA, attitudes about AA, and provides some answers to some controversial issues, including:

- Does discrimination still occur? (Answer: you bet)
- What is the economic impact of AA on target groups? (A: it's complicated)
- What is the economic impact of AA on organizations? (A: apparently very little)
- Does AA lead to stigmatization of target group members by others? (A: it can)
- Does AA lead to self-stigmatization of target group members? (A: hard to say)

The article then wraps up with some great practical recommendations, the two most important of which are strong, visible, ongoing support of management and the development of an appropriate culture.

Last but not least, don't miss the other great content in this issue, including Mount et al.'s article, Incremental validity of perceptual speed and accuracy over general mental ability and Taylor et al.'s article The transportability of job information across countries.

Now get out there and get some free content!

1 comment:

HRagitator said...

The "diversity-validity" dilemma is an extremely sensitive subject for anyone to try to tackle, because it puts one between a rock and a hard place.

If tests are predictive and minorities score lower than non-minorities, then minorities must, in fact, be the poorer choice.

If, however, you assume that the reason that non-minorities score lower is not because they are poorer choices, but because of some other unknown factor. Than the test is not as predictive as you thought and shouldn't be used.

It places you between believing in the validity of tests (and thus being forced to accept a race-based difference) or believing the scientific method is fallacious (despite years of empirical support).

This makes me glad I'm not an academic!