I confess that I'm late to the party on this one. I meant to write about it ages ago, but forgot. Apologies.
A couple months ago, The Economist ran an article about workplace personality testing. The DiSC test is obviously not the only such assessment in the market. (Though it is clearly the best, we assert as a disinterested and completely objective party, with absolutely no vested interest whatsoever in DiSC. Oh, wait.)
Anyway, the article discusses talent management in general, and the extent to which subjects like leadership can or cannot be quantified. It's an interesting article, and certainly worth a read.
Te last paragraph quotes John Rust, director of the Psychometrics Research Center at Cambridge, as criticizing hiring tests for being too transparent and easy to game. That may be true of personality tests in general, but it doesn't apply to our particular test of choice, in which there are no right or wrong answers.
Moreover, I'd suggest that this failure is, at least in some circumstances, a failure of process. For instance, if a firm posts a job description that says, "Required: a hard-bitten, fast-paced, no-nonsense go-getter," and then administers a personality test (DiSC, natch) to the applicants, then of course the results will be demonstrate this bias. But this isn't (necessarily) a failure of the test mechanics or efficacy; rather, it's a failure of the administrative/hiring process.
Anyway, good article. Go read it.