From the Wall Street Journal, a well-written and interesting article on the human side of cheating on personality tests. It does make you wonder whether the personality test is becoming either 1) a test of intelligence or 2) the motivation to cheat. Nevertheless, the scientific evidence does indicate such tests are still 1) useful 2) cost effective 3) valid, although probably not as valid as though and 4) seen in a positive light by most candidates.
And another WSJ article on that tough interview question about “what is your greatest weakness?” Mine is that I am so good looking people don’t take me seriously.
{ 3 comments… read them below or add one }
The line in the article I found most intriguing was this: “Turnover at Finish Line declined after the chain adopted the test in 2003, saving about $1 million a year, according to marketing material for the test. Finish Line had no comment.”
Now, I know there is good solid evidence for the incremental validity of personality indicators/measures in predicting job performance, so I certainly won’t dismiss the claim as spurious. But here is what I wonder: When claims like this are made, is this based on longer tenure-per-employee associated with *individual* scores on such batteries, or is it based on overall turnover/tenure rate since the adoption of such testing? Makes a difference.
Adopting a test is no different than any other historical event on the landscape. So, a selection test can be adopted at a time when turnover is bound to decrease, simply because of the job market and segment of the labour pool one seeks out. Steps need to be taken to assure that the test itself, as a selection tool, is reponsible for observed changes, and not some overall change ion the landscape that applies to multiple organizations. One needs to be able to assert that prediction of turnover can be done on an individual basis in order to be able to provide that assurance.
That’s certainly not a criticism of the test, the test-provider, or the client organizations. Rather, it is a caveat to the casual reader of such stories.
I disagree with Unicru that this isn’t an issue. Criterion-related validity may be one thing (and there’s a lot of research to back them up), but the impact of articles like this and the general perception that the tests are “beatable” affects applicant perceptions of the organization, willingness to apply, etc.
This may not be as much of an issue for hourly positions (particularly in this economy), but it becomes more of a problem for higher-level positions, where the competition for talent becomes fierce.
Oh, and the “weakness” question is one of my most hated, although your answer was possibly the best I’ve heard!
Thanks Bryan and Mark for the comments. This is probably worthy of a whole blog, but I was especially struck by Bryan’s comments. I would argue it is not just articles like this. Starting in grade school, I believe teachers deliver to their students the message that tests are beatable and bad; that the testing enterprise is a corrupt one with no real benefit. In Intro to Psychology classes, one hears more about the problems with testing than about the benefits. As an assessment profession, we have probably failed to protect our professional identity and our products.