Saturday, January 20, 2007
New Journal of Applied Psychology (v92, #1)
There's a new issue of the Journal of Applied Psychology (one of the most read research-focused journals related to recruitment/assessment) out, so let's see what it's got! Caution: this issue is a whirlwind smorgasborg, so there's a lot to look at.
First up, an study of conditional reasoning tests (CRTs) and how "fakeable" they are. Don't know what a conditional reasoning test is? Here's an example. Anyway, this study found that CRTs were fakeable when the purpose of the assessment was given. On the plus side, no differences were found between student, incumbent, and applicant samples. Will we ever figure out the faking problem? That's one of the big questions in assessment right now.
Next, a meta-analysis that looked at the unique contribution self-efficacy measures might make to the prediction of job performance after accounting for personality (Big 5), general mental ability, and job/task experience. Results? Not a big contribution after accounting for the other factors. Self-efficacy did predict performance in low-complexity jobs, but not medium- or high-complexity jobs. It also predicted task performance, but not job performance.
On a related note, the third article relevant for us reviewed the research on goal orientation (GO) and whether it predicts performance. (GO is essentially how you view performance--are you focused on learning or on completing, or not completing, the task? More details here) The authors of the current study found that having a learning orientation or avoiding performance predicted performance (in opposite ways) and state GO tended to out-predict trait GO. (State being related to the task at hand, trait being a persistent aspect of someone) Even more interesting, state GO added predictive power above and beyond cognitive ability and personality. I may have to get this one.
The next article is, I must confess, one of those that you really need to read in order to fully understand. And by "you", I mean me, because I can't get the whole gist from just the abstract. Essentially what the researchers found is that bias against Black men can be reduced through a structured free recall intervention. Donations to the "get Bryan more subscriptions to research journals" are accepted with gratitude.
Speaking of articles I need to actually read, the next study looks at how we might group college-bound individuals beyond simply looking at standardized test scores. After studying over 2,700 entering college students, the researchers found that individuals could be grouped into one of five categories based on factors such as biographical data and scores on situational judgment tests. To be honest, the press release does a better job of describing this than I can.
Next one up is a doozie. Dr. Collins, who has done other work in the area of employer knowledge and applicant behavior, found that when applicants aren't very knowledgeable about an employer's products or services, less information (e.g., a banner ad, a sponsorship) is better in terms of generating interest. On the flip side, when products and services are well known, more information is required to attract applicants (e.g., brochures, employee endorsements). I'm thinking an employee blog would be an example of the latter, as well? Good stuff, with a more readable summary here.
Next, an examination of what predicts workplace aggression. This meta-analysis of 57 research studies found that both individual and situational factors predict aggression and the pattern is specific to the target of said aggression. A must for those interested in predicting workplace violence, I'm thinking. A previous article by some of the same authors is available here.
Last but definitely not least is a study of the impact that emotions (affect) have on perceptions of organizational justice. As we know, perceptions of organizational justice (e.g., procedural justice) are strongly correlated with important behaviors like likelihood of applying for a job, referring the employer to others, and filing lawsuits. So what does this study have to tell us? That both state and trait affect influence justice perceptions, with state effects (get it? affect, effect?) generally slightly stronger.
Why is this last study so important? Because people generally don't file lawsuits because they believe Title VII of the Civil Rights Act of 1964, as amended in 1991, has been violated and the employer will fail to show that the selection process was job-related and consistent with business necessity. No, they are honked off because they were treated poorly or the process seemed squirrelly. They talk to other people who feel the same way, and BAM!, class action lawsuit. So take your time when coming up with your assessment process, and treat your applicants with respect, if not because they're human beings, then because you're likely saving yourself thousands of dollars and countless hours defending a lawsuit.
Whew! What an issue. Shelley Zedeck (the editor) must have had a busy last few months!