Thursday, September 27, 2007

Autumn 2007 Personnel Psychology

The Autumn 2007 issue of Personnel Psychology is out with plenty for us to sink our teeth into, particularly for you personality testing fans out there. Let's take a look:

First up, Luthans et al. present the results of a study that focuses on positive psychology, which is gaining more and more interest these days. The authors describe support for a survey instrument that purports to measure four aspects of "positive psychological capital"--hope, resilience, optimism, and efficacy--and then looked at whether results predicted job performance and satisfaction. Results? A "significant positive relationship", with the composite of the four aspects outperforming each individually. (Side note: two of the authors published a book last year that focuses on this topic)

Next, Judge & Erez look at how two of the Big 5 personality dimensions--emotional stability and extraversion--predicted job performance at a health and fitness center. Not only did both predict performance on their own, but they did even better in combination. The authors suggest that the combination of emotional stability and extraversion reflects a "happy" or "buoyant" personality that may be more important to predicting performance than each trait in isolation. Great study that goes beyond the "which of the Big 5 are the best" mentality.

Next up, Buckley et al. with a study of race and interview panels. Ten White and ten Black raters viewed videotaped responses of 36 White and 36 Black police officers applying for a promotion. Results? Well, there's good news and bad news. The bad news is that there was a same race bias (i.e., White raters rated White applicants better, Black raters rated Black applicants better) and a significant difference between the panels depending on the ethnic makeup. Good news? The effect size was small and "net reconciliation" (the difference between initial and final scores) was significant (but small) only among Black raters.

Pay attention to the next study, recruiters: Zhao et al. present the results of a meta-analysis on the impact of psychological contract breach on 8 work-related outcomes, including attitude and individual effectiveness. An example of contract breach: telling an applicant you have great work-life balance policies and then never approving leave. So what did they find? Breach was related to all eight outcomes except for actual turnover. Affect mediated this relationship, which suggests to me that if you have to break a contract, you may be able to somewhat manage the impact by being smart about how you present it and being sensitive about the reaction.

Next up, a bevy of big names in the field (let's just call them "Morgeson, et al.") drop a bombshell on personality testing: they argue that because of the low validities associated with self-report personality measures, they should be discontinued for personnel selection! They don't write personality tests off completely, but suggest that alternatives to self-report measures need to be developed (someone may want to tell Judge & Erez; see article above). What might this look like? Conditional reasoning tests are mentioned as a possibility. And, (this is just me talkin') "ability" type measures could be developed (e.g., if you're conscientious you should be able to demonstrate certain behaviors) or we could integrate personality measurement into the reference checking process (hey, I didn't say it would be easy). Oh and hey, here's the article if you're interested; thanks to Dr. Morgeson for making so much of his work available.

Ironically (or is it coincidentally? curse you, Alanis Morissette), the very next article is about the development of a new self-report personality measure, the Five Factor Model Questionnaire. Gill & Hodgkinson criticize existing measures (e.g., they contain too many generic items, they use culture-specific language) and find support for their measure using five separate diverse samples, including close convergent and divergent validity with the NEO PI-R.

So that's the end of the research articles, but not the end of this journal issue. It also contains reviews of several books, including:

- Using individual assessments in the workplace: A practical guide for HR professionals, trainers, and managers by Goodstein and Prien (which looks to be a very useful introductory guide, along the lines of Aamodt et al.'s statistics book)

- Foundations of psychological testing: A practical approach (2nd ed.) by McIntire and Miller, which is designed for an undergraduate-level course.

- and for those of you looking for something a little more advanced, a review is also included of Dr. Viswanathan's Measurement error and research design.

No comments: