Huffcutt, A. I., Conway, J. M., Roth, P. L., & Stone, N. J. (2001). Identification and meta-analytic assessment of psychological constructs measured in employment interviews. Journal of Applied Psychology, 86, 897-913.
Hurtz, G. M., & Donovan, J. J. (2000). Personality and job performance: The big five revisited. Journal of Applied Psychology, 85, 869-879.
Raymark, P. H., Schmit, M. J., & Guion, R. M. (1997). Identifying potentially useful personality constructs for employee selection.Personnel Psychology, 50, 723-736.
**Arthur, Jr., W., Woehr, D. J., & Graziano, W. G. (2001). Personality testing in employment settings: Problems and issues in the application of typical selection practices. Personnel Review, 30, 657-676.
There is a distinction that must be made between constructs and methods. This construct/method distinction (aka the content/method distinction) is a touchy subject for many I/O psychologists. We briefly touched on this in earlier modules (see PRV – Module 4, with the discussion on Schmidt vs. Landy) but now I’ll elaborate some more.
Constructs are properties that we posit to exist in the world. They guide our definitions for the purpose of developing measurements. Constructs are ideas in our own minds, not (as some writers imply), characteristics that exist in the real world. When we develop measures, they are tapping events in the world that we use to alter our mental constructs. You can think of this as surveyors measuring the land, then bringing information back to make maps (constructs).
We can measure these constructs using various methods. As you’ll see in the Huffcutt, Conway, Roth, and Stone (2001), one methodology (interviews) can actually tap many different constructs depending on the format of the method (e.g., structured vs. unstructured). Similarly, many different methods can tap the same construct. For example, cognitive ability can be assessed through paper and pencil measures, interviews, and assessment centers.
Why is this construct/method distinction important? Basically, it does not make sense to compare and contrast a methodology with a construct. For example, to assert that interviews have greater predictive validity than does personality is illogical when we know that interviews can be used to assess personality. An interview that assesses personality can have greater predictive validity over paper and pencil measure of personality, or interviews that assess cognitive ability could be more predictive of future performance than interviews that assess personality, but to compare a method with a construct is inappropriate. This happens a lot, however, so it is up to you to know when this error is made.
Anything that we can imagine would possibly be important or predictive of job performance that we cannot actually measure can constitute a construct. Raymark, Schmit, and Guion (1997) provide examples of constructs that are important for selection, placement, and classification. The Hurtz and Donovan (2000) article focuses on one of the most widely studied construct (other than, perhaps, cognitive ability) in the selection domain: personality.
Important terms from the readings:
predictor construct, personality, performance, results, effectiveness, conscientiousness, extraversion, openness, neuroticism, agreeableness, NEO, motivation, structured interview, unstructured interview
Please answer three of the following questions. The first question is required. Each question is worth 5 points.
1.Huffcutt, Conway, Roth, & Stone (2001) showed that High- and Low-structured interviews tend to focus on different constructs. What constitutes a high-structured interview? What constitutes a low-structured interview? What constructs are associated with each, and why might there be differences in the constructs tapped depending on structure?
2.What constitutes the “Big Five”? Describe each of these main personality constructs. Are they valid predictors of job performance? Are some better than others? Explain.
3.What is personality? How is it measured? Is it stable? (Hint: this question is best approached if you read the optional reading).
4.Why do researchers and practitioners care about using different constructs in selection practices? Also, if cognitive ability has been shown to be such a good predictor of performance, why is there such a push to tap such constructs as integrity, conscientiousness, and work ethic?