PS390 Unit 2 Personnel Selection Paper Example

The Interview That Predicted Almost Nothing: A PS390 Unit 2 Personnel Selection Paper

[Student Name]

Purdue University Global

PS390: Introduction to Industrial/Organizational Psychology

Unit 2 Personnel Selection Paper

[Instructor Name]

[Date]

The writer, the call center, the applicants and every figure are composite and illustrative; the articles cited are real.

The Hiring Problem

Our Charlotte customer service center hires about ninety representatives a year to answer billing and outage calls for a regional electric utility. Last year, forty-six of them left within ninety days, most during or soon after the six-week training class. Each early departure cost the center the recruiter's time, a trainer's seat and weeks of a supervisor's coaching.

Our selection process was a résumé screen and a thirty-minute interview with whichever supervisor was free. Supervisors asked whatever they liked, often about hobbies or previous jobs, and rated candidates on a single overall impression. When I compared interview ratings with which hires stayed past ninety days and with their first-quarter quality scores, the correlation was close to zero.

Two Summaries of the Evidence

Schmidt and Hunter (1998) pooled eighty-five years of studies on which hiring tools forecast later job performance. In their tables, mental ability tests, job tryouts and structured interviews led the field, the free-form interview trailed far behind, and adding either a structured interview or an honesty test to a mental ability test improved the forecast beyond the test alone.

Sackett et al. (2022) went back over those numbers and concluded that the statistical adjustments behind them had inflated many of the figures. Nearly every number came down, and the order changed: structured interviews now sat at or near the top, and cognitive tests dropped well below where earlier work had placed them. Unstructured interviews remained near the bottom.

Our interview was not just weak; it was the kind of method the research ranks lowest in both summaries.

What Structure Means

Campion et al. (1997) reviewed the components that make an interview structured. Some of the features they described concern what gets asked, such as questions drawn from the duties of the job, identical questions for everyone and questions about past or hypothetical situations; others concern how answers get judged, such as scoring each reply against described examples, pairing interviewers and preparing them. More structure generally meant more reliable and valid interviews.

Designing a Better Process

I began with a short job analysis. With three high-performing representatives and two supervisors, I listed what the job requires: handling upset callers calmly, following billing rules accurately, typing notes while listening, and working fixed shifts including weekends. Then I designed three steps.

First, a structured interview of six questions, the same for every candidate, each tied to a job requirement. Two are behavioral ("Tell me about a time you calmed down an upset customer"), and four are situational ("A caller says her power was cut off by mistake and she has a baby at home. What do you do first?"). Interviewers score each reply from one to five, guided by written samples of weak, middling and excellent replies. Two interviewers rate independently.

Second, a fifteen-minute work sample: a recorded simulated call in which the candidate looks up a billing question in a mock system and explains it to an actor playing a frustrated customer, scored on a checklist.

Third, a realistic job preview at the start, a short video of the floor and the schedule, so candidates who cannot work weekends or dislike constant calls can withdraw early.

Why Supervisors Liked the Old Interview

The supervisors were not careless. They liked the unstructured interview because it felt natural, let them judge whether they would enjoy working with someone and gave them a sense of control over who joined their teams. Several told me they could "just tell" who would last. The data said otherwise, but telling them that would not have changed their minds. What helped was showing them the list of last year's early leavers alongside the ratings each supervisor had given: several of the highest-rated candidates were gone within a month. I also kept one thing they valued. The structured interview ends with five minutes for the candidate's questions, which supervisors do not score, so they still get a sense of the person without letting it drive the decision. Winning their support mattered, because a structured interview delivered by reluctant interviewers soon drifts back into conversation.

The Résumé Screen

The résumé screen needed attention too. Recruiters had been favoring candidates with previous call center experience, but our data showed no difference in ninety-day retention between experienced and inexperienced hires. Requiring experience was narrowing the pool without improving it. The new screen checks only for the schedule availability and basic computer skills the job analysis identified.

Fairness and Adverse Impact

A selection process must be fair as well as predictive. A hiring step shows adverse impact when people from a protected group are turned away at clearly higher rates; federal guidance treats a pass rate below four-fifths of the top group's pass rate as a signal to look closer. I chose not to add a general cognitive ability test, partly because such tests often show larger group differences than structured interviews and work samples, and partly because Sackett and colleagues' revised estimates gave less reason to expect large gains from one. I will track selection rates by group every quarter and report them to the center manager alongside retention.

Training the Interviewers

Structure fails if interviewers ignore it. All supervisors who interview will complete a two-hour session covering the six questions, how to score them and the habit of letting the first minute decide everything. They will practice on recorded mock interviews until their ratings agree closely.

Evaluating the New Process

The new process starts in January. I will compare ninety-day retention and first-quarter quality scores for hires selected with it against last year's hires, and check whether interview and work sample scores predict those outcomes. If one step adds nothing, it will be revised or dropped. I will also ask new hires, at the end of their training class, whether the realistic preview matched what they found on the floor, since a preview that oversells the job defeats its purpose and may even raise early turnover.

Conclusion

An unstructured interview chose last year's hires and predicted almost nothing about who stayed. Schmidt and Hunter's summary and Sackett and colleagues' revision agree that unstructured interviews are among the weakest methods, and Campion and colleagues show what structure requires. A structured interview, a work sample and a realistic preview, checked for adverse impact and evaluated against retention, give the center a better chance of hiring people who stay.

References

Campion, M. A., Palmer, D. K., & Campion, J. E. (1997). A review of structure in the selection interview. Personnel Psychology, 50(3), 655-702. https://doi.org/10.1111/j.1744-6570.1997.tb00709.x

Sackett, P. R., Zhang, C., Berry, C. M., & Lievens, F. (2022). Revisiting meta-analytic estimates of validity in personnel selection: Addressing systematic overcorrection for restriction of range. Journal of Applied Psychology, 107(11), 2040-2068. https://doi.org/10.1037/apl0000994

Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124(2), 262-274. https://doi.org/10.1037/0033-2909.124.2.262