The Science of Selecting a Hire

The Science of Selecting a Hire

You have posted an advertisement for a job and now have a pile of applicants. You know your business better than anyone, so you figure you will just interview applicants and pick one. Unfortunately, unstructured interviews are (scientifically) nearly the worst possible approach to pick an applicant.1 Interviewers are notoriously inconsistent. They also provide different ratings than objective tests2 and are influenced by non-performance-related factors.3 Also, the applicants themselves are notoriously poor at estimating their own abilities even when they are not motivated to lie: self-estimates of abilities explain only 11% of the variance in measured abilities.4 Relying on interviews alone is a high-risk approach to hiring.

Need to Hire?  Post Jobs for Free

Worried about the subjectivity of unstructured interviews, some have started using measures of personality. Hiring someone who is not neurotic and is conscientious sounds like a great idea. However, candidates successfully fake these measures5 and are most likely to “fake good” on measures of conscientiousness6 often used in candidate selection. Furthermore, the relationships of personality and those with the highest levels of “conscientiousness” and “emotional stability” ultimately get lower job performance evaluations from their supervisors.7 Faking appears pervasive. In one study 50% of applicants successfully faked their performance on these questionnaires, and this faking was associated with poorer job performance.8

Industrial-Organizational Psychology is a scientific field that studies hiring practices. Science has some specific ideas for how to improve your candidate selection. The main difference in these approaches is just how much time you have to improve your candidate selection process.

This is necessarily a limited review. For example, the fit of an applicant to an organization is thought to be more important to turnover, while the fit of the applicant to the job is thought to be more important to job performance. Different selection procedures might be more important for one than the other. Any of these topics individually warrants more coverage than it is going to get here.

Referral hiring

Hires from referrals are far less likely to quit, have better job performance, and are about 25% more profitable than non-referred workers.9 These improvements were evident for hires in call centers, trucking, and high-tech sectors. Referrals from current workers with the highest cognitive ability were likely to also have the higher abilities and job performance.10 On the other side, workers with the highest education are the least likely to use network contacts when looking for work,11 which suggests how desirable a referral is might vary by hiring sector.

The risk in referral hires is that they also tend to share racial, age, and gender characteristics of the referrer,12 which can introduce undesirable hiring biases. However, it has the desirable effects of reducing income differences in black and Hispanic hires, because referred hires have less income differentiation by ethnicity.13

Cognitive tests

Goertz_plot
Sample interpretation: If you are hiring for a clerk position, do not use perceptual speed or numerical tests to select your hire; Reasoning tests probably will not help; Verbal and Memory tests should be used.

General intelligence has been described as the “single most useful worker attribute for predicting job performance” (p. 380).14,15 Cognitive testing used to refer to general intelligence tests. These are long-recognized as the best selection criteria for employment, but many of the tests suffer from some bias that may violate employment law.16 Cognitive ability could be gathered from an applicant (e.g., GRE scores) or assessed as a part of the selection process.

Specific cognitive abilities have long been demoted as not adding any prediction to job satisfaction or performance beyond general intelligence. However, there are exceptions and some reason for resurgence. Perceptual speed and spatial abilities are demonstrated to be important specific cognitive abilities for some jobs, but some are advocating including memory tests. For example, context memory predicts performance in job training well,17 and verbal memory is less affected by age to avoid adversely impacting older applicants.18 A meta-analysis of studies on specific cognitive abilities has highlighted which abilities appear to be the best predictors for different types of jobs (see Figure).19

Job sampling (aka performance assessment, aka work sample)

The goal of this assessment is to collect and evaluate work from the applicant that closely resembles real work tasks and evaluate it as a work sample. The burden of developing, scoring, and testing this assessment falls to the organization. The cost of making a good job samples measure might be why they are not used often, despite their strengths. Specifically, job samples tend to predict job performance well (r = .39).20 Computer-simulated job samples also predict job performance beyond relevant work history and personality factors.21 Job samples are often found to have less or no bias for categories protected by employment law,16 with some exceptions.22

Social media selection

If you are considering being clever by getting information the applicant did not give you through Facebook or Twitter, hold off. Those data do not give any information above traditional methods, are unrelated to job performance and turnover, and introduce gender bias in selection.23 This may be due to screening of irrelevant information, such as the use of “text-speak” in spelling on a post.24 When companies have chosen to screen social media secretly, the search is superficial (not beyond face page) and decreases the likelihood of inviting the applicant for an interview.25 Add the potential liability as privacy laws evolve around these issues, and social media screening appears to be a bad hiring move for many reasons.

SUMMARY

You are more likely to acquire a strong performer who will stay by hiring based on their general intelligence, cognitive abilities specific to your profession, and performance of a work sample. Referrals help identify people likely to fit meet these criteria. Screening candidate’s social media would lead you to miss qualified candidates. Unstructured, in-person interviews and personality tests rarely improve candidate selection.

  1. Schmidt, F. L. & Hunter, J. E. The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychol. Bull. 124, 262–274 (1998).
  2. Kuncel, N. R., Klieger, D. M., Connelly, B. S. & Ones, D. S. Mechanical versus clinical data combination in selection and admissions decisions: a meta-analysis. J. Appl. Psychol. 98, 1060–72 (2013).
  3. Roulin, N., Bangerter, A. & Yerly, E. The uniqueness effect in selection interviews. Journal of Personnel Psychology 10, 43–47 (2011).
  4. Freund, P. A. & Kasten, N. How smart do you think you are? A meta-analysis on the validity of self-estimates of cognitive ability. Psychol. Bull. 138, 296–321 (2012).
  5. Jerneic, Ž. Questionnaire Responses and How Successful are Their Attempts?? A Case of Military Pilot. 20, (2012).
  6. Vecchione, M., Alessandri, G. & Barbaranelli, C. The Five Factor Model in personnel selection: Measurement equivalence between applicant and non-applicant groups. Pers. Individ. Dif. 52, 503–508 (2012).
  7. Le, H. et al. Too much of a good thing: Curvilinear relationships between personality traits and job performance. Journal of Applied Psychology 96, 113–133 (2011).
  8. Donovan, J., Dwight, S. & Schneider, D. The Impact of Applicant Faking on Selection Measures, Hiring Decisions, and Employee Performance. J. Bus. Psychol. 29, 479–493 (2014).
  9. Burks, S., Cowgill, B., Hoffman, M. & Housman, M. You’d Be Perfect for This”: Understanding the value of hiring through referrals. Inst. Study Labor Discuss. Pap. (2013).
  10. Beaman, L. & Magruder, J. Who gets the job referral? Evidence from a social networks experiment. Am. Econ. Rev. 102, 3574–3593 (2012).
  11. Ioannides, Y. M. & Loury, L. D. Job Information Networks, Neighborhood Effects, and Inequality. J. Econ. Lit. 42, 1056–1093 (2004).
  12. Stainback, K. Social Contacts and Race/Ethnic Job Matching. Soc. Forces 87 , 857–886 (2008).
  13. Burks, S. et al. The Facts About Referrals?: Toward an Understanding of Employee Referral Networks ?. (2013).
  14. Gottfredson, L. S. Societal consequences of the g factor in employment. J. Vocat. Behav. 29, 379–410 (1986).
  15. Ree, M. J., Earles, J. a. & Teachout, M. S. Predicting job performance: Not much more than g. J. Appl. Psychol. 79, 518–524 (1994).
  16. Sackett, P. R., Schmitt, N., Ellingson, J. E. & Kabin, M. B. High-stakes testing in employment, credentialing, and higher education. Prospects in a post-affirmative-action world. Am. Psychol. 56, 302–318 (2001).
  17. Cucina, J. M., Su, C., Busciglio, H. H. & Peyton, S. T. Something more than g: Meaningful Memory uniquely predicts training performance. Intelligence 49, 192–206 (2015).
  18. Klein, R. M. et al. Journal of Applied Psychology Among Business Executives Cognitive Predictors and Age-Based Adverse Impact Among Business Executives. (2015).
  19. Goertz, W., Hülsheger, U. R. & Maier, G. W. The Validity of Specific Cognitive Abilities for the Prediction of Training Success in Germany. J. Pers. Psychol. 13, 123–133 (2014).
  20. Roth, P. L., Bobko, P. & Mcfarland, L. a. A meta-analysis of work sample test validity: Updating and integrating some classic literature. Pers. Psychol. 58, 1009–1037 (2005).
  21. Fluckinger, C. D., Dudley, N. M. & Seeds, M. Incremental Validity of Interactive Multimedia Simulations in Two Organizations. Int. J. Sel. Assess. 22, 108–112 (2014).
  22. Bobko, P., Roth, P. L. & Buster, M. A. Work Sample Selection Tests and Expected Reduction in Adverse Impact: A Cautionary Note. Int. J. Sel. Assess. 13, 1–10 (2005).
  23. Van Iddekinge, C. H., Lanivich, S. E., Roth, P. L. & Junco, E. Social Media for Selection? Validity and Adverse Impact Potential of a Facebook-Based Assessment. J. Manag. (2013). doi:10.1177/0149206313515524
  24. Scott, G. G., Sinclair, J., Short, E. & Bruce, G. It’s Not What You Say, It’s How You Say It: Language Use on Facebook Impacts Employability But Not Attractiveness. Cyberpsychol. Behav. Soc. Netw. 17, 562–566 (2014).
  25. Manant, M., Pajak, S. & Soulié, N. Online social networks and hiring?: a … eld experiment on the French labor market. 1–57 (2014).

 

Written by

More Articles by Nicole Prause, PhD