Validity and Utility Of Psychometric Assessments
A concern often raised is whether or not the assessments used are indeed helpful in discriminating against potentially high and low performing candidates. This has been researched as far back as 1913. In 1998, a meta-analytic review was conducted of over 85 years of research to evaluate how well intelligence, personality, and integrity tests can predict job performance. They found that the employees’ general mental ability can predict up to 40% of the difference between job performance of candidates (Schmidt & Hunter, 1998). Similarly, personality and integrity tests added to the effectiveness of predicting job performance. For context, the relationship between these tests and job performance are compared to well established relationships (smoking and lung cancer) in the table below.
Relationship | Correlation (r) |
Smoking and lung cancer within 25 years | .08 |
Antihistamines and reduced snot and sneezing | .11 |
Viagra and improved sexual functioning | .38 |
Conscientiousness and Job performance | .18 |
General mental ability and job performance | .21 |
Integrity tests and job performance | .21 |
Note. From “In defense of personality measurement” by R, Hogan, 2005, Human Performance, 18(4), p.338.
However, despite the consistency in correlation between psychometric assessments and overall job performance, some have argued that these tests can be faked, especially personality tests. However, this is very unlikely. Firstly, it is difficult to fake a whole profile. You would need to consistently fake your score on numerous questions. Secondly, these questions do not always give a clear indication of what they measure, the candidate would therefore not know how to best fake the test. Thirdly, it is not always possible to know what profile the recruiter is looking for. Finally, sophisticated techniques exist for detecting faking. These include statistical analysis, and the introduction of socially desirable responding scales.
The sceptical reader may wonder what the actual impact of psychometric assessment is with regards to the company’s bottom line. Moerdyk (2015) mentions that the difference in productivity of poor and well performing workers is about double. In other words, outputs are twice as much. In marketing and sales, well performing salespeople can even be more than twice as productive. The effects of selecting the wrong employee for a leadership position can be even more devastating. In South Africa, a lack of integrity and poor leadership skills have led to poor service delivery, corruption, and deterioration of public institutions. However, the utility of selection methods can also have a high return on investments for blue collar workers as well. Smith and Smith (2005), gives and excellent example where a hotel chain gained £5 565 a year in work value by utilising a valid selection method for their painters and decorators. While the data required to calculate the utility of psychometric tests is difficult to come by, it is still possible to calculate the utility of selection methods for your individual business after a certain period of time has elapsed (Smith & Smith. 2005).
Why use the skills of a registered Industrial Psychologist?
The Employment Equity Act No. 55 of 1998 give strict guidelines with regards to employee selection. Firstly, when selecting employees, the onus is on the employer to ensure that the selection methods are fair. In other words, it may not discriminate against anyone on the basis of race, gender, sex, pregnancy, marital status, or any other arbitrary grounds. Discrimination is only allowed on the basis of the inherent requirements of the job. Herein lies many potential pitfalls for untrained or uninformed recruiters.
When evaluating an employee with the use of psychometric constructs, it must have been proven by previous research that these construct are related to job performance. For example not hiring a candidate because they may be more disagreeable than another potential candidate constitutes illegal discrimination if you cannot show that this trait is related to effective job performance.
Employee Equity Act
Secondly, while many psychometric tests are available, they are not all equally valid across population groups. In section 8 of the Employee Equity Act No. 55 of 1998, it is explicitly stated that psychological tests may not be administered unless it:
- has been scientifically shown to be valid and reliable
- can be applied fairly to all employees
- is not biased against any employee or group
It is difficult to understand the validity and reliability of psychometric assessments for anyone outside the field of psychometrics. Indeed research into the validity and reliability of tests is a speciality in its own right. For example, while the 16 PF and the BTI are both personality assessments administered in South Africa, it may be beneficial in some circumstances to rather use the 16 PF over the BTI and vice versa. The rationale for the use of psychometric assessments can be quite complex. Finally, administering psychometric assessments constitutes a psychological act and may therefore only be legally administered by psychologists and psychometrists registered with the HPCSA.
In summary, psychometric assessments have many potential benefits for employee selection and job performance. However, if not done carefully and skilfully, may lead to uninformed selection procedures that are at best ineffective, illegal, and at worst may cause harm to the candidate. It is always advisable to use the expertise of trained industrial psychologists so maximise the benefits and minimise the risks.
References
Employment Equity Act, No 55 of 1998. Available at: https://www.labour.gov.za/DocumentCenter/Acts/Employment%20Equity/Act%20-%20Employment%20Equity%201998.pdf
Hogan, R. (2005). In defense of personality measurement: New wine for old whiners. Human Performance, 18(4), 331–341. https://doi.org/10.1207/s15327043hup1804_1
Moerdyk, A. (2015). The Principles and Practice of Psychological Assessment (2nd Ed). Van Schaik Publishers.
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
Smith, M., & Smith, P. (2005). Testing people at work: Competencies in psychometric testing. Blackwell Publishing.
Whether you’re an individual seeking career guidance, a team aiming for optimal performance, or an organisation striving for growth, Ernestco is your trusted partner. We offer a comprehensive range of services designed to address the unique challenges you face.
Contact us today to explore how our expertise can empower you to achieve your goals. Let’s work together to unlock your full potential.
