Leveraging the Predictive Power of Integrity

Recent meta-analytical studies have confirmed the decades-old findings that personality and cognitive ability - among other important factors – remain some of the best predictors of work performance. Sackett et al. (2021) highlight that some of the best predictors of work performance include structured interviews, contextualised personality, integrity, and cognitive assessments amongst others. What is interesting to note in such leading empirical reviews is that authors split integrity measures and other personality assessments (Schmidt et al., 1998; Sackett et al., 2021). We often get asked why there is a need to have a standard personality assessment as well as an integrity measure in our selection/development batteries. This often stems from the notion that these measurement approaches tap into personality, and if that is the case, why have two assessments of personality as surely one can infer similar results from the use of just one robust assessment. Before unpacking this question let’s define what is often meant by the terms contextualised personality and integrity:

1. Contextualised personality refers to personality traits that are relevant to a specific job or work context (Schaffer et al., 2012). For example, conscientiousness is a personality trait that is generally associated with strong ‘on the job’ performance. However, conscientiousness may be even more important for jobs that require a high degree of attention to detail or that involve working independently. For other roles that require flexibility and adaptability, high conscientiousness may be disadvantageous.

2. Integrity can refer to a person’s honesty, trustworthiness, adherence to rules, and other guiding principles (Berry et al., 2007; Ones et al., 2012). Integrity is important in the workplace because it can help to reduce employee theft, aggression, fraud, sabotage, and other forms of counterproductive work behaviours (CWB). Integrity can help to build trust and cooperation among employees, which can also lead to improved job performance.

Both contextualised personality and integrity are important in workplace performance prediction but for different reasons. Contextualised personality as mentioned can help to predict how well a person will perform doing daily tasks inherent within a specific role, whereas integrity and risk assessments focus on the prediction of counterproductive behaviours. Integrity assessments emerged in the United States as a solution to detect potential deceitful employment candidates following the enactment of laws that forbade the usage of polygraph examinations (Ones et al., 2012). Over time, these integrity evaluations have progressed significantly, expanding beyond their initial role as alternatives to lie detector tests in order to help predict how likely a person is to engage in CWB’s and workplace deviance. Such measurement also began to focus on the psychological underpinnings of counterproductive work behaviours. 

Assessing such predictors of performance can be done through using any number of tools available. There is a plethora of options when it comes to contextualised personality which normally draw on the Five Factor Model - FFM (Costa & McCrae, 1992). Also popularly referred to as the Big 5, the FFM is a well-validated model of personality that assesses five broad dimensions of personality: extraversion, agreeableness, conscientiousness, neuroticism, and openness to experience. Integrity assessment, however, is not as well established and there is a growing need to screen out individuals that have a greater propensity for risk-type behaviours and/or to highlight development areas that can derail a person’s potential.  As such the Work-Related Risk and Integrity Scale (WRISc) was developed to heed the call of the Human Science and Research Council (HSRC) that highlighted the need for integrity assessments (Van Zyl, 2016). More information about the WRISc can be found here.

The Problem: Is it not all personality?

At this point we return to our previous question, are not personality and integrity essentially the same i.e., that integrity is merely a higher order factor of personality over and above the FFM (Berry et al., 2007; Ones et al., 1994), or even a single or narrower facet from various five-factor domains (Marcus et al., 2007; Schneider et al., 1996)? Well, the short answer is, integrity is complex, and although there may be various approaches such as lie detection through physiological responses in a polygraph test, the way integrity is measured in such assessments as the WRISc, is personality based. As mentioned, we often get asked, if the WRISc is a personality measure, then why use other personality measures in a selection battery when one assessment of personality will suffice.

Bright side and dark side approaches to understanding personality

Various personality traits can often be subsumed under even broader conceptualisations or meta-traits, for example Stability and Plasticity (De Young, 2015). There are often sound theoretical reasons for such an approach during research but there is also sometimes a need for more precise measurement of traits and their associated facets in selection contexts - facets can be thought of as subcategories of a higher order factor e.g., Gregariousness is a facet of the higher order factor of Extraversion. The ‘Bright side’ and ‘Dark side’ is another way of parcelling out the complexity of personality measurement (Hogan, 2009). ‘Bright side’ personality measures generally focus on typical everyday traits that individuals exhibit when they are at their best. They often draw from models such as the FFM mentioned previously which measure traits like Openness, Conscientiousness, Extraversion, Agreeableness, and Emotional Stability. The WRISc can be used to assess ‘dark side’ personality traits that are not explicitly captured by the FFM, and which refers to dysfunctional or maladaptive aspects of personality, often not immediately apparent or frequently displayed by an individual. These can include aggression, Machiavellianism, and narcissism (Van Zyl, 2016). As such, the WRISc focuses more on the potential ‘dark side’ of personality, assessing traits related to risk and integrity. This allows organisations to identify potential challenges or risks that may not be evident in ‘bright side’ measures, providing a more balanced and comprehensive understanding of an individual’s personality.

Traditionally, personality tests have been “overt” measures, meaning they directly assess personality traits by asking individuals to rate their own behaviours or tendencies. The WRISc represents a slight departure from this traditional approach by being a more “covert” personality measure (Van Zyl, 2016). It focuses on assessing work-related risks and integrity but does so by indirectly probing personality traits associated with these qualities. This means that the respondent may not be aware of the specific traits being assessed, which can help to reduce response bias and provide a more authentic picture of the individual’s character. This covert nature makes the WRISc an invaluable tool in high stakes contexts where there is a concern that individuals may attempt to present themselves in an overly positive light or where it’s important to measure traits that people may not be aware of and may not feel comfortable disclosing.

Sophisticated techniques have been employed by the test developers of the WRISc to achieve this and are more fully expanded on in the technical manual (this is available on request for those who are trained). These techniques, such as quantile regression, model how the measurement of seemingly everyday personality traits is used to predict risk when there are higher levels of a particular trait. This allows one to examine how behaviour changes not just on average as is the case with traditional assessments, but specifically when certain traits are on the higher end of the population distribution and therefore come with more pronounced behaviours. It emphasises that while these personality traits may seem ordinary, their impact on behaviour can be significant and varied when elevated, possibly signalling risk or counterproductive potential. This offers a more nuanced understanding of how changes in a particular personality trait at higher levels affect behaviours and ultimately performance outcomes.

The importance of combining the WRISc and Big Five measures

Incorporating both the WRISc and Big Five (FFM) personality measures into a selection battery provides a more holistic and nuanced view of an individual’s personality. The Big Five assessment can highlight an individual’s strengths and typical everyday behaviours which may both enhance or inhibit performance under certain conditions, while the WRISc can illuminate potential areas of risk or concern related to integrity where such behaviours may not be immediately apparent, and which could lead to deviant behaviour. This dual approach can help organisations make more informed decisions about hiring and team composition. It can also provide a foundation for personalised training and development initiatives, helping individuals understand and manage their potential risk areas while leveraging their strengths (Barrick et al., 1991).

Two studies conducted by JVR have proved useful in driving home the utility of having both Big Five measures (bright side) and Integrity (dark side) measures of personality in a selection process. In the most recent study, which was commissioned by a client and in collaboration with another test publisher, we were able to explore the congruence between the WRISc and a second assessment used by the client. These assessments were used in their selection process, and the aim of the study was to determine if both tools were assessing similar constructs or providing distinctive insights – the second assessment claiming to assess performance inhibitors (n = 193). The findings suggested that there was some overlap as would have been expected with both tapping into personality, but the relationships were largely weak positive relationships. In discussion with the test developer, it was they who stated that this was no surprise, as their inhibitor identification was not designed to tap into dark side traits but was rather a measure of ‘normal’ functioning where strengths may be overutilised and where other inhibitors of performance may be evident based on everyday functioning e.g., high stress tolerance that may translate into a lack of urgency. Although further detail of the study is beyond the scope of this article, the takeaway is the important distinction between how normal behaviour can impact everyday work performance versus identifying dark side tendencies that heighten the risk of counterproductive behaviours.

In the second study conducted by JVR (n = 3 056), using Dominance Analysis, we considered which factors from the Basic Traits Inventory (BTI being a bright side measure) best explained the variance in the WRISc scales and to what extent. Given the conceptual commonality of both instruments measuring personality, we were essentially curious to understand (a) the level of overlap encountered and (b) any unique contributions made by using both instruments.  Expected overlap was discovered between scales and factors that are conceptually linked. For example, we expected to see a strong negative correlation between Low Effortful Control (WRISc) and factor Conscientiousness (BTI). However, we found that low Conscientiousness does not completely predict Low Effortful Control. What this means practically is that not all occupations necessarily require the same amount of Conscientiousness - think of an entrepreneur who may often work on new projects, driven by a sense of urgency as well as need for innovation and yet who may be less reliant on being orderly and highly disciplined. Looking at such a profile, low Conscientiousness scores may not be worrisome this could allow them to respond in an agile, flexible manner to new demands. However, having the WRISc profile can add a lot of additional insight, because looking at if the person scores above the 80th percentile on Low Effortful Control could then be indicative of more problematic disorganisation and/or questionable strategies to reach their goals. Another very interesting finding concerned the WRISc scale Egotism, which was most prominently explained by factor Extraversion (BTI). Higher scores on Egotism should not necessarily be negatively interpreted when under the 80th percentile. However, considering the emphasis on the potential for grandiose/narcistic behaviours in the Egotism scale e.g., excessive self-esteem and the need to be admired, coupled with higher scores on Extraversion, this could help in predicting leadership emergence in combination with other proven critical traits but could also help in mitigating risks that could derail leadership effectiveness. There are many more such examples that can be derived from the study, but it is beyond the scope of this article.

These studies highlight the benefit of examining both the ‘bright side’ and ‘dark side’ in combination, as we are then able to get a comprehensive view of an individual’s behaviour across a range of situations, making conclusions more nuanced and complete. This multifaceted approach not only enhances the predictive validity of assessments but also ensures fairness by considering a more comprehensive spectrum of an individual’s personality, rather than just focusing on the aspects that are readily observable or more easily measured.

Summary

In summary, understanding both personality and integrity is crucial for predicting workplace performance. Traditional personality assessments that generally report on the ‘bright side’, capture an individual’s typical behaviours and strengths, while integrity assessments like the WRISc provide insights into the ‘dark side’ and report on potential risk behaviours and counterproductive tendencies. The WRISc offers a unique, covert perspective on personality that complements traditional ‘bright side’ personality measures. Incorporating both types of measures into a selection or development process is essential as it provides a more comprehensive and balanced assessment of an individual, ultimately contributing to more effective hiring and development practices.

Reference List

Barrick, M.R. & Mount, M.K. (1991). The big five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44, 1-26. https://doi.org/10.1111/j.1744-6570.1991.tb00688.

 Berry, C. M., Sackett, P. R., & Wiemann, S. (2007). A review of recent developments in integrity test research. Personnel Psychology, 60(2), 271-301. https://doi.org/10.1111/j.1744- 6570.2007.00074.

 Costa, P. T., & McCrae, R. R. (1992). The five-factor model of personality and its relevance to personality disorders. Journal of Personality Disorders, 6(4), 343–359. https://doi.org/10.1521/pedi.1992.6.4.343.

DeYoung, C. G. (2015). Cybernetic Big Five Theory. Journal of Research in Personality, 56, 33–58. https://doi.org/10.1016/j.jrp.2014.07.004.

Hogan, R. (2009). Hogan Development Survey Manual.  Hogan Assessment Systems.

 Marcus, B., Lee, K., & Ashton, M. C. (2007). Personality dimensions explaining relationships between integrity tests and counterproductive behaviour: Big five, or one in addition? Personnel Psychology, 60(1), 1-34. https://doi.org/10.1111/j.1744-6570.2007.00063.

 Ones, D. S., Schmidt, F. L., & Viswesvaran, C. (1994). Do broader personality variables predict job performance with higher validity? [Conference presentation]. 1994 Conference of the Society for Industrial and Organisational Psychology, Nashville, TN, U.S.A.

 Ones, D. S., Viswesvaran, C., & Schmidt, F. L. (2012). Integrity tests predict counterproductive work behaviours and job performance well. A comment on Van Idekkinge, Roth, Raymark and Odle-Dousseau. Journal of Applied Psychology, 97(3), 537-542.

 Sackett, P. R., Zhang, C., Berry, C. M., & Lievens, F. (2021). 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

Shaffer, J. A., & Postlethwaite, B. E. (2012). A matter of context: A meta analytic investigation of the relative validity of contextualized and noncontextualized personality measures. Personnel Psychology, 65(3), 445–494. https://doi.org/10.1111/j.1744-6570.2012.01250.

Schneider, R. J., Hough, L. M., & Dunnette, M. D. (1996). Broadsided by broad traits: How to sink science in five dimensions or less. Journal of Organisational Behavior, 17(6), 639–655. https://doi.org/10.1002/(SICI)1099-1379(199611)17:63.0.CO;2-9.

Van Zyl, J.J. (2016). The Development and validation of a model of work-related integrity. [Unpublished doctoral dissertation]. University of Johannesburg. https://uicontent.ui.ac.za/vital/access/manager/lndex?site name-Research%200utput.