Research paper: Personality tests

Talent management processes within organisations are to a great extent based on understanding, predicting and changing people’s behaviour. The classic elements of talent management in an organisation are selection and development, (Tippins, 2013). Selection is all about who should be recruited, promoted, moved to a new role, sent overseas, trained, added to a team or board, or even “let go”. At each juncture decisions have to be made based on what will, ultimately, be best for the organisation in question. Bound up in the need to make judgements in selecting people, it is often important to avoid unfair discrimination. If an organisation’s judgements in selection could be based on a reliable and valid prediction of contextually relevant behaviour then its decisions would likely be high quality, and bias-free. The other key element of talent management is the development of employees; this includes skills and behaviour training, mentoring, coaching, appraising and rewarding staff. Again, if we have a good understanding of behaviour, and even how people might react to those development interventions, talent management will surely be more effective in meeting an organisation’s goals.

In order to critically evaluate the use of personality tests in talent management, this paper will firstly explore personality – what is it comprised of and how understanding it helps us understand about the way people behave. It will then look at whether personality is actually a significant determinant of job performance, and lastly it will look at how effectively tests can actually measure personality.

Hippocrates (460-370 BCE) categorised four personality types based on the proportions of blood, phlegm, black bile and yellow bile in a person’s constitution: the more blood, the more sanguine, or leader-like and optimistic; the more black bile, the more choleric, or irritable and bad-tempered; the more yellow bile, the more melancholic, or reflective and thoughtful; and the more phlegm, the more phlegmatic, or relaxed and calm, (Glalgan, 2015). So began the search for the determinants of human behaviour in the modern canon. Two-and-a-half millennia later we are still looking for a scientifically robust way to explain and predict how people will behave, and why they behave differently to one another. We are searching, in other words, for an understanding and classification of those consistent, internal, personal characteristics that drive recognisable behaviours, emotions and cognitions across heterogeneous situations – in short, “personality traits”, (Chamorro-Premuzic, 2007).

The most commonly accepted set of traits in use is the “Big Five”; there are competing inventories, but since Costa and McRae published their approach it has underpinned much of the thinking on the subject, (Chamorro-Premuzic & Furnham, 2010). The Big Five dimensions of personality are as follows: Extraversion, Agreeableness, Neuroticism, Conscientiousness and Openness. Individuals all possess these distinct traits to a greater or lesser degree, and their unique combination gives rise to what we call personality. Extraversion relates to a number of facets including warmth, gregariousness, assertiveness, and excitement-seeking, so someone with high degrees of extraversion will come across as talkative, social, dominating and energetic, whereas someone with low extraversion will be more withdrawn, thoughtful, solitary, reserved and pessimistic. Someone with high levels of Agreeableness will be trusting, altruistic, eager-to-please, tender and empathetic, whereas low levels imply cynicism, independence of thought, indifference to convention and single-mindedness. Neuroticism relates to levels of anxiety, hostility, depression, self-consciousness, impulsivity and vulnerability; the highly neurotic can be irritable, excitable and wired, whereas the emotionally stable may be complacent, over-confident and lazy. Openness to experience deals with characteristics such as fantasising, actions, aesthetics, willingness to explore new ideas and activities. Lastly, Conscientiousness measures the level of commitment to mastery, competence, dutifulness, striving for achievement, self-discipline and deliberation, (Costa, McCrae, & Dye, 1991). In 2004 Ashton et al published research positing a sixth factor, Honesty/Humility, which they found to be significant particularly when looking across cultures outside Europe and America. This sixth factor measures facets such as sincerity, modesty, fair-mindedness and (lack of) greed, (Ashton, Lee, Perugini, De Vries, Di Blas, & De Raad, 2004).

What, then, is the role of personality in driving job performance? Psychological research in the 1960s was deeply equivocal, eventually concluding that personality measures were not of help in making decisions about selecting people for jobs. Guion and Gottier could find no real correlation between measures of job performance and personality metrics, (Guion & Gottier, 1965). But these conclusions were based on research conducted at a time when there was little consistent taxonomy about personality measurement, so it was hard to conduct meaningful meta-analytic studies of research results across different experiments, (Chamorro-Premuzic & Furnham, 2010). As a result over the next decade or so personality was not much explored as a driver of job performance. Rather, research concentrated on the significance of factors such as cognitive ability, job knowledge, task proficiency and experience, (Schmidt, Hunter, & Outerbridge, 1986). But the tide turned, largely as a result of the development of the “Big Five” taxonomy that enabled psychology research to adopt more consistent definitions of what was meant by personality. By 1995, Kanfer, Ackerman, Murth and Goff found themselves able to declare that the argument about whether personality tests could be treated as useful for the prediction of an employee’s performance was “no longer pertinent”, (Kanfer, Ackerman, Murtha, & Goff, 1995).

Personality traits might be expected to have an influence on a number of aspects of an employee’s performance such as occupational interests, motivation to learn, training outcomes, job knowledge, employment status, organisation choice, job satisfaction, attainment level, entrepreneurship, leadership, managerial effectiveness, sales effectiveness, customer orientation, interpersonal effectiveness, counterproductive behaviour (theft, substance abuse, damage, absenteeism), teamwork and tenure, (Hough & Connelly, 2013). In 2001, for example, Barrick. Mount and Judge reported from a meta-analytic study across broad job types that each of the Big Five personality constructs correlated to job performance to a greater or lesser degree as follows: Extraversion, 0.15; Conscientiousness, 0.27; Neuroticism, 0.13; Agreeableness, 0.13, Openness, 0.07, (Barrick, Mount, & Judge, 2001). In a separate survey, Hough, Ones and Viswesvaran established that the effectiveness of managers correlated to the traits factors dominance (0.27), energy levels (0.20) and achievement-focus (0.17), (Hough, Ones, & Viswesvaran, 1998). Overall, Barrick and Mount estimated that personality traits account for 15% of work success. (Barrick & Mount, 1991).

Within the list of traits it seems that “conscientiousness” is the one that stands out as correlating most highly with overall job performance, (Hough & Furnham, 2003) and is perhaps the most important predictor of the success of an employee, (Schmidt & Hunter, 1998). This is intuitively appealing – the highly conscientious are likely to be hard-working, diligent, dutiful and to set themselves challenging goals, (Barrick, Mount, & Strauss, 1993). Putting it the other way around, those who lack conscientiousness are likely to be careless, impulsive, irresponsible lead-swingers who will do badly in almost any role, (Chamorro-Premuzic & Furnham, 2010).

So far we have only considered the importance of personality in determining an individual’s job performance. In reality, nearly all jobs involve some element of teamwork. Team effectiveness depends on a number of factors, but one of them is how well the people who in the team work together. The Big Five traits may be able to provide some insight. Conscientiousness once again is important, but needs to be seen in the context of how the team is operationalized, (LePine, Hollenbeck, & Ilgen, 1997). Extraversion would likely be positive for teamwork in that it drives towards helping behaviour, but the reasonableness and legitimacy of the request made can make a significant difference, (Poerter, Hollenbeck, Ilgen, West, & Moon, 2003). A team made up of individuals with high agreeableness as a trait might have a good chance of cooperating well, but is dependent also on the extent to which the reward system in place mitigates for or against the provision of mutual help, (Beersma, Hollenbeck, Moon, Conlon, & Ilgen, 2003). Lastly, a team of people with low neuroticism might be expected to be calm in a crisis and to perform well, but again this is dependent on context, in this case the nature of situational stresses that applies, (Hollenbeck, Moon, Ellis, West, Ilgen, & Sheppard, 2002).

So if we have an understanding of an individual’s personality traits does that mean we understand how they will behave in given situation, and how their behaviour will differ from one another? It seems that context is important; indeed as suggested by Mischel, amongst others, does the power of the situation overwhelm personality traits in determining how people will behave, (Mischel, 1963)? Mischel’s point was that the context dominates people’s behaviour. Before going on to a platform to make a speech, we are all likely to be nervous; at home watching television, we are all likely to be relaxed. However, despite Mischel’s point it seems clear that personality has an influence within a given situation – someone who is highly neurotic will likely be more anxious before giving a presentation than a counterpart who is not, and someone who has extreme levels of conscientiousness may find it hard to relax at home in front of the TV. Personality traits should not be ignored as drivers of broad behavioural patterns, (Eysenck & Eysenck, 1980). So we can understand behaviour – what we actually do or think, how we actually feel – as being a function of personality and situation combined, (Chamorro-Premuzic & Furnham, 2010).

In the context of work, therefore, whether as a team or as an individual, we can expect behaviour to be determined by the person andthe context or situation, (Mitchell, 1983). Mitchell posits that job performance depends upon an individual’s arousal, motivation and ability in the context of the nature of the tasks that have been set, the social factors surrounding the role, the organisational factors pertaining to the job and how role evaluations are conducted. Arousal at a given time is a measure of how stimulated and energised a person is, and motivation for a given task in a given role is a function of personality. Ability, by contrast, is a result of skill, training and experience, and to a large extent depends on cognitive power. Indeed, cognitive ability has been given credit for a full 25% of variances in job performance, (Schmidt & Hunter, 1998). Personality tests, therefore, should ideally be set in the context of the jobs and roles for which they will be applied, but for complex positions this can be very difficult to define, (Dakin, Nilakant, & Jensen, 1994).

How do personality tests actually work? Do they take into account context, and can they give a reliable and valid picture of personality? A definition of a personality test is a measurement of attributes that drive observable actions, thoughts or emotions thus allowing inferences of characteristics and predictions of behaviour, (Hubley & Zumbo, 2013). Putting it another way, personality tests are attempts to find the independent variables of behaviour which are different to the measures of simple cognitive processing capability, (Chamorro-Premuzic & Furnham, 2010). The word “test” is moot – a better word might be “assessment”. A test is appropriate to describe an objective process by which we try to establish maximal performance of an ability (we test for mathematic ability, or for knowledge of Latin vocabulary), whereas when looking at personality we are more interested in assessing typical (not maximal) behaviour, (Chronbach & Gleser, 1965).

A personality assessment is typically subjective. It relies upon self-report or peer-report, and is usually designed to collect data showing how an individual behaves in comparison to some norm or high-performing group, (Glalgan, 2015). At first blush it may seem that a subjective approach is inferior to an objective measure, but when trying to assess behavioural tendencies it may be actually be more pertinent. Objective measures typically rely on some sort of test environment which is in itself unnatural, (Chamorro-Premuzic & Furnham, 2010). There are plenty of variations in format (multiple choice, Likert scales, scaling criteria), but typically the way the subjective data is collected is as a set of responses to questions that are either normative or ipsative. The questions in a survey are usually developed based on factor analysis; i.e. items are selected for use depending on whether they load on to a particular factor such as one of the Big Five facets. Alternatively, they maybe set following a rational / theoretical understanding of the subject area, (Hubley & Zumbo, 2013).

But the key question is whether personality assessments are reliable and valid such that they can be confidently used and interpreted based on evidence and theory, (Messick, 1989). Hubley and Zumbo provide a useful framework for considering this problem. They suggest that personality assessments be appraised along (amongst others) the dimensions of reliability, content-related evidence, and criterion-related evidence. A reliable assessment will produce data that is repeatable and consistent; in other words, two forms of the same test at the same time will give them same answer (equivalence), the same test at the different times will give the same answer (stability) and two different markers will give the same score for the same responses (inter-rater consistency). To be valid, it is a necessary but not sufficient condition that an assessment is reliable. Content-related evidence of validity tests the extent to which the assessment measures what it intends to measure, (Haynes, Richard, & Kulbany, 1995), whereas criterion-related evidence looks at the extent to which data from a response is related to a criterion. An alternative but similar set of construct validity challenges is posed by Chamarro-Premuzic and Furnham: How do the questions look to people taking the test, do they seem sensible and relevant, (“Face Validity”); do they measure what we want them to measure? (“Content Validity”); does the data correlate to how people actually behave outside the assessment process? (“Predictive Validity”); and does the assessment connect the same items to the same dimensions? (“Factorial Validity”), (Chamorro-Premuzic & Furnham, 2010).

We may (theoretically) design a personality assessment that meets all the requirements of reliability and validity, but what is the consequence if assessment-takers choose to fake their responses? Individuals or peers may want to convey a distorted impression of themselves / the subject of the test. If the stakes are high, as in, for instance, a job application, applicants are motivated to try to project themselves what they think is the “right” way. In an extensive literature review by Campion, 39 studies confirmed that scores can be distorted, that people do not distort by as much as they theoretically could (i.e. when compared to when actively prompted to try to fake), that faking does affect criterion-related validity, and that faking cannot be detected adequately, (Morgeson, Campion, Dipboye, Hollenbeck, Murphy, & Schmitt, 2007). Is the whole exercise then redundant? No; studies have shown that when respondents try to be “extra-desirable”, their responses are only marginally different to when they are being exact, (Hough, Eaton, Dunnette, & McCloy, 1990). Another way of looking at it is to argue that if people are good at adapting their responses to the situation in an assessment, they’ll probably be good at it in “real life”, (Tippins, 2013).


If we had a perfect description and quantification of an individual’s personality traits, would our talent management problems be over? No. There are two broad reasons. Firstly, how would we know what would be the “right” personality for a given role? This is not clear, especially for roles that have multiple dimensions like many senior management jobs; one person needs to be a high-energy communicator but also able to think strategically, to take a strong lead but also build a team. Moreover, it is difficult to assess the right balance even of a single trait – unlike testing for, say, running speed where faster is unequivocally better than slower, there are advantages to every point on the scale of, for instance, neuroticism. It is even possible to be too conscientious such that one becomes overly perfectionist and/or overly dutiful and thus can never complete anything or take any initiative, (Benson & Campbell, 2007). Secondly, personality only sets, to use a musical analogy, the key in which the music is written. It does not create the melody, the harmony, the tempo or instrumentation. That all depends on ability, skill, experience on top of personality, and even then the crucial factor is context. Context is infinitely complex, dynamic and unpredictable, so to suppose that we can analyse personality and then come up with the perfect selection of employee is otiose.

And, of course, we don’t have a perfect way of describing or quantifying personality traits. Personality assessment is crude, subject to distortion, and troubled by weaknesses in reliability and validity. So are assessments a waste of time and money? (A lot of time and money, as it happens – the personality assessment market is worth USD 2bn-4bn per annum, (Kinley & Shomo, 2013)). If put to the wrong use, if relied upon as a sole arbiter for selection and development, then the answer is yes, they are a waste. But if used intelligently as part of a portfolio of tools that help the talent manager do a better job, then assessments can be very valuable.

Alongside recommendation letters, structured interviews, multi-source feedback, cognitive ability tests, CV reviews, performance appraisals and simple observation, personality assessments provide a valuable place to start a talent management process, (Chamorro-Premuzic & Furnham, 2010). One good example is coaching; at the commencement of a coaching assignment for an individual or team, a set of personality assessments revealing Big Five factors and hence leadership / working style inventories can be a powerful way of raising self-awareness and beginning a process of change and growth, (Hawkins, 2014). This is not to say coaching or any other talent management process such as selection or development depends upon personality assessment for success, and nor should it, but they can certainly be enhanced by it.


Ashton, M., Lee, K., Perugini, M., De Vries, R., Di Blas, L., & De Raad, B. (2004). A six- factor structure of personality-descriptive adjectives: Solutions from psycholexical studies in seven languages. Journal of Personality and Social Psychology , 86, 356-366.

Barrick, M., & Mount, M. (1991). The Big Five personality dimensions and job performance: A meta-analysis. Personnel Psychology , 44, 1-26.

Barrick, M., Mount, M., & Judge, T. (2001). Personality and performance at the beginning of the new millennium: What do we know and where do we go next? International Journal of Selection and Assessment , 9, 9-30.

Barrick, M., Mount, M., & Strauss, J. (1993). Conscientiousness and performance of sales representatives: test of the mediating effects of goal setting. Journal of Applied Psychology , 75 (5), 715-722.

Beersma, B., Hollenbeck, J., Moon, H., Conlon, D., & Ilgen, D. (2003). Co-operation, competition, and team performance: Towards a contingency approach. Academy of Management Journal , 46, 572-590.

Chamorro-Premuzic, T. (2007). Personality and Individual Differences. Oxford: Blackwell.

Chamorro-Premuzic, T., & Furnham, A. (2010). The Psychology of Personnel Selection. Cambridge: Campbridge University Press.

Chronbach, L., & Gleser, G. (1965). Psychological Tests and Personnel Decisions. Urbana, IL: Universsity of Illinois Press.

Costa, P., McCrae, R., & Dye, D. (1991). acet Scales for Agreeableness and Conscientiousness: A Revision of the NEO Personality Inventory. Personality and Individual Differences , 12 (9), 887-898.

Dakin, S., Nilakant, V., & Jensen, R. (1994). The Role of Personality Testing in Managerial Selection. Journal of Managerial Psychology , 9 (5), 3-11.

Eysenck, H., & Eysenck, S. (1980). Mischel and the concept of personality. British Journal of Psychology , 71 (2), 191-204.

Glalgan, P. (2015). Tester Beware. Talent Development , 69 (9).

Guion, R., & Gottier, R. (1965). Validity of personality measures in personnel selection. Personnel Psychology , 18, 135-164.

Hawkins, P. (2014). Leadership Team Coaching. London: Kogan Page.

Haynes, S., Richard, D., & Kulbany, E. (1995). Content validity in psychological assess- ment: A functional approach to concepts and methods. Psychological Assessment , 7, 238-247.

Hollenbeck, J., Moon, H., Ellis, A., West, B., Ilgen, D., & Sheppard, L. (2002). Structural contingency theory and individual differences: Examination of external and internal person-team fit. Journal of Applied Psychology , 87, 599-606.

Hough, L., & Connelly, B. (2013). personality Measurement and Use in Industrial and Organizational Psychology. In K. Geisinger (Ed.), PA Handbook of Testing and Assessment in Psychology: Vol. 1. Test Theory and Testing and Assessment in Industrial and Organizational Psychology. (pp. 501-531).

Hough, L., & Furnham, A. (2003). Importance and use of personality variables in work settings. In W. I.B. (Ed.), Comprehensive handbook of psychology: Vol. 12. Industrial and organizational psychology (pp. 131-169).

Hough, L., Eaton, K., Dunnette, M., & McCloy, R. (1990). Criterion-related validities of personality constructs and the effect of response distortion on those validities. Journal of Applied Psychology , 75, 581-595.

Hough, L., Ones, D., & Viswesvaran, C. (1998). Personality correlates of managerial performance constructs. Personality determinants of managerial potential performance, progression and ascendancy. Dallas: Symposium conducted at the 13th Annual Conference of the Society for Industrial and Organizational Psychology.

Hubley, A., & Zumbo, B. (2013). Psychometric Characteristics of Assessment Procedures: An Oveview. In K. Geisenger (Ed.), APA Handbook of Testing and Assessment in Psychology: Vol. 1. Test Theory and Testing and Assessment in Industrial and Organizational Psychology. (p. 3).

Kanfer, R., Ackerman, P., Murtha, T., & Goff, M. (1995). Personality and intelligence in industrial and organizational psychology. In S. D.H., & Z. M. (Eds.), International handbook of personality and intelligence. New York: Plenum.

Kinley, N., & Shomo, B. (2013). Talent Intelligence: What you need to know to identify and measure talent. San Francisco: John Wiley & Sons.

LePine, J., Hollenbeck, J., & Ilgen, D. (1997). Effects of individual differences on the performance of hierarchical decision making teams: Much more than g. Journal of Applied Psychology, , 82, 803-811.

Messick, S. (1989). Validity. In Educational Measurement (pp. 13-103). New York, NY: Macmillan.

Mischel, W. (1963). Toward a cognitive social learning reconceptualisation of personality. Psychological Review , 80, 252-283.

Mitchell, T. (1983). The Effects of Social, Task, and Situational Factors on Motivation, Performance and Appraisal. In F. Landy, S. Zedeck, & J. Cleveland (Eds.), Performance Measurement and Theory. new Jersey: lawrence Erlbaum.

Morgeson, F., Campion, M., Dipboye, R., Hollenbeck, J., Murphy, K., & Schmitt, N. (2007). Reconsidering the use of personality tests in personnel selection contexts. Personnel Psychology , 60, 683-729.

Poerter, C., Hollenbeck, J., Ilgen, D., West, B., & Moon, H. (2003). Backing up behaviors in teams: The role of personality and legitimacy of need. Journal of Applied Psychology , 88 (391-403).

Schmidt, F., & Hunter, J. (1998). The validity and utility of selection methods in personnel psychology: practical and theoretical implications of 85 years of research finding. Psychological Bulletin , 124, 262-274.

Schmidt, F., Hunter, J., & Outerbridge, A. (1986). Impact of job experience and ability on job knowledge, work sample performance, and supervisory ratngs of job performance. Journal of Applied Psychology , 71, 432-439.

Tippins, N. (2013). Assessment of Leadership. In K. Geisinger (Ed.), APA Handbook of Testing and Assessment in Psychology: Vol. 1. Test Theory and Testing and Assessment in Industrial and Organizational Psychology (pp. 457-478).