How to adopt personality assessments in selection

December 22nd, 2022
Tariq Shaban, Senior IO Psychologist
General

Everyone is heads down, strategizing for the new year, and talent teams in particular are evaluating what worked well and what they need in what's sure to be another complex year in the industry. Will the Great Resignation still pose problems? Are layoffs in your sector imminent? 

As you look ahead at which tools can set your talent experience apart, it’s a great time to evaluate the way personality assessments can fit into the puzzle. But before you make any choices, let’s separate fact from fiction.

What are personality assessments?

Personality assessments are questionnaires that attempt to measure traits, or the ways in which one person differs from another. They are a popular method of assessing people’s preferences, behavioral styles, interests, and motivations, and there are myriad benefits to incorporating personality assessments into your selection practices: Better diversity and inclusion, a better understanding of people’s preferences, improved performance, and a general fit to the role are some of the most important. 

In order to truly benefit from these assessments, companies should closely follow a few common sense recommendations to ensure they deliver in a selection context.

Research-backed models and impression management

Personality assessments vary in what traits they measure, how they measure them, and how well or accurately they can be measured. Companies looking to implement these assessments in their hiring processes should only use reliable and valid assessments that are based on research-backed models such as the Five Factor Model of personality. 

Additionally, practitioners should consider how assessments address candidate motivations and test-taking attitudes. It is well established in research that people responding to a personality questionnaire are motivated to agree with positive statements and disagree with negative statements. It is not wrong or deceptive for candidates to manage their impressions and respond in this socially desirable way, but personality assessments must address how they tackle faking and impression management tendencies in applicants. 

It isn’t enough for vendors to simply say that faking isn’t a concern or that the use of images or social desirability scales solve the problem. Vendors need to demonstrate in their technical manuals that they have adopted methods that improve personality assessments and reduce faking. 

These methods could include statistical techniques or artificial intelligence scoring methods that are able to produce normative scores which are essential for comparing applicants from ipsative or forced-choice designs that are harder to fake analyzing candidate responses to paired item choices.

Example of a normative item

Normative assessments ask test-takers to rate how they feel about a series of statements such as “I am very organized.” 

Survey question prompt

Candidates would have the option to respond on a Likert scale that is typically on a five-point scale ranging from “strongly disagree” to “strongly agree.” Their responses would be summed and then directly compared to those of other respondents. These types of personality assessments are called normative assessments and they are great for comparing test-takers. However, they are very susceptible to a tendency that test takers have to manage impressions.

Example of an ipsative item

Ipsative assessments present test-takers with two or more item choices and they are asked to select which is most like them. This method attempts to address the issue of impression management by “forcing” test-takers to decide between a number of equally appealing choices. 

When all the options are equally desirable then test-takers are left with no option other than to answer honestly. While ipsative methods succeed in making it difficult for test-takers to fake their responses or game the assessment, they are not placed to be used for selection unless advanced statistical techniques are used to normalize their results.

Survey prompt: Select the response that is most like you.

The job analysis is the backbone of ethical use

The use of any assessment in selection is mostly dependent on its ability to predict real-world outcomes. This is where the job analysis to understand what aspects of personality should be included in the assessment is critical. The analysis should be conducted at the job or job-family level; focusing on the Knowledge, Skills, Abilities, and Other characteristics (KSAO) required for the role.

The most recent meta-analytic research on personality measurements shows that while they add value, their impact is greatest when they are contextualized with a job analysis.

The table below shows the correlation coefficients of the factors from the Big Five Model of personality to work performance. The TL;DR version is that you must conduct a job analysis to determine what aspects of personality are related to the job before applying the assessment, and a one-size-fits-all approach isn’t likely to add much value.

General Validity Estimate Contextualized Validity Estimate
Conscientiousness 0.19 0.25
Emotional Stability 0.09 0.23
Extraversion 0.10 0.21
Agreeableness 0.10 0.19
Openness to Experience 0.05 0.12

Based on a scale of 0.0 to 1.0, where .11 is unlikely to be useful, a correlation between .11 and 0.20 is useful depending upon the situation (e.g., where a high volume of candidates is being assessed), a correlation between 0.21 and 0.35 is likely to be useful, and a correlation above .35 is very beneficial in predicting job performance (U.S. Department of Labor, 1999). 

Pair personality assessments with other selection tools 

Well-rounded hiring practices rely on multiple sources of information when making decisions. After all, people are complex and can succeed in many different ways. Personality, cognitive ability, and structured interviews each measure distinct aspects of how people may perform in a job. By combining their unique perspectives, we achieve a better and more comprehensive view of potential performance. This is a practice that is well supported in research that shows improved job performance outcomes when multiple predictors are used as part of the selection process.

Personality assessments, therefore, should be used along with other measures such as ability and skills-based assessments to form a more holistic view of the person and their fit for the job. Skills-based assessments should be the primary focus when assessing candidates where possible, with personality being a supplementary or added-value assessment to help understand a person’s preferences and fit to the role.

Putting best practices into action with Portrait

HireVue brings together the recommended best practices for personality assessments in the game-based assessment, Portrait. Built upon the Big Five Model of personality, the assessment classifies candidates by their predisposition to five high-level traits: openness to experience, conscientiousness, extraversion, agreeableness, and emotional stability. 

  • People high in openness are typically imaginative, artistic, feeling-oriented, action-oriented, and ideational. 
  • People high in conscientiousness are responsible, dutiful, self-controlled, and enjoy planning. 
  • People high in extraversion seek excitement, and are typically warm, gregarious, assertive, and focused on external sources of stimulation. 
  • People high in agreeableness tend to be trusting, altruistic, compliant, and modest. 
  • People high in emotional stability typically remain composed in the face of challenges or setbacks, and are generally calm and able to cope well with stress and criticism.

Portrait's unique image-based format offers a refreshing and fast-paced candidate experience. Instead of asking candidates to answer a series of text-based items, they are presented with image pairs where they are asked to select which image is most like them. This semi-ipsative design is key to accounting for test-taking motivations. When paired with artificial intelligence scoring algorithms, they are more effective at representing a candidate's personality as well as improving diversity outcomes by reducing subgroup differences for gender, age, and ethnicity.

The Portrait assessment is only implemented after a rigorous job analysis and is always paired with another measurement tool– this allows us to improve job prediction by contextualizing the assessment to on-the-job requirements and incorporating multiple forms of assessment.

To learn more about how personality assessments can help you create a better hiring experience, book a time to speak with one of our IO psychologists.