Personality assessments can add valuable information in selection contexts. They are widely used in personnel selection and can contribute to the prediction of job and study performance. In higher education admissions, however, they are used far less frequently.
One major concern is that candidates may deliberately adjust their answers to align with the perceived profile of an ideal applicant rather than reporting their true standing — a phenomenon known as faking. In selection contexts, this usually takes the form of faking good or impression management: applicants distort their responses in order to create a specific, overly positive impression of themselves.
This raises a central tension. On the one hand, faking may threaten the validity of personality assessments and distort selection decisions. On the other hand, some level of self-presentation may reflect socially expected and potentially success-relevant behaviour.
So the question is not simply: Can applicants fake personality tests?
The more relevant question is: Does faking make personality assessments useless for admissions?
The answer is: not necessarily — but it must be taken seriously.
The “faking is bad” perspective
From a psychometric viewpoint, faking is problematic because it reduces construct validity. If applicants systematically inflate their scores, the test no longer measures personality as cleanly as intended. Deliberate distortion in socially desirable directions can:
- increase mean scores, especially on traits as conscientiousness and emotional stability
- distort rank orders, so that more strategic applicants may be ranked higher
- make individual profiles harder to interpret, because some of the variance reflects impression management rather than the trait itself
In selection decisions, these effects matter especially when the test is used to identify the top candidates. In such cases, faking can lead to suboptimal admissions decisions and may reduce fairness.
The “faking is not all bad” perspective
However, there is an alternative perspective. Making a positive impression is widely accepted — and often expected — in daily life. In job interviews, applications, networking situations, or presentations, people rarely show their full, unfiltered selves. They adapt to the situation.
Some researchers therefore argue that self-presentation in selection contexts may be a legitimate and adaptive response to situational expectations. From this perspective, the ability to manage one’s self-presentation may even involve success-relevant skills, such as:
- the analytical ability to infer what is expected
- the willingness to adapt to expectations
- behavioural flexibility in presenting oneself appropriately
As a result, even if faking reduces construct validity, criterion-related validity may still remain meaningful. In simpler terms: personality assessments may no longer measure “pure” personality under high-stakes conditions — but they may still predict relevant outcomes to some extent.
How big is the faking problem in practice?
Hu and Connelly (2021) meta-analysed studies comparing personality responses in low-stakes contexts, where test results are not important for selection, and high-stakes contexts, where test results can influence selection decisions.
Their main finding: Faking is real — but not always as strong as many people expect.
Key findings include:
- Applicants tend to fake in high-stakes contexts.
- The effects are moderate, for example around d = 0.44 for extraversion and d = 0.59 for conscientiousness. As a rough orientation, effect sizes of d = 0.20, 0.50, and 0.80 are often considered small, medium, and large, respectively.
- Faking can lead to rank-order changes, which is relevant in admissions contexts.
Studies comparing the predictive validity of personality assessments in low-stakes and high-stakes situations suggest that predictive validity does not disappear under high-stakes conditions — but it tends to become smaller. Loy, Christiansen, Tett, Klein, and Toich (2025), for example, found that personality test validity differs between low-stakes and high-stakes employment settings.
These findings do not rule out the use of personality assessments in high-stakes contexts. But they are substantial enough to show that faking should be treated as a relevant practical challenge.
Can faking be detected or reduced?
Methods to detect faking
Several methods have been proposed to identify applicants who distort their answers. However, the evidence is mixed, and many detection methods are less reliable than one might hope.
- Social desirability scales: Social desirability scales, such as the Marlowe-Crowne Social Desirability Scale, are designed to flag potential faking. They often include items that describe socially desirable but statistically unlikely behaviours, for example: “I have never lied in my life.” However, empirical research suggests that using these scales to “correct” personality scores often does not improve validity — and may even reduce it.
- Overclaiming techniques: Overclaiming techniques aim to detect faking by including items that refer to non-existent facts, concepts, or information. Applicants who claim familiarity with these “bogus” items may be flagged as likely overclaimers. However, research suggests that the practical utility of these techniques is limited.
- Machine-learning approaches: Machine-learning models can identify characteristic response patterns associated with faking. However, their accuracy remains inconsistent. They are also prone to classification errors. This is especially problematic in admissions, where false accusations of faking could have serious consequences for applicants. For this reason, such tools should be used with great caution and are currently better suited for research than for direct high-stakes decision-making.
In short: current methods for detecting faking have not fully met expectations yet.
Methods to prevent or reduce faking
Because detection is difficult, many researchers favour preventing or reducing faking.
- Warnings: Warnings can reduce response distortion. Applicants may be told that deception can be detected, that it may have consequences, or that faking can undermine the purpose of the assessment. Moon et al. (2025) found that such warnings reduce faking with a significant, small to moderate effect of d = 0.31. However, warnings also need to be used carefully. They may increase anxiety in honest but self-critical candidates. And claiming that all faking will be detected is factually false, which may be problematic from a legal or ethical perspective.
- Neutralized items: Neutralization involves rephrasing desirable and undesirable items so that they sound more neutral. This can reduce the temptation to change answers purely for the sake of impression management. The risk is that if all items become too neutral, the assessment may lose bandwidth and become less able to identify individuals at the extremes of normal personality.
- Forced-choice formats: Forced-choice formats ask respondents to choose between several statements that are similarly desirable. For example: “I am a highly reliable team player.” or “I am an effective problem-solver.” Because candidates cannot simply endorse all positive statements at the same time, it becomes harder to inflate scores on multiple traits simultaneously. A meta-analysis by Speer et al. (2023) suggests that faking can still occur in forced-choice formats, but that it is generally smaller than in classical rating-scale formats.
- External and verifiable data: Another option is to reduce reliance on unverifiable self-reports. Observer reports, references, structured interviews, biodata items, or documented experiences can complement self-report personality measures. These data sources are not immune to bias or distortion, but they can help provide a broader and more robust picture of the applicant.
Practical implications for higher education admissions
What does this mean for higher education institutions that are considering personality assessments in admissions? Based on the empirical findings, we suggest three practical principles.
1. Prioritize prevention over detection
Institutions should focus on reducing the opportunity and motivation to fake. This can include:
- carefully worded instructions with appropriate warnings
- neutralized items
- forced-choice formats
- complementary data sources
This is usually more promising than relying on post-hoc detection of individual fakers.
2. Be cautious when using personality tests to rank top candidates
Faking is particularly problematic when personality scores are used to identify and rank the “best” candidates. In such cases, even moderate score inflation or rank-order changes can affect selection decisions. For admissions, personality assessments may therefore be more appropriate as part of a broader assessment strategy than as a single decisive ranking tool. Depending on the context, they may be more useful for identifying very low levels of clearly relevant traits than for selecting the very highest-scoring applicants.
3. Combine personality assessments with interviews or other methods
Takeaway
Faking is real. It is not trivial. And it should not be ignored. But faking does not make personality assessment useless. The empirical evidence suggests that personality assessments can still provide relevant information in selection contexts — even though their validity may be reduced under high-stakes conditions. For higher education admissions, this means that personality assessments require careful handling. They should be designed and used in ways that reduce faking, avoid overreliance on self-report, and combine multiple sources of information.
Personality assessments can be particularly useful when combined with structured interviews, Multiple Mini-Interviews, situational tasks, or other assessment tools. For example, a personality questionnaire may provide hypotheses that can later be explored in a structured interview. If candidates know that their responses may be followed up in a personal interview, this may also reduce the incentive to fake. Taken together, personality assessments and structured interviews can provide a more complete picture than either method alone.
A promising approach is to:
- prevent faking rather than trying to detect it
- avoid using personality scores as the only basis for ranking applicants
- combine personality assessments with structured interviews or other admissions tools
- use personality scores as one part of a broader, evidence-based admissions process
In short: Personality assessments are not useless because applicants can fake. They can be useful if the faking problem is taken seriously.
Further reading
Hu, J., & Connelly, B. S. (2021). Faking by actual applicants on personality tests: A meta-analysis of within-subjects studies. International Journal of Selection and Assessment, 29, 412–426. https://doi.org/10.1111/ijsa.12338
Loy, R. W., Christiansen, N. D., Tett, R. P., Klein, K., & Toich, M. (2025). Personality test validity differs between low-stakes and high-stakes employment settings. International Journal of Selection and Assessment. https://doi.org/10.1111/ijsa.70018
MacCann, C., Ziegler, M., & Roberts, R. D. (2011). Faking in personality assessment: Reflections and recommendations. In M. Ziegler & C. MacCann (Eds.), New perspectives on faking in personality assessment (pp. 309–329). Oxford University Press.
Moon, B., Daljeet, K. N., O’Neill, T. A., Harwood, H., Awad, W., & Beletski, L. V. (2025). Comparing the efficacy of faking warning types in preemployment personality tests: A meta-analysis. Journal of Applied Psychology, 110, 131–147. https://doi.org/10.1037/apl0001224
Speer, A. B., Wegmeyer, L. J., Tenbrink, A. P., Delacruz, A. Y., Christiansen, N. D., & Salim, R. M. (2023). Comparing forced-choice and single-stimulus personality scores on a level playing field: A meta-analysis of psychometric properties and susceptibility to faking. Journal of Applied Psychology, 108, 1812–1833. https://doi.org/10.1037/apl0001099
