Can People Fake an EQ Test During Hiring?

Can people fake an emotional intelligence test during hiring?

Yes. Applicants can present themselves more favorably on some emotional-intelligence questionnaires when they are told to imagine a job-selection situation. That response shift is evidence that scores can be influenced by the testing context; it does not establish that an applicant has lied. Researchers call this response distortion or favorable self-presentation. A high score alone cannot distinguish deliberate impression management from an applicant's genuine self-view or interpretation of a question.

Two studies illustrate why the answer depends on the instrument. In “Faking emotional intelligence (EI): comparing response distortion on ability and trait-based EI measures,” participants completed the EQ-i under simulated applicant instructions and later under instructions to answer honestly; the EQ-i shifted more than the ability-based MSCEIT. The abstract does not establish how often this occurs in hiring. In the 2024 study “Faking Good on Self-Reports Versus Informant-Reports of Emotional Intelligence,” 81 community volunteers and 151 university students were assigned honest or simulated-selection instructions. Both TEIQue-SF self-ratings and informant ratings were higher in the simulated condition. Those instructed-response results show susceptibility for these measures, not real-world prevalence or effects on job performance. So the practical question is what evidence supports using a named assessment for a defined job.

Sources: Faking Good on Self-Reports Versus Informant-Reports of Emotional Intelligence; Faking emotional intelligence (EI): comparing response distortion on ability and trait-based EI measures

The label 'EQ test' does not name one task

An EQ label may describe a questionnaire about typical emotional behavior, an informant rating, or a task involving emotional problems. A self-report records what a person says about their own typical behavior; a performance task records how they respond to particular problems under the test's scoring rules. Calling both results an EQ score can hide that difference.

In measurement research, trait emotional intelligence refers to perceived emotional dispositions and typical functioning: for instance, whether someone sees themselves as able to stay composed or notice another person's feelings. The adjective “trait” here describes a pattern people report about themselves, not a fixed identity. Ability emotional intelligence instead refers to performance on tasks involving emotional information. The Office of Personnel Management's “Emotional Intelligence Tests” page describes examples such as identifying emotions in faces, interpreting emotional scenarios, and selecting responses to emotional problems. These tasks ask for answers to set prompts; they do not ask applicants to rate how empathic or calm they believe they are.

A third format changes who supplies the description. An informant rating asks someone else to report tendencies they have observed. It reports observed tendencies rather than direct task performance. The informant's answer depends on what they have had a chance to see, while a task score depends on responses to its prompts. Neither should be merged with the other as though both read the same meter.

The 2019 critical review “The Measurement of Emotional Intelligence: A Critical Review of the Literature and Recommendations for Researchers and Practitioners” examines the field's different measurement traditions. Some instruments combine features or use broader competency language, so the label on a test alone may not tell a candidate what was measured. The named instrument and its method matter.

This distinction helps explain why studies of response distortion can reach different results without contradicting one another. In “Faking emotional intelligence (EI): comparing response distortion on ability and trait-based EI measures,” participants completed the EQ-i and MSCEIT under applicant instructions and later under honest-response instructions. The publisher abstract reports greater susceptibility to distortion for the EQ-i than for the MSCEIT in those conditions. That comparison concerns those two instruments and that simulated setup. It does not establish a rule that every self-report is easy to manipulate or every ability task resists it.

When someone says an applicant “faked EQ,” ask what the applicant was asked to do. A more favorable description of one's own habits, a colleague's account of observed behavior, and a changed answer to a scored emotional problem are distinct events. Naming the format ties the conclusion to what could have shifted and what the score can support.

Sources: Emotional Intelligence Tests; The Measurement of Emotional Intelligence: A Critical Review of the Literature and Recommendations for Researchers and Practitioners; Faking emotional intelligence (EI): comparing response distortion on ability and trait-based EI measures

Why does a hiring situation make an answer easier to shift?

A self-report questionnaire asks someone to describe their usual emotional habits. In hiring, those descriptions can also signal what the employer wants. When a desirable answer is easy to infer, an applicant can present a favorable version of their behavior. That may be deliberate exaggeration, but a high score alone cannot establish intent. Someone may sincerely describe an ideal self, interpret a broad item differently, or answer from a narrow slice of experience. The assessment records responses; it cannot reveal why a person chose them.

The 2012 study “Faking on Self-Report Emotional Intelligence and Personality Tests: Effects of Faking Opportunity, Cognitive Ability, and Job Type” examined this under simulated selection conditions. Its 150 undergraduate participants completed self-report emotional-intelligence and personality measures honestly, then answered as if seeking to appear ideal for one of three jobs: nurse practitioner, marketing manager, or computer programmer. The researchers reported greater faking among participants with higher cognitive ability, those whose honest scores left more room to rise, and on traits considered relevant to the assigned job. These factors together produced the largest difference between honest and applicant-style responses in the study.

The design makes the mechanism clearer. A test taker has more room to move a score that began low than one already near the favorable end. A prompt that appears important for the role also gives a clearer clue about the impression to make. The findings complicate the claim that every applicant can inflate every scale by the same amount: opportunity to shift depended partly on starting score and the match between a trait and the hypothetical job. They provide no rate for real applicants. The participants were undergraduates following study instructions, and the reported differences do not show what an employer would observe on the job.

A later experiment points to the role of instructions. In “Faking Good on Self-Reports Versus Informant-Reports of Emotional Intelligence,” 81 community volunteers and 151 university students completed the short-form Trait Emotional Intelligence Questionnaire under honest-response or simulated job-selection instructions. Both self-reports and informant reports were higher in the simulated-selection condition. Even when a questionnaire asks about typical behavior, changing the response goal can change answers. Because this study also used instructed simulation, it cannot show how common deliberate distortion is among applicants or how a score shift affects hiring outcomes.

For a candidate, an unexpectedly flattering result is not automatic evidence of lying. For an employer, a warning flag or score pattern is not a lie detector. The studies show that self-descriptions can move when stakes and response goals change; they do not identify the motive behind an individual answer. The practical question is whether the named instrument supports the employment decision and whether its design gives applicants an obvious script for presenting a preferred image.

Sources: Faking on Self-Report Emotional Intelligence and Personality Tests: Effects of Faking Opportunity, Cognitive Ability, and Job Type; Faking Good on Self-Reports Versus Informant-Reports of Emotional Intelligence

Which test format changes the answer?

A self-report asks applicants to describe their emotional habits; an ability test asks them to solve emotional problems. In a direct comparison of the EQ-i and MSCEIT, the EQ-i shifted more under simulated applicant instructions. In the study directly comparing the EQ-i and the MSCEIT, the self-report EQ-i shifted more under simulated applicant instructions. This makes format relevant, though it does not establish that ability tests cannot be manipulated or are suitable hiring verdicts. In “Faking emotional intelligence (EI): comparing response distortion on ability and trait-based EI measures,” participants completed both the EQ-i and the MSCEIT in two sessions. In the first, they received a job description and answered as if applying for that job. Two weeks later, they repeated both tests with instructions to answer honestly. The published abstract reports a larger condition difference for the EQ-i than for the MSCEIT. It also says that the study’s faking indices predicted EQ-i scores in the applicant condition after the honest-condition EQ-i scores were taken into account. The same indices were not related to applicant-condition MSCEIT scores once honest-condition MSCEIT scores were controlled. Under a simulated top-down selection rule, participants were more likely to be selected on the basis of their applicant-condition EQ-i score than their honest-condition EQ-i score; selection likelihood was equal across conditions for MSCEIT scores. The EQ-i was more susceptible to response distortion than the MSCEIT in this study. That difference fits the tasks’ basic formats. The U.S. Office of Personnel Management’s “Emotional Intelligence Tests” guidance describes self-report as a person’s account of their perceived competence or self-image. An ability-based measure instead presents questions with responses scored by expert judgment or by agreement among a reference group. Its examples include identifying emotions in faces, interpreting emotional scenarios, and solving emotion-related problems. A flattering claim about one’s usual conduct can be chosen directly on a self-report questionnaire. On an ability task, a candidate has to produce an answer that scores well. The task narrows the route available for favorable self-description. The study does not tell us that every self-report test shifts by the same amount, or that all ability tests resist distortion. It examined two named measures under a laboratory simulation. The study did not test actual applicants or subsequent job performance. Its simulated selection result is not evidence about real employers’ decisions or workplace outcomes. A separate experiment helps check whether this difference is simply a contrast between one questionnaire and one task test. “Faking Good on Self-Reports Versus Informant-Reports of Emotional Intelligence” examined the short-form Trait Emotional Intelligence Questionnaire with community volunteers and university students. Participants assigned simulated job-selection instructions gave higher self-ratings than participants instructed to respond honestly; informants asked to rate someone else also gave higher ratings under the selection condition. Because it used another self-report instrument and design, it cannot validate or overturn the EQ-i/MSCEIT comparison. Together, the studies suggest reports of typical emotional tendencies can move under selection incentives, including informant reports. An ability score answers a different question from a self-rating: how the candidate responded to the test’s emotional problems. It does not directly show how that person listens during a tense exchange, recognizes when a colleague needs clarification, or repairs a misunderstanding over time. Conversely, a questionnaire may capture how someone sees their regular behavior, while remaining open to favorable self-description in a consequential setting. Neither format alone establishes the full pattern of conduct required in a particular role. The practical verdict is therefore about the route to distortion, not a hierarchy of universally safe tests. The comparison suggests ability-task format reduced the distortion observed for the EQ-i under those instructions. The OPM guidance similarly distinguishes self-report from ability tasks and notes ability models may be more acceptable when self-report fakability is a concern, but this is not instrument-specific proof. A buyer still needs evidence that the named test supports the interpretation and employment decision being proposed. Format can change what to scrutinize; it cannot supply that job connection by itself.

Sources: Faking emotional intelligence (EI): comparing response distortion on ability and trait-based EI measures; Faking Good on Self-Reports Versus Informant-Reports of Emotional Intelligence; Emotional Intelligence Tests

Can a colleague's rating solve the problem?

A colleague's rating may seem to check an applicant's account of handling emotion at work. Yet the second report is still a judgment made by a person. In “Faking Good on Self-Reports Versus Informant-Reports of Emotional Intelligence,” Walker and MacCann tested whether informants would also raise ratings when the stated purpose resembled job selection. Their result challenges a simple fix: another rater did not make the score immune to favorable presentation.

The researchers used the Trait Emotional Intelligence Questionnaire-Short Form (TEIQue-SF), a questionnaire about typical emotional tendencies. In a 2 × 2 between-person design, participants completed a self-report or informant-report under honest or simulated job-selection instructions. The sample comprised 81 community volunteers and 151 university students, 232 people altogether. It did not involve actual applicants in an employer's selection process, so it shows what respondents can do under favorable-presentation instructions, not how often candidates do so in ordinary hiring.

Scores were higher in the job-simulation condition for both report types. The standardized mean difference was d = 1.47 for self-reports and d = 1.56 for informant-reports, relative to honest instructions. These figures describe gaps between group averages in this experiment. They are not percentages of people who faked or estimates of real-world prevalence. The authors concluded that participants could present higher emotional-intelligence ratings for themselves and for another person under simulated selection instructions.

This informant result tests the assumption that another person's view must be more objective than self-report. A colleague may add perspective when they have repeatedly observed a specific behavior, but an informant questionnaire still asks someone to interpret another person's tendencies. A rater may also want to help the applicant or meet perceived employer expectations. The study supports the narrower point that informant reports can shift under favorable instructions; it does not show that every colleague will do so or that all multi-rater processes are unreliable.

A higher rating alone does not prove that anyone lied. In an application, scores may also reflect self-perception, what the rater has observed, or how either person understands the questions. This study cannot explain a particular candidate's result because it did not assess real hiring cases. A flag for favorable responding should prompt scrutiny of the measure and process, not become a verdict about an individual's honesty.

A second report can contribute evidence when the behavior is clear, the observer has relevant experience, and the reason for collecting the rating is explicit. If an employer wants to know whether a candidate can listen to a frustrated customer, a general informant score does not show what the candidate did in that situation. The study did not compare informant ratings with observed work behavior or establish that they predict performance. Another person's rating can broaden the view; it cannot act as a truth machine.

Sources: Faking Good on Self-Reports Versus Informant-Reports of Emotional Intelligence

What should the job decide before a test is chosen?

The job should be defined before anyone chooses an EQ test. The U.S. Office of Personnel Management’s “Designing an Assessment Strategy” says assessment tools should be selected in light of job analysis and the intended employment decision. A general score cannot establish by itself that the quality it summarizes is needed for a particular role. Take a customer-support position where a person may need to respond to frustration. “Needs high emotional intelligence” is too broad to guide an assessment. A more useful requirement names what the work calls for: listen long enough to identify the complaint, clarify what happened, explain the available next step, and keep a respectful boundary if the customer continues to be angry. Those are observable actions. They also give the employer a basis for asking whether the role requires them on entry, or whether training and supervision can develop them after hiring. A requirement that can be learned later may not justify screening applicants for a broad personal quality. Job analysis is the process of identifying important duties and the knowledge, skills, abilities, or other characteristics needed to perform them. OPM’s guidance describes using critical duties and competencies to shape an assessment strategy. Without a named behavior, it is difficult to explain what a high or low result means for the work. The label “EQ” does not make that connection automatically. Then ask what evidence supports interpreting the result for that job and decision. In “Designing an Assessment Strategy,” OPM uses validity to describe the relationship between assessment performance and job performance, and notes that the appropriate evidence depends on how an assessment will be used. For a work sample that imitates real tasks, the evidence question includes whether the exercise matches job content identified through job analysis. For a test intended to forecast later performance, evidence would need to connect scores with subsequent work outcomes. These are different claims; showing that a questionnaire consistently produces a score would not, on its own, show that the score supports a hiring inference. This is the point at which a candidate can ask a clear question: “Which job behavior is this assessment intended to measure, and what evidence connects this instrument’s results to that behavior or decision?” OPM’s assessment-strategy guidance also describes considering whether different tools add information about different job-related factors. That is a design principle, not proof that combining measures validates an EQ test or removes error. The employer still needs evidence for the specific tool and intended use. A hiring request that begins and ends with “we use an EQ score” skips the reasoning that should precede selection. Define the duty, make the competency observable, decide whether it is required at entry, and examine whether the named instrument supports the intended conclusion. Without that chain, a score risks standing in for an undefined impression of a good colleague.

Sources: Confirm the Job Analysis and Assessment Strategy; Assessment Strategy

When the concern is behavior, what evidence is closer to the work?

A self-report asks someone to describe a pattern; a work sample asks them to perform a task. When the hiring question is whether a person can listen, clarify a complaint, or set a boundary in a particular role, the second format can make the behavior visible. It still shows performance in one designed exercise, not a permanent trait or a guarantee about every future workday. The U.S. Office of Personnel Management’s “Work Samples and Simulations” guidance describes exercises that mirror tasks employees perform, with scores based on trained assessors’ observations, task outcomes, or both. Its example of handling upset or hostile customers turns a vague label such as “people skills” into observable actions: does the applicant let the customer finish, check what went wrong, explain a limit clearly, and offer an appropriate next step? This is an illustration, not a reported study or sample test item. A role-play can be designed around a job’s demands, while a questionnaire may ask applicants to summarize how they usually respond. Those are different kinds of evidence. The exercise has to fit the job. OPM says work samples should generally be reserved for competencies applicants are expected to bring on entry; if the employer plans to teach the task after hiring, testing it beforehand may be a poor use of the exercise. An employer should first identify the relevant duty and the level of skill needed at entry. A short customer exchange may fit a service role requiring immediate complaint handling; it may say less where employees receive substantial training before doing that work independently. Scoring takes design too. Assessors need shared criteria tied to the exercise, and the task should represent the work closely enough that the observed response has a clear meaning. OPM notes that trained assessors may observe behavior or score outcomes. Without criteria for effective clarification or an appropriate boundary, a role-play can replace a broad EQ label with an interviewer’s unstructured preference. There are costs and limits. OPM notes that these exercises can take time and money to develop and administer, may require individual observation, and can need updating as work changes. A candidate’s response to one scenario remains a sample. Still, when the question concerns a defined interpersonal action, a carefully constructed work sample is closer to that action than a favorable self-description. This is about evidence fit; it does not mean simulations expose dishonesty or predict all workplace conduct. The Office’s “Assessment Strategy” guidance places method choice within a broader process of selecting and using assessments, so one exercise should not carry the whole decision. EQ Test’s Emotional Skills Profile is a private educational reflection tool, not an instrument for hiring, ranking, or selection.

Sources: Work Samples and Simulations; Assessment Strategy

Why is catching a liar the wrong design goal?

A hiring assessment should support a job decision, not identify a dishonest applicant from a score pattern. A favorable response may raise a question about how a result was produced; it cannot establish intent. In “Faking Emotional Intelligence (EI): Comparing Response Distortion on Ability and Trait-Based EI Measures,” EQ-i scores shifted more than MSCEIT scores under simulated applicant instructions. Walker and MacCann’s “Faking Good on Self-Reports Versus Informant-Reports of Emotional Intelligence” found shifts in both self- and informant-ratings under instructed selection conditions. These findings concern study responses. They do not show that a candidate lied or that a flagged answer predicts poor work. A score may reflect an ideal self-description or a person’s view of typical behavior.

That uncertainty makes “catch the liar” a poor foundation for selection. A flag matters only if an employer can explain what it detects, how well it distinguishes response strategies from ordinary variation, and what decision follows. The EI studies did not establish a universal deception threshold. Treating a favorable self-description as proof could turn an uncertain signal into a judgment about character. The better question is whether the test measures a defined competency in a way that supports the employment decision.

The U.S. Office of Personnel Management’s “Designing an Assessment Strategy” guidance starts with the job and its relationship to performance. It says validity evidence depends on how a tool will be used: a work sample that mirrors job tasks raises a different question from a questionnaire intended to forecast later performance. For an EQ measure, an employer should name the relevant behavior, explain why it matters at entry, and show that the instrument supports the inference. A broad claim that emotional intelligence matters at work does not establish that a score should affect hiring.

Consistent procedures and different kinds of evidence can reduce dependence on one response format. OPM describes combining tools that measure different job-related factors as a way to add information. That principle does not show that a combination removes bias, exposes faking, or validates an EQ instrument. A structured interview might examine responses to the same work-related situation; an exercise might let assessors observe a defined behavior. Each needs clear criteria and a reason to be included. This comparison concerns the evidence each method supplies, not which reveals a candidate’s character.

The conclusion would change if an employer could point to validation evidence for the named instrument, role, and decision, based on an appropriate applicant population, and explain consistent administration and scoring. Until then, more elaborate lie detection cannot repair a missing link between a result and the job. A candidate asked to complete an EQ assessment can ask what competency it measures, how the result will be used with other evidence, and how responses are handled.

Sources: Assessment Strategy; Faking emotional intelligence (EI): comparing response distortion on ability and trait-based EI measures; Faking Good on Self-Reports Versus Informant-Reports of Emotional Intelligence

What should you ask when an employer requests an EQ test?

If an employer asks you to take an EQ test, start with the name: which instrument and model is it, and does it ask for self-ratings, another person's ratings, or answers to emotional problems? Ask what job behavior it is meant to assess and what decision the result will inform. In “Confirm the Job Analysis and Assessment Strategy,” the U.S. Office of Personnel Management says assessment tools should follow identified duties and competencies. Its “Designing an Assessment Strategy” guidance says evidence should support the intended relationship between scores and job performance. A broad EQ label does not answer either question. You can ask how the result is combined with other standardized evidence, who can see your responses, and how they are used and retained. These are questions about the process, not an accusation that the test is fake. If the employer cannot explain the instrument or its job connection, ask how the score affects the decision. The EQ Test Emotional Skills Profile serves a different purpose: private individual reflection on emotional habits and responses to authored scenarios. It has no validated norms or established ability-test validity, so it is not a hiring score or selection measure. To examine your own patterns for development, explore the [Emotional Skills Profile](/assessment). Employment use requires evidence supporting the named instrument for that job decision.

Sources: Confirm the Job Analysis and Assessment Strategy; Assessment Strategy

Questions readers ask