Can a High EQ Score Coexist With Harmful Workplace Behavior?

Can a high EQ score coexist with harmful workplace behavior?

Yes. A high result on a named EQ instrument can coexist with harmful conduct at work. The score describes performance or responses under that instrument’s conditions; it does not certify ethical behavior, and it cannot establish whether a particular exchange caused harm or what someone intended. It may still tell you something useful about the ability, habits, or self-perceptions the instrument actually measures. To interpret a workplace concern, first ask what produced the score and what the test asked its respondent to do. A useful result can guide a person toward a skill to examine or practice, provided its meaning stays tied to the instrument. When someone reports a troubling exchange, the score should not replace the account, records, or other evidence about what occurred.

What does a high EQ score describe?

Before treating a high EQ score as evidence about conduct, identify how the instrument produced it. The title “EQ test” does not tell you whether a person solved emotion problems, described their usual behavior, or received a broader competency profile. In “The Measurement of Emotional Intelligence: A Critical Review of the Literature and Recommendations for Researchers and Practitioners,” the authors distinguish ability, trait, and mixed approaches while noting that their labels and measures overlap. The practical distinction is about the task and the inference it permits, not a contest to name one approach the best.

An ability test asks someone to work through emotion-related problems. A result therefore reflects their responses to those particular tasks under the test’s scoring method. It can be relevant to how they perform on that kind of assessment; it is not a transcript of how they acted during a disagreement with a colleague. A person might identify a likely emotion in a written situation and still respond poorly when challenged, tired, rushed, or responsible for a decision. The item samples an answer to a problem; it does not observe the later interaction.

Trait measures take a different route. They commonly ask respondents to describe how they usually feel, think, or behave, so a high score reflects reported tendencies or self-perception. That report may help someone notice the habits they believe they have, but it depends on what they recognize and are willing or able to report. Agreement with statements about staying calm or understanding others does not independently verify how those habits appeared in one specific exchange. The person is reporting a general pattern, not submitting a record of a dated workplace event.

Mixed approaches can combine self-reported traits with competencies, and some include ratings from other people. Their breadth may suit a broader development question, yet the combined score still needs to be read according to its components and method. An observer rating describes how a rater sees behavior; it does not automatically settle what happened in a disputed incident. Nor does putting several kinds of information under one profile make them interchangeable.

These approaches are not sealed categories: theories and instruments can share concepts even when they collect answers differently. The useful question is what evidence sits behind the number. If it comes from emotion-task answers, interpret it as performance on those tasks; if it comes from a questionnaire, interpret it as reported typical behavior; if it combines inputs, inspect what each contributes. None of those readings turns a score into direct evidence about a particular disagreement. That requires attention to the exchange itself, which the following research examines through the outcomes its studies actually measured.

Sources: The Measurement of Emotional Intelligence: A Critical Review of the Literature and Recommendations for Researchers and Practitioners

What does the workplace manipulation study actually show?

The study titled “The dark side of emotional intelligence: the role of gender and the Dark Triad in emotional manipulation at work” surveyed 765 employees from varied occupations. Its publisher abstract describes questionnaire measures of emotional intelligence, Dark Triad traits, and willingness to use manipulation at work. That wording matters: respondents supplied reports about their willingness, rather than researchers documenting a series of workplace exchanges and independently deciding whether manipulation occurred.

The abstract says emotional intelligence contributed to reported disingenuous manipulation in a multiple regression analysis. This is a result about variation in questionnaire responses after the study's regression model considered its measured factors. It does not tell us that employees with higher EQ committed more manipulative acts, how often any act happened, or whether a colleague experienced harm. The abstract does not provide an incident count, observation protocol, or independent verification of conduct; those details cannot be filled in from the finding.

The distinction is especially important because a willingness response and an observed act sit at different points in a chain. Someone may endorse an item about being prepared to use a disingenuous tactic without using it in a particular situation; conversely, a person may behave harmfully without endorsing a generalized statement about willingness. The survey captures what respondents said about a tendency under questionnaire conditions. It does not follow them through a disagreement, deadline, or power imbalance to record what they chose and what happened next.

The result still has a defensible use. It indicates that, in this employee sample and under the study's measures, reported emotional intelligence contributed statistically to stated disingenuous manipulation. That association makes the possibility worth examining rather than assuming that emotional skill always serves considerate aims. But regression does not turn an answer into a witnessed event, and the abstract does not establish that emotional intelligence produced the willingness. It also gives no basis for estimating how common such willingness is among high-scoring workers as a whole.

So the study helps formulate a research question about reported tendencies; it cannot answer whether a specific person manipulated a colleague. A self-report measure can be informative about what respondents recognize or say they might do. For a claim about behavior at work, the missing step is evidence of the interaction itself: what was communicated, what options people had, and what followed. Those are different observations from a survey response, and the abstract reports no such observations.

The wording also keeps the study's scope honest. The abstract confirms the employee count, varied occupations, questionnaire approach, and regression finding; it does not expose enough full-text method detail to reconstruct every item, sampling step, or model specification. A careful account should stop at what is reported rather than infer the exact scenarios respondents imagined. That restraint preserves the finding while avoiding a stronger claim than the accessible record supports.

Sources: The dark side of emotional intelligence: the role of gender and the Dark Triad in emotional manipulation at work

Why do the manipulation results point in different directions?

The apparent disagreement becomes easier to understand when the meta-analysis, “A Meta-Analytic Investigation of the Relationship Between Emotional Intelligence and Emotional Manipulation,” is read by the names of its measures. It combined 24 studies with 5,687 participants, but did not pool every outcome under one undifferentiated label. It compared different emotional-intelligence measures with distinct emotional-manipulation scales. The resulting pattern is not one universal EQ-to-manipulation relationship; it changes with what researchers called manipulation and how they measured emotional intelligence.

For self-reported ability emotional intelligence, the meta-analysis found a small positive association with the EMS non-prosocial manipulation scale, with a corrected association of about .10. The word “non-prosocial” marks the scale's harmful or self-serving direction; it is part of what makes this pairing different from a measure of emotional influence in general. Even here, the result describes an association among measured constructs across the included studies. It does not mean a given person has a ten-percent chance of manipulation, nor does it count harmful episodes in workplaces.

The trait-EI comparisons with the MEOS looked different. Trait EI was positively associated with MEOS Enhance, about .40, and Divert, about .34, while its association with Worsen was negative, about -.16. These are scale-specific associations reported by the meta-analysis, not three interchangeable readings of one behavior. MEOS distinguishes ways of managing another person's emotions: Enhance and Divert are treated as prosocial forms in the analysis, while Worsen is the non-prosocial factor. A broad label such as “emotional manipulation” can therefore make a supportive mood shift and an attempt to worsen someone's mood sound like the same outcome when the scales separate them.

The labels do not settle every real-world moral question. A person could describe an effort to lift a colleague's spirits as helpful, while the colleague might experience unwanted pressure; a scale score cannot resolve that particular exchange. The point is narrower: researchers operationalized outcomes differently, and those definitions help explain why the associations point in different directions. The EMS result concerns its non-prosocial category. The MEOS pattern separates Enhance, Divert, and Worsen rather than treating all influence as harmful. Combining them verbally into a single “manipulation” outcome obscures the distinctions that produced the findings.

That distinction also changes what can be inferred from a positive association. The positive MEOS associations with Enhance and Divert cannot be recast as evidence that trait EI predicts coercion simply because both sit under a broad emotional-management umbrella. Nor does the negative MEOS association with Worsen establish that people with higher trait EI never behave harmfully. The scales describe reported patterns, and the meta-analysis summarizes relationships between those scale results. It does not inspect a particular conversation to determine whether influence respected another person's choice.

The comparison is useful precisely because it resists a headline that treats all EQ and all manipulation as single quantities. A small positive EMS association and differing MEOS associations can coexist when the EI construct and outcome category change. The meta-analysis supports that measurement-sensitive reading; its results should remain attached to the named scales rather than be compressed into a verdict about emotionally intelligent employees. The contrast identifies how studies define their outcomes, not what any one worker did.

There is also a boundary to the synthesis: the meta-analysis assembled associations from studies of measured constructs, not verified workplace incidents. Its systematic differences are informative and worth preserving, but they cannot establish a causal direction or classify an individual act. In practical terms, the evidence says that the answer depends partly on which emotional-intelligence measure is paired with which manipulation scale. Whether a specific influence was supportive, unwelcome, or coercive still requires attention to the conduct and circumstances, beyond a score on either scale.

This is why the meta-analysis should be used to refine the question, not supply a shortcut answer. Its scale pairings show where the broad term splits; they do not tell a manager how to classify a disputed interaction before examining what took place.

Sources: A Meta-Analytic Investigation of the Relationship Between Emotional Intelligence and Emotional Manipulation

How might emotional skill be used harmfully without making harm inevitable?

The strongest defensible explanation is conditional: emotional skill may give someone information about another person’s feelings or likely reactions, while the purpose and use of that information depend on other choices. The review “Does Emotional Intelligence have a ‘Dark’ Side? A Review of the Literature” describes an emerging literature on possible negative outcomes of both trait and ability emotional intelligence. Its abstract characterizes proposed pathways as conditional and calls for stronger methods that account for context. That is a cautious review of possibilities, not a demonstrated sequence in which greater EQ produces manipulation or misconduct. It does not provide a rule for identifying which individual will use emotional skill well or badly.

One plausible mechanism, offered here as editorial inference rather than a result established by that review, is that recognizing distress or anticipating a reaction can widen a person’s options. They might use that awareness to make a difficult exchange clearer, reduce avoidable pressure, or check whether the other person wants support. The same awareness could help someone select a more persuasive moment or frame a request to secure agreement. The skill supplies information; it does not by itself specify an ethical objective. What matters in the interaction includes the purpose being pursued and whether the other person can understand the request, consider alternatives, and make a meaningful choice.

The response to hesitation or refusal can make that distinction more observable. A person who notices discomfort might pause, clarify what is optional, adjust a request, or accept a no. Someone seeking compliance might instead use the signal to increase pressure, obscure alternatives, or attach consequences to refusal. These are possible uses of interpersonal information, not a diagnostic checklist: one behavior alone does not reveal motive, and context can change what an action means. Harmful behavior can also arise for reasons unrelated to emotional skill. The review supports treating a dark-side pathway as a question for contextual research; it does not show that EQ caused harm or predict an individual’s conduct. A plausible mechanism therefore helps explain how capacity could be used in different ways, while evidence about the actual conduct remains necessary to judge a case.

This account also keeps ability and responsibility separate. Noticing a likely emotional response may improve prediction, but prediction is not permission to exploit it, and difficulty reading a situation does not excuse a harmful choice. In practice, the useful question is whether the person used what they noticed to preserve the other party’s agency or to narrow it. That judgment concerns actions and their consequences, not a presumed inner motive inferred from a score. It also leaves room for explanations the available evidence cannot distinguish: deliberate pressure, a poorly handled request, or a conflict shaped by constraints unrelated to EQ. The review identifies a possible research problem; only contextual evidence can resolve a concrete case.

Sources: Does Emotional Intelligence have a ‘Dark’ Side? A Review of the Literature

How does the broader workplace evidence complicate the dark-side claim?

A Bangladesh survey points in a different direction for a broader workplace outcome. “Emotional Intelligence and Counterproductive Work Behavior - An Empirical Investigation for Asia” reports a purposive sample of 412 respondents drawn from manufacturing, merchandising, financial, and service institutions in Bangladesh. Participants completed the Emotional Quotient Index and a 22-item measure of counterproductive work behavior. The article reports negative correlations between five emotional-intelligence dimensions and that conduct measure, ranging from -.51 to -.55. In its regression analysis, the authors report that the dimensions explained 35% of variation in counterproductive-work-behavior scores within this sample. Those are questionnaire associations in the surveyed group, not observed workplace incidents.

The outcome matters as much as the direction. Counterproductive work behavior is a broad category of workplace conduct; it is not the same measure as a respondent’s stated willingness to use disingenuous emotional manipulation, or a scale separating different forms of emotional influence. The Bangladesh result therefore does not directly replicate or overturn the manipulation findings. A person could score on a broad self-reported conduct measure for behaviors that do not involve reading or steering another person’s emotions. Conversely, a willingness-to-manipulate response does not establish broad counterproductive conduct. The studies ask different questions, so their associations need not cancel each other out.

The survey’s setting and recruitment also bound what its coefficients can say. Purposive sampling means the 412 respondents were selected through a non-random approach; the result cannot be treated as an estimate for every Bangladeshi employee, much less a universal workplace relationship. The measures were questionnaires, and the study was correlational. Its regression describes how the measured dimensions related to variation in reported counterproductive behavior under that model; it does not show that emotional intelligence reduced harmful acts or caused a change in conduct. Work conditions, reporting choices, and other unmeasured factors could matter, but the survey does not settle their contribution.

The reported 35% also needs its denominator kept visible: it refers to variation in the study’s counterproductive-behavior scores among these respondents, as modeled from the measured dimensions. It does not mean that EQ prevented 35% of misconduct, that the dimensions explain 35% of any employee’s behavior, or that an individual score predicts a future incident. The correlations summarize how questionnaire scores moved together in this sample. They do not tell whether a given respondent harmed a colleague, how often a particular act occurred, or whether changing a dimension would change conduct.

Read together, this study and the manipulation research support a narrower conclusion than either a dark-side headline or a protective-EQ claim. Direction depends partly on what investigators measured: broad counterproductive behavior in one survey is not interchangeable with manipulation willingness or specific emotional-influence scales elsewhere. The Bangladesh findings are relevant counterevidence to a claim that higher EQ generally predicts every form of workplace harm, yet they cannot establish a general protective effect. Likewise, the other questionnaire associations cannot establish that emotional skill causes harmful conduct. Each finding stays attached to its measure, sample, and design; neither substitutes for evidence about a particular workplace exchange. This distinction is essential when comparing studies: coefficients belong to the outcomes and samples that produced them, not to EQ in the abstract.

Sources: Emotional Intelligence and Counterproductive Work Behavior - An Empirical Investigation for Asia

What does a workplace exchange show that a score cannot?

A workplace exchange can show what information and choices were actually available. Consider this invented illustration: a senior coworker asks a junior coworker to take on extra work before a deadline. To understand how much room the request left for choice, first establish what the senior coworker explained: the deadline, the work involved, which existing priority could move, and whether another person or a later handoff was an option. Without that information, the junior colleague cannot compare the request with current responsibilities.

Power over consequences matters too. The senior coworker may assign work, influence evaluations, or control access to an opportunity; those facts affect the practical cost of saying no. A request can leave formal permission to decline while making refusal costly through an implied penalty. Conversely, a real deadline or limited staffing may constrain the options without making the request a deliberate attempt to control someone. The relevant record is what authority each person had and what consequences were stated or reasonably attached to the decision, not an assumption based on job title alone.

Timing can sharpen the account. Note whether the junior coworker learned about the deadline before accepting the extra task or only after the commitment, and whether the senior coworker knew about competing assignments. A later change in scope or timing may alter what a workable choice looked like. Where accounts differ, contemporaneous messages or task records can clarify sequence; they may still leave the meaning of an unrecorded exchange uncertain. Keeping chronology separate from interpretation makes the questions more precise.

Next, look at whether the junior coworker could ask questions, identify conflicting priorities, propose an alternative, or refuse without being cut off. Then consider what followed hesitation: did the senior coworker clarify the tradeoff or revisit priorities, or did the available options narrow after the hesitation? These observations help reconstruct the choice as it happened. They do not reveal private intent by themselves; the same strained response can have more than one explanation, and a single exchange may be incomplete without its surrounding pattern.

This is the practical difference between a score and an account of conduct. A score cannot supply the missing deadline, authority, alternatives, or consequences. Record those details, separate what was directly observed from what is inferred, and compare accounts where possible. An awkward request or tense response alone does not establish manipulation. The point is to ask whether meaningful choice existed in that setting and whether the response to hesitation preserved or restricted it, while leaving any motive conclusion open until relevant evidence supports one.

What evidence should guide a workplace response to a report?

When someone reports harm, the response should begin with the reported conduct and relevant evidence, not either person's EQ result. In the United States, the Equal Employment Opportunity Commission's “Small Business Fact Sheet: Harassment in the Workplace” advises employers to consider a prompt, impartial, thorough investigation of harassment allegations. The EEOC says the steps, evidence to gather, and people to interview depend on the facts and circumstances of the particular allegation. That is guidance for U.S. workplace harassment matters; it does not make every harmful, unfair, or upsetting act unlawful harassment, and it should not be treated as a universal procedure for every country or workplace concern.

A high score cannot be used to discount a report or establish that the person named could not have caused harm. The same score also cannot tell an investigator what happened. A report deserves attention as an account that needs a fair response; by itself, it does not settle another person's intent or the facts of a disputed event. The investigation should test the specific concern through relevant material and accounts, rather than interpreting a personality or emotional-skills profile as evidence for or against either party.

The EEOC's case-specific approach has a practical consequence: gather material that can clarify this allegation instead of relying on a fixed checklist detached from it. As practical, non-legal guidance, preserve relevant messages, schedules, documents, and other records in their original context; note when and where the reported events occurred; and distinguish what someone directly saw or heard from conclusions they drew. If accounts differ, record the difference accurately rather than smoothing it into agreement. Follow the applicable workplace policy and reporting process, including any stated route for raising concerns or preserving records. A useful note can keep each event separate: describe the action, the date or approximate timing, the people present, and the record that may help clarify it. Preserve the original message or document rather than relying on a cropped excerpt, and avoid adding conclusions to a contemporaneous record. If a detail is uncertain, mark it as uncertain instead of filling the gap from memory. These are practical ways to keep relevant material usable; the workplace policy and the allegation determine what should be collected and who should handle it. When possible, keep a note of the question that remains unresolved, so follow-up can address it directly. Avoid circulating sensitive material more widely than the policy or process requires.

A careful record helps keep observation and interpretation distinct. For example, a dated message may establish that a task was assigned and when; it may not, on its own, establish what alternatives were explained in a conversation or how a consequence was intended. Mark what the material directly shows, what witnesses recall, and what remains uncertain. Those distinctions make it possible to ask focused follow-up questions without treating confidence, distress, composure, or an EQ result as a credibility shortcut.

Fair fact-finding takes a report seriously while checking the claim against relevant evidence. Its purpose is to understand what conduct occurred and how the circumstances bear on the allegation, not to award a psychological label. The EEOC guidance is specifically about U.S. harassment allegations; other concerns and jurisdictions may call for different procedures. For any report, the workplace's applicable policy and the evidence about the actual events should guide the response.

Sources: Small Business Fact Sheet: Harassment in the Workplace

What can a development profile help someone do next?

The Emotional Skills Profile has a narrower job than deciding who was right in a difficult exchange: it gives an individual a structured prompt for examining their own emotional habits and choosing something to practice. Its 32 questions cover recognition, understanding, regulation, and responding to others. That makes it relevant when the question is personal—how do I notice tension, make sense of it, manage my response, or answer another person?—rather than evidentiary: what exactly did a colleague do, and what should a workplace conclude?

The profile keeps two kinds of reflection distinct. Recent-behavior summaries concern how someone describes their own patterns; authored scenario judgments concern how they respond to situations presented in the assessment. A person can use both as prompts for attention, while keeping the questions separate: what do I tend to do, and what response seems appropriate in a described situation? The difference can help turn a broad wish such as ‘be better with people’ into a more observable practice area, without treating either summary as a record of a particular meeting.

This is an educational reflection profile, not a standardized ability measure. It has no validated norms or established ability-test validity, so its result should not be read as a population standing or proof of how someone will perform under pressure. The product facts describe what the assessment covers and how its report separates the two kinds of responses; they do not establish that a score predicts workplace conduct. Its practical value lies in the questions a person chooses to examine, not in a verdict attached to a number.

Suppose someone wants to handle disagreement more constructively. They might choose one behavior within their control: pause before replying to criticism, ask a clarifying question before defending a decision, or state a boundary together with the reason and an alternative. The behavior should be small enough to notice in a real conversation. Afterward, the person can record what they tried, what response followed, and what they would adjust next time. This is a self-directed practice loop, not a test of whether a complaint is true or whether another person is trustworthy.

A practice target also needs a clear cue. ‘Stay calm’ is difficult to observe and can become a demand to hide a reasonable reaction. ‘When I notice I am interrupting, let the other person finish and summarize the point before I answer’ describes an action that can be reviewed afterward. The cue should fit the person's actual goal and circumstances; the profile can prompt reflection, while the person decides which behavior is relevant and feasible in their work.

Specific feedback can add another perspective when the person wants it and the relationship makes asking appropriate. A useful request is bounded: ask what was observable in a recent exchange, such as whether the response left room for disagreement, rather than asking for a global judgment of character. The other person may have only one perspective, so feedback is material to consider, not an official score. A person can compare it with their own recollection and decide what to practice, while a reported incident continues through the relevant workplace route. To explore that personal reflection tool, visit the Emotional Skills Profile at /assessment.

Sources: Small Business Fact Sheet: Harassment in the Workplace

Choose the next step that fits the question

Start with what needs attention now. If you are raising a specific incident or a repeated pattern, use the applicable workplace reporting route and follow its policy. Keep the concern with the people or process responsible for addressing it; an EQ result cannot establish what happened or determine the response. For serious or recurring concerns, that route matters especially, and local policy may set steps or timelines. If unsure which channel applies, consult the policy or ask the designated contact how to raise the concern.

If your aim is your own development, name one behavior you can observe in yourself: asking a question before disagreeing, pausing before responding to sharp feedback, or explaining a boundary clearly. Choose a situation where you can try it, then notice what you did and what you might adjust. The private Emotional Skills Profile can offer structured prompts across recognition, understanding, regulation, and responding to others. Explore it at /assessment if that would help you choose a practice focus. The next step depends on whether you need a concern addressed or a behavior to work on.

Questions readers ask