The Confidence Paradox: Why Competent People Doubt Most

The most capable person in the room is often the least certain-sounding. That is not false modesty and it is not a communication defect waiting for presence coaching.

The usual explanation reaches for Dunning-Kruger, and that explanation needs handling with more care than it normally gets.

The Problem With the Usual Explanation

Kruger and Dunning published their study in the Journal of Personality and Social Psychology in 1999, and it has since become the most quoted chart in business psychology: novices overrate themselves, experts underrate themselves.

The chart is contested, and the objection is specific rather than vague. Gignac and Zajenkowski argued in 2020 that “regression toward the mean in combination with other cognitive biases, like the better-than-average effect, can explain most of the empirical findings.”

Those two mechanisms are unglamorous and well understood. Regression toward the mean means that if you select people with an extreme score on one measure, their score on a second measure tends to sit closer to average, from measurement noise alone. The better-than-average effect means most people rate themselves slightly above average on most things, which flattens the self-assessment line regardless of actual skill. Put together in a simulation, they produced correlations “almost the same as actual experiments.”

Nuhfer and colleagues went further, concluding “there is no strong tendency to overly positive self-assessment and that the label ‘unskilled and unaware of it’ applies only to few people.”

Dunning has defended the work, and the state of play is unsettled. In the wording of the summary of this debate, the scientific community “remains divided on whether the effect represents genuine psychological phenomena or largely reflects statistical artifacts combined with general overconfidence biases.”

So the mechanism this article would most like to invoke, that incompetence damages the ability to perceive incompetence, is the part under dispute. An earlier version of this piece asserted it as established and called it “the mechanism, and the mechanism is what matters”. That was overstated and has been corrected.

What Survives Without the Chart

Quite a lot, as it turns out, and it needs no curve.

Expertise in a complex domain consists partly in knowing the ways a confident answer can be wrong. A junior analyst presents without qualification because the variables she has missed are not yet visible to her. A senior partner qualifies because two decades of deals have shown her how many apparently solid analyses failed, and where.

That qualification is precision. In a culture that reads confidence as competence, it gets heard as hedging. The observation stands on its own without any claim about what happens inside the novice’s head, which is the part nobody has settled.

The Distinction Worth Keeping

There are two different things that both present as uncertainty, and conflating them is where the damage happens.

The first is uncertainty about the world: outcomes, variables, the reliability of data, how systems behave. This tracks reality, and a sophisticated practitioner in a complex domain should carry more of it than a beginner. It belongs in the communication and in the decision.

The second is uncertainty about oneself: whether you are competent, whether you belong in the room, whether you are adequate. This does not track anything external. It is the same humility pointed at the wrong object. “This outcome is uncertain” has become “I am uncertain as a person.”

The first should be stated. The second should be examined before it is acted on, and the labelling matters here: this publication is in no position to call anyone’s self-doubt a disorder or a distortion, and an earlier version of this article used clinical vocabulary it has no standing to use.

Where the Second Kind Comes From

Crocker and Wolfe gave this a name in Psychological Review in 2001: contingencies of self-worth, the specific domains in which a sense of being worthwhile is staked on how things turn out.

Their account explains the mechanics. In domains of contingency, “people typically have the goal to prove themselves rather than the goal to learn, grow, or improve themselves.” If competence is the domain where worth is staked, then every genuine uncertainty about an outcome arrives as a question about the person holding it. The two kinds of uncertainty are not merely confused; one is manufacturing the other.

They also describe the arrangement as “a trade-off between increased motivation and increased emotional vulnerability”, which is worth holding onto. The sensitivity is attached to the drive, not bolted on beside it.

Clance and Imes described the adjacent experience in 1978 after clinical work with 150 highly successful women who explained achievements away while retaining criticism as evidence of deficiency. They were describing a pattern rather than measuring how common it is, and later work using structured questionnaires found men reporting it with comparable frequency. Claims about how the pattern is distributed across achievement levels need their own evidence, and this article does not have it.

Hewitt and Flett conceptualised perfectionism in 1991 as multidimensional, separating self-oriented perfectionism from other-oriented and from socially prescribed perfectionism, the last being the belief that others demand flawlessness of you. Later research has linked that third form to psychological distress such as depression. For a senior person, the third kind is the one that makes a qualified answer feel dangerous rather than precise.

What It Costs an Organisation

What follows is a reading rather than a finding.

If qualified answers read as weak and unqualified answers read as strong, the selection pressure runs the wrong way. The people best equipped for an ambiguous problem are the ones most aware of why it is hard, and therefore the least likely to volunteer briskly. The brisk volunteers may simply be seeing less of it.

The usual management response makes this worse. Boost presence, work on confidence, communicate assertively. Each treats the tentativeness as the fault, when some of the tentativeness is the most accurate thing in the room.

Four Questions Worth Sitting With

These produce no score and no verdict. They are places where the two kinds of uncertainty come apart.

Where is the worry pointed? When you hesitate over a decision, is the concern that the outcome might be wrong, or that you might be inadequate? They feel similar and have different centres of gravity.
What happens when it goes well? If a good result arrives and part of you immediately assigns it to luck, that is the second kind at work.
Where does attention go before you speak? To the substance, or to how you will be received?
Can you keep feedback on the work? Hearing a remark about an analysis as a remark about your worth is the conflation happening in real time.

And if the self-doubt has stopped looking like a question about work and is affecting how you function, that belongs with a licensed professional. Nothing here substitutes for one, including the assessment below.

A Different Standard for Confidence

What a senior room could reward instead is calibration: saying what you are confident about, naming what you would monitor, and declining to manufacture certainty that the evidence does not support.

“This looks like the right call, and I want to stress-test these three assumptions before we commit” is a more informative sentence than either false certainty or visible self-doubt. It states the epistemic position and leaves the person out of it.

Whether organisations that work this way perform better over time is a reasonable guess and not something this article can demonstrate. What can be said is that the sentence carries more usable information than the alternatives.

Which Kind of Uncertainty Is Yours

Separating uncertainty about outcomes from uncertainty about yourself starts with knowing which domains your sense of worth is staked on. The Executive Identity Profile is a confidential assessment of where your sense of worth currently rests across five dimensions, and which of them is carrying more than it should. It takes about ten minutes, and it opens a conversation rather than closing one. Begin your profile here.

If this was worth your time, there is a letter. One short piece a week on identity, self-worth and the arithmetic of senior work. It costs nothing, and leaving takes one click. See what arrives.

References

Kruger, J., & Dunning, D. (1999). Unskilled and unaware of it. Journal of Personality and Social Psychology, 77(6), 1121–1134. https://doi.org/10.1037/0022-3514.77.6.1121
Gignac, G. E., & Zajenkowski, M. (2020). The Dunning-Kruger effect is (mostly) a statistical artefact. https://doi.org/10.1016/j.intell.2020.101449
Crocker, J., & Wolfe, C. T. (2001). Contingencies of self-worth. Psychological Review, 108, 593–623. https://doi.org/10.1037/0033-295x.108.3.593
Clance, P. R., & Imes, S. A. (1978). The imposter phenomenon in high achieving women: Dynamics and therapeutic intervention. Psychotherapy: Theory, Research & Practice, 15(3), 241–247. https://doi.org/10.1037/h0086006
Hewitt, P. L., & Flett, G. L. (1991). Perfectionism in the self and social contexts: Conceptualization, assessment, and association with psychopathology. Journal of Personality and Social Psychology, 60(3), 456–470. https://doi.org/10.1037/0022-3514.60.3.456

The Confidence Paradox: Why the Most Competent People Feel the Least Certain