People who perform worst at a task are often the least accurate judges of how poorly they performed and may be the most certain they did well. This is known as the Dunning–Kruger effect.
Confidence is sometimes treated as a shortcut for competence. The person who speaks first, speaks crisply, and sounds certain often becomes the default “expert in the room.” It’s an uncomfortable truth that confidence and competence are only loosely related.
The cost of mis-calibrated confidence isn’t just bruised ego. It’s flawed strategy, overcommitted plans, under-tested assumptions, and teams that follow the loudest voice rather than the best thinking.
What the Dunning–Kruger Effect Really Is (and Isn’t)
The original evidence came from studies where participants completed tasks (such as tests of logical reasoning, grammar, or humor), then estimated how well they had done relative to others. Those who scored in the bottom quartile dramatically overestimated their performance; those who scored near the top were more accurate and sometimes slightly underestimated themselves (Kruger & Dunning, 1999). Later work extended and refined these findings, showing that mis-calibration is not just about inflated self-esteem. It is often about limited self-insight (Ehrlinger et al., 2008).
Note the effect is about judgment accuracy, not arrogance. Many low performers aren’t boasting; they simply don’t possess the knowledge required to evaluate quality. The most useful managerial interpretation is not “the incompetent are deluded.” It is that in situations where standards are subtle, feedback is delayed, and performance is hard to measure, mis-calibration is likely and it tends to be worst among those who most need correction.
Why It Happens
The most influential explanation is often called the dual-burden account. That is to say, the skills needed to perform well are frequently the same skills needed to evaluate performance. If you lack those skills, you not only make mistakes, you also fail to recognise them (Kruger & Dunning, 1999; Dunning, 2011).
Consider a senior leader reviewing a strategic deck. To assess the quality of the analysis, you need to notice missing variables, weak causal logic, non-comparable benchmarks, and overconfident inference. If you don’t have those mental models, the deck may feel “tight” because it’s fluent and coherent and your internal alarm system never triggers.
People vary in their ability to monitor their own knowledge, detect uncertainty, and revise beliefs in light of evidence. Recent research has explored “second-order judgments”, i.e. how confident people are about their own confidence. It turns out that mis-calibration is particularly pronounced among low performers (Coutinho, 2024). Put bluntly: some individuals don’t just have wrong answers; they have weak internal signals that their answers might be wrong.
The Organizational Multipliers That Make It Worse
Three multipliers show up repeatedly in practice.
1) Weak or ambiguous feedback loops
In many leadership roles, feedback is delayed, noisy, political, or absent. If a strategic decision “works,” we rarely know whether it was skill or luck. If it fails, the causes are contested. This is fertile ground for overconfidence, especially when people can craft narratives that protect self-image.
2) Social reward for certainty
Cultures that equate decisiveness with competence select for confident displays. Over time, people learn that doubt is penalized and certainty is promoted. The result is not better judgment. It’s better performance of judgment. A subtle but important difference.
3) Relative comparisons that distort confidence
Confidence is not formed in a vacuum. Confidence isn’t only based on your own skill. It’s influenced by who you're comparing yourself to. Research shows that even when people get irrelevant or misleading comparisons (like being told they're above average when they're not), their confidence can increase anyway. Over time, this false boost can "snowball," leading someone to feel more confident without actually improving. In business, this might happen when a leader frequently works with less informed teams — they seem smart by contrast, and their confidence increases, even though their own ability hasn't improved.
Why High Performers Sometimes Underestimate Themselves
The mirror-image of Dunning–Kruger is also important: skilled people can be cautious. They often have richer mental models of the domain, which means they see complexity, ambiguity, and the many ways a plan can fail. They may also assume that what is obvious to them is obvious to others, leading them to understate their advantage.
In organisations, the danger is predictable: the people with the best judgment may speak with the least certainty, while the least reliable judgments arrive packaged as clarity.
Practical Tools to Reduce Miscalibrated Confidence
The Calibration Loop
Before showing results, ask people to predict outcomes and assign a confidence level (e.g., 60/80/95%). Then compare prediction to reality and track calibration over time.
Calibration improves when people are required to make explicit judgments and receive feedback on accuracy. Intervention research shows that training protocols involving confidence judgments plus feedback can shift confidence and reduce bias (Engeler et al., 2020).
In project reviews, require a one-slide “forecast” before the update: delivery date, cost range, key risks, confidence.
In hiring panels, ask each interviewer: probability this candidate will be top quartile in role after 12 months.
In strategy, ask: probability this initiative hits target by quarter X.
Over time, you’ll see who is systematically miscalibrated, overconfident or underconfident and you’ll normalize learning rather than blame.
“Show Your Work”
The Dunning–Kruger dynamic thrives when quality is hard to observe. Making evaluation criteria explicit reduces the “unknown standards” problem that low-skill evaluators face (Kruger & Dunning, 1999; Dunning, 2011). It also helps high performers articulate what they see implicitly.
Use “Decision Quality Checklist”:
- What would have to be true for this to work?
- What evidence would change our mind?
- What are the base rates / reference class?
- What’s the strongest alternative explanation?
- What are we least certain about?
“Consider the Opposite”
Mis-calibration is reinforced when people only search for confirming evidence. Structured disagreement forces leaders to engage with disconfirming data and surfaces gaps in understanding. It also reduces the likelihood that confidence is mistaken for correctness.
For major decisions, run a 20-minute “opposite case” segment. Rotate the role so it’s not a personality trait (“the skeptic”) but a process. End by updating confidence levels rather than declaring winners.
Don’t Weaponise the Label
The Dunning–Kruger effect is often used as an insult. That is a mistake, strategically and ethically. First, because the phenomenon is partly a function of the environment: unclear standards, poor feedback, and cultures that reward certainty. Second, because all of us are novices in some domains. The moment you step into unfamiliar territory — a new market, technology, regulation you are temporarily vulnerable to mis-calibration. Treat confidence as a hypothesis, not a signal of status.
Ⓒ Dr Dominic Irvine 2026
References
Coutinho, M. V. C. (2024). Unskilled and unaware: Second-order judgments increase with miscalibration for low performers.
Dunning, D. (2011). The Dunning–Kruger effect: On being ignorant of one’s own ignorance. In Advances in experimental social psychology (pp. xx–xx). Academic Press.
Ehrlinger, J., Johnson, K., Banner, M., Dunning, D., & Kruger, J. (2008). Why the unskilled are unaware: Further explorations of (lack of) self-insight among the incompetent. Organizational Behavior and Human Decision Processes, 105(1), 98–121.
Engeler, N. C., Gilbert, S. J., & Fleming, S. M. (2020). The effect of metacognitive training on confidence and strategic behavior. PLOS ONE, 15(10), e0240858.
Gignac, G. E., & Zajenkowski, M. (2020). The Dunning–Kruger effect is (mostly) a statistical artefact: Valid approaches to testing the hypothesis with individual differences data. Intelligence, 80, 101449.
Kruger, J., & Dunning, D. (1999). Unskilled and unaware of it: How difficulties in recognizing one’s own incompetence lead to inflated self-assessments. Journal of Personality and Social Psychology, 77(6), 1121–1134. https://doi.org/10.1037/0022-3514.77.6.1121
Lebuda, I., et al. (2024). No strong support for a Dunning–Kruger effect in creativity. Scientific Reports, 14, 61042.
Magnus, J. R. (2022). A statistical explanation of the Dunning–Kruger effect.
Martin, M., et al. (2025). Calibration feedback with the Practical Scoring Rule: Does it improve confidence accuracy?
Murad, Z., et al. (2021). Confidence snowballing and relative performance feedback. Journal of Economic Behavior & Organization, 191, 1–xx.
Stone, E. R., et al. (2000). Training to improve calibration and discrimination: The effects of performance and environmental feedback. Organizational Behavior and Human Decision Processes, 83(2), 282–309.