If I were to ask you how good of a driver you think you are, or what score you think you would receive when given a standardized test, what would you say? Now if you were to actually undergo said driving or standardized test, would your score align with what you inferred? Probably not. This concept is known as the Dunning-Kruger effect and states that most people tend to overestimate their own abilities, and others, who are “experts” on the presented topic will underestimate their own abilities (Dunning and Kruger, 1999; Sawler, 2021).
The Dunning-Kruger effect results from a combination of individuals not being able to make competent decisions and subsequently not being able to realize their faulty choices (Dunning and Kruger, 1999). Founded by David Dunning and Justin Kruger in 1999, the hypothesis is a highly cited theory that has been used as the basis for many psychological studies. Through four separate studies regarding humor, grammar, and two forms of logic-based tasks, the pair determined that the bottom quartile of subjects consistently overestimate and the top quartile consistently underestimate their abilities (Figure 1) (Dunning and Kruger, 1999; Burson Larrick and Klayman, 2006). Since its publishing, various other studies have investigated this theory. One study published in 2018 used the Dunning-Kruger effect to explain anti-vaccine policy attitudes by showing how anti-vaccine supporters were also individuals that were overconfident in their knowledge regarding autism (Motta et al., 2018). However, since the development of the Dunning-Kruger effect, many studies have also noted that while reasonable in some scenarios, the effect of this theory is much less drastic than previously reported (Gignac and Zajenkowski, 2020).

In 2020, Gignac and Zajenkowski began to dismantle the Dunning-Kruger Hypothesis by identifying its failure to take into account how the regression towards the mean plays a role (Gignac and Zajenkowski, 2020). When Kruger and Mueller released followed-up on Kruger and Dunning’s 1999 research in 2002, they did account for the regression towards the mean, however they then neglected to factor in the fact that an individual’s perception of themself is not uniform across populations (Kruger and Mueller, 2002; Burson, Larrick, and Klayman, 2006; Gignac and Zajenkowski, 2020). While including these confounding variables is necessary, they largely reduce the efficiency of the test, which is not ideal for research (Gignac and Zajenkowski, 2020).
Another study by Burson, Larrick and Klayman (2006) identified the role of task difficulty and its effect on the Dunning-Kruger Hypothesis. They proved that difficult tasks are difficult to all individuals, and easy tasks are easy to all individuals, regardless of cognitive ability. Thus, higher performers would be well-calibrated for easy tasks and poorly-calibrated for difficult tasks, and vice versa. Without taking into account the net task-induced bias of their trials, Kruger and Dunning’s work cannot be completely relied upon (Burson, Larrick and Klayman, 2006). If bias was considered, results would either deviate more from lower quartiles (positive net task-induced bias) or deviate more from higher quartiles (negative net task-induced bias) (Figure 2).

While not completely disregarding the work of Kruger and Dunning, recent research has been able to adapt their initial hypothesis to evolving theories and statistical methods. As noted, many of the excluded variables work to reduce the efficiency of research trials. Thus, it is evident that further research is needed to enhance efficiency through new statistical methods and grow the field of both statistics and psychology.
References
Burson, K.A., Larrick, R.P. and Klayman, J., 2006. Skilled or unskilled, but still unaware of it: How perceptions of difficulty drive miscalibration in relative comparisons. Journal of Personality and Social Psychology, [online] 90(1), pp.60–77. https://doi.org/10.1037/0022-3514.90.1.60.
Gignac, G.E. and Zajenkowski, M., 2020. The Dunning-Kruger effect is (mostly) a statistical artefact: Valid approaches to testing the hypothesis with individual differences data – ScienceDirect. [online] 80. Available at: <https://www-sciencedirect-com.libaccess.lib.mcmaster.ca/science/article/pii/S0160289620300271> [Accessed 30 Jan. 2022].
Krueger, J. and Mueller, R.A., 2002. Unskilled, unaware, or both? The better-than-average heuristic and statistical regression predict errors in estimates of own performance. Journal of Personality and Social Psychology, [online] 82(2), pp.180–188. https://doi.org/10.1037/0022-3514.82.2.180.
Kruger, J. and Dunning, D., 1999. Unskilled and Unaware of It: How Difficulties in Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments. [online] p.14.
Motta, M., Callaghan, T. and Sylvester, S., 2018. Knowing less but presuming more: Dunning-Kruger effects and the endorsement of anti-vaccine policy attitudes – ScienceDirect. [online] 211, pp.274–281. https://doi.org/10.1016/j.socscimed.2018.06.032.
Sawler, J., 2021. Economics 101-ism and the Dunning-Kruger effect: Reducing overconfidence among introductory macroeconomics students. International Review of Economics Education, [online] 36, p.100208. https://doi.org/10.1016/j.iree.2020.100208.