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.
Comments
7 Responses to “The Dunning-Kruger Effect”
Hi iSci,
The idea of the Dunning & Kruger effect is one that has been brought forth in a few of my courses over the past year. Bridging ideas from each of the courses, Human Perception and Cognition and Art History: Colours of the World, as well as mathematical and statistical components from iSci lectures and Research Project work and ideas brought forth from first year SciLit classes about the trustworthiness of sources, I was able to write this blog post. It was really interesting to look at how this theory was formed and how other papers have both supported and altered it,. I hope you enjoy reading and can’t wait to hear your feedback!
-Sarah 🙂
Hi Sarah,
This was a really interesting blogpost! You did a great job at engaging the reader with your introduction and all the topics flowed really well. I liked how you bought up the more recent work that has been done on this topic and you were able to effectively synthesize a lot of information. I have a few small suggestions:
– In the second test, I think you meant to write ‘standardized’ rather than ‘standardize’.
– In the third sentence in the second paragraph, I just wanted to clarify something. You say that there were four separate studies but state three topics (humour, grammar, logic). Is there another study topic or were the four studies only done on these topics?
– In figure 2, you say that there is a dashed line in the figure but that isn’t really clear. Is there a figure that is a little more clear?
– In your reference list, we usually need to include how we accessed the article in square brackets before the volume number. For example, you could add in [e-journal]
Overall, great blogpost! I look forward to reading the final draft.
-Navneet
Hello Navneet,
Thanks for your comments. I fixed up the few grammatical and reference errors you pointed out, thanks. I clarified the third sentence from the second paragraph to show that two different forms of logic tasks were used. As for the dashed line in figure 2, unfortunately this is the only figure that displays what I am attempting to achieve, however I attempted to clarify which line I was discussing in my figure caption. Hope that helps.
Thanks again for the comments,
Sarah
Hi Sarah,
Super interesting blog post, it peaked my interest because of the connection to PNB 2XA3! I do have a few suggestions though:
– In paragraph two, there seems to be a comma missing in the in-text citation for “(Burson, Larrick, and Klayman, 2006)”
– You should try to stay consistent for capitalizing “effect” whenever you mention the Dunning-Kruger effect, as in the second paragraph you capitalized the word effect
– This might just be a formatting thing but it looks like you centered figure 1 and its caption while figure 2 and its caption was left aligned, so it might be better to stay consistent
Other than that I thought you did a wonderful job and I look forward to reading the final product!
– Yash
–
Hi Yash,
Thanks for your comments. I made sure to insert that missing comma, center my figure caption and ensure effect is left uncapitalized throughout the piece. Thanks for catching all of those!
Thanks again,
Sarah
Hiya Sarah,
Very interesting blog post! I know how excited you are about this content!! I remember you asking my if I thought I was a better than average driver when you were studying for exams, makes sense now!
Here are a few suggestions:
– Not sure if this is the proper use for single quotation marks? ‘experts’ in the first paragraph
– Picky working, change the the to this “Since its publishing, various other studies have investigated the theory.” to “Since its publishing, various other studies have investigated this theory.”
– In the Figure 1 caption, this seems to run on “The circles represent the final selection task presented which was a Wason selection task which is a specific form of logic puzzle.” maybe change to “The circles represent the a Wason selection task which is a specific form of logic puzzle.”
– Also in the Figure 1 caption more information would be useful. Were these tests conducted one after the other? Did each participant do a humour test, then logical reasoning, next grammar and finally the Watson test. This is implied by the language in the Watson explanation, but it could be clearer.
– The original study was published in 2020 but the follow up was 2002?? Maybe make this paragraph clearer, because right now it sounds like there’s been a Back to the Future situation! “When Kruger and Mueller released follow-up research in 2002,” or say “In 2002, when Kruger and Mueller released follow-up research to their 1999 work,”
– I would change your conclusion for better flow “While not completely disregarding the work of Kruger and Dunning, recent research has been able to adapt their initial hypothesis as understanding and statistical methods evolve. 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.” to:
“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 their original research trials. Thus, it is evident that further research is still needed to enhance efficiency through new statistical methods and grow the field of both statistics and psychology.”
Overall terrific work Sarah!
-Clara
Hi Clara,
Thanks for you thorough and detailed comments. They were very helpful! I integrated many of the formatting, clairity, and wording changes you suggested. I hope it makes a bit more sense now.
Thanks again,
Sarah