The positive illusion effect is definitely real— most drivers think they're above average etc.— but I think that there's a little more going on in this particular example:
"For example, a 2018 study asked US Air Force Academy cadets to rate their leadership skills compared to their fellow cadets on a scale of 0 to 99. The average rating was 80, significantly above the 50 that would be expected if the cadets had been unbiased."
My intuition is that 17 year old air cadets are probably more egotistical than average. As the paper cited suggests: "Inflated leadership ability self-assessment might also be associated with dark-side personality traits related to narcissism". But also, although the study would have been anonymised, I wouldn't necessarily expect a teenager in the military to feel fully assured that the results wouldn't be seen by commanding officers, which could lead to more boosterish self-reporting.
I assume if that were a problem, the psychologists who conducted the study and their peer reviewers would have caught it. Anyway, I link to a meta-analysis of 124 better than average effect studies, so you can probably find one that satisfies you.
Like I said, I wasn't doubting the effect, just saying there's probably more going on in this particular instance. The meta-analysis makes three related points: the effect is bigger in younger people, the effect bigger among Americans, and the methods of the study impact the size of effect (and can even reverse it under certain conditions).
The positive illusion effect is definitely real— most drivers think they're above average etc.— but I think that there's a little more going on in this particular example:
"For example, a 2018 study asked US Air Force Academy cadets to rate their leadership skills compared to their fellow cadets on a scale of 0 to 99. The average rating was 80, significantly above the 50 that would be expected if the cadets had been unbiased."
My intuition is that 17 year old air cadets are probably more egotistical than average. As the paper cited suggests: "Inflated leadership ability self-assessment might also be associated with dark-side personality traits related to narcissism". But also, although the study would have been anonymised, I wouldn't necessarily expect a teenager in the military to feel fully assured that the results wouldn't be seen by commanding officers, which could lead to more boosterish self-reporting.
I assume if that were a problem, the psychologists who conducted the study and their peer reviewers would have caught it. Anyway, I link to a meta-analysis of 124 better than average effect studies, so you can probably find one that satisfies you.
Like I said, I wasn't doubting the effect, just saying there's probably more going on in this particular instance. The meta-analysis makes three related points: the effect is bigger in younger people, the effect bigger among Americans, and the methods of the study impact the size of effect (and can even reverse it under certain conditions).