Channel · confidence vs. evidence
Lake Wobegon Effect
The tendency for most people to rate themselves as above the average peer on desirable traits and abilities, a logical impossibility for the majority of any group.
Free-run conditions
Across hundreds of studies, a large majority of people place themselves above the median on desirable, common, or controllable traits and skills — most famously driving, where roughly 80-90% of people call themselves safer and more skilled than average. The pattern (more neutrally called the better-than-average effect) is one of the most robust findings in social psychology, with a 2020 meta-analysis of 291 samples and over 950,000 people putting the pooled effect at dz = 0.78. But it is not universal: it reverses to a below-average effect on hard tasks (Kruger 1999) and is far weaker or absent in East Asian samples, which fuels an ongoing debate over whether the cause is motivated self-enhancement or a cold cognitive bias (egocentrism/focalism) in how people process comparisons.
Channel readings
Svenson (1981) — driving skill and safety. A large majority placed themselves above the median; e.g. among US subjects roughly 88% judged themselves safer than the median and ~93% rated themselves above median in skill, mathematically impossible for the group.
Ola Svenson, 1981 · ~161 students (US + Swedish)
Alicke (1985) — trait desirability and controllability. Self-minus-other ratings grew more positive as traits became more desirable and more controllable; foundational demonstration of the better-than-average effect for personality traits.
Mark D. Alicke, 1985 · three samples of college students
Kruger (1999) — 'Lake Wobegon be gone' / below-average effect. Above-average ratings on easy domains but a reliable BELOW-average effect on hard domains; load worsened the bias — consistent with egocentric anchoring on the self and insufficient adjustment for the peer group.
Justin Kruger, 1999 · undergraduate samples (multiple studies)
Zell, Strickhouser, Sedikides & Alicke (2020) — meta-analysis. BTAE is robust across studies with little evidence of publication bias; moderated by trait type and culture.
Ethan Zell, Jason E. Strickhouser, Constantine Sedikides, Mark D. Alicke, 2020 · 124 articles, 291 independent samples, >950,000 participants
Why the signal misleads
Two broad accounts compete. The motivational account says people enhance their self-image to protect self-esteem (self-enhancement). The cognitive account, championed by Alicke & Govorun and Kruger, says the effect arises from egocentrism/focalism: when comparing self to 'the average person,' people overweight what they know about themselves (the focal object) and underweight the diffuse, abstract comparison group. Kruger (1999) showed this predicts a reversal — a below-average effect on difficult tasks — and that cognitive load worsens it, supporting an anchor-on-self-and-under-adjust process rather than pure ego-protection.
Motivated self-enhancement vs. non-motivational egocentrism/focalism; most reviewers conclude both contribute. The cross-cultural pattern (strong in Westerners, weak/absent in East Asians) is itself disputed: Heine and colleagues argue it shows self-enhancement is culturally contingent, while Sedikides and colleagues argue self-enhancement is pancultural but expressed on different, culturally valued attributes.
Calibration verdict
Signal confirmed — the trace settles to a stable, replicated level.
The aggregate better-than-average effect is one of the most replicated findings in social psychology (Zell et al. 2020 meta-analysis: dz = 0.78 across >950,000 people, little publication bias). Specific classics also replicate: Ziano, Mok & Feldman (2021) ran two pre-registered replications of Alicke (1985) with 1,573 MTurk participants and reproduced the desirability and desirability-by-controllability effects, though with a reduced magnitude (replication sr2 = .54, 95% CI [.43,.65], vs. original ηp2 = .78, 95% CI [.73,.81] — note these are different effect-size metrics and are not directly comparable). The important nuance is that 'robust' applies to the effect's existence, not its universality — it reliably reverses on hard tasks (Kruger 1999) and is weak or absent in East Asian samples, so claims of a universal human above-average illusion are NOT supported.
Recorded over-runs
Cannell's standardized-testing scandal: all 50 states 'above the national average' · 1987
West Virginia physician John Jacob Cannell documented in 1987-88 that every US state reported elementary achievement scores above the national norm — statistically impossible. This is the case that gave the effect its name; Cannell traced causes including outdated norms, lax test security and teaching to the test.
K. Patricia Cross: over 90% of college faculty rate themselves above-average teachers · 1977
Cross (1977) reported survey data in which more than 90% of faculty rated themselves above-average teachers and about two-thirds placed themselves in the top quarter — a frequently cited real-world instance in academic self-evaluation.
CEO overconfidence and value-destroying acquisitions · 2008
Malmendier & Tate found CEOs classified as overconfident (over-estimating their own ability to generate returns) make more acquisitions, overpay, and tend to destroy value; merger-announcement market reactions were about -90 bps for overconfident vs -12 bps for others — a documented behavioral-finance manifestation of above-average self-belief.
Damping
Force concrete, distributional thinking: replace 'rate yourself vs. the average person' with an explicit rank ('what % of people are better than you?'), require objective benchmarks or external/peer assessment, and make the comparison group vivid and specific rather than abstract — since the effect shrinks when people focus on the comparison target instead of egocentrically on themselves, and it can even reverse on hard tasks.
Kruger (1999) showed the bias stems from egocentric anchoring on the self and under-adjustment for peers, and reverses to below-average on difficult tasks; making the comparison group salient reduces it.
Reading the trace in the wild
Watch for self-ratings where most members of a group land above the midpoint of a comparative scale — surveys where 80-90% call themselves 'above average,' near-zero people picking 'below average,' or claims of superiority on common, desirable, controllable, or easy attributes (driving, ethics, teaching, getting along with others). The tell is a comparative judgment against a vague 'average person' rather than an objective benchmark.
Adjacent channels
- Better-than-average effectsiblingThe neutral academic name for the same phenomenon; 'Lake Wobegon Effect' is the popular label.
- Illusory superiorityparentThe broad family of overestimating one's standing relative to others, of which the above-average effect is the canonical instance.
- Dunning-Kruger effecteasily-confusedOften conflated, but distinct: Dunning-Kruger is about miscalibration between actual skill and self-assessment (low performers most overestimate); the above-average effect is about comparative ranking against peers.
- Overconfidence effectsiblingExcessive certainty in one's judgments/predictions; overlaps but concerns confidence in accuracy rather than comparative standing.
- Below-average effectoppositeOn difficult tasks the same egocentric mechanism flips people to rate themselves WORSE than average (Kruger 1999).
- Egocentrism / focalismmechanistically-linkedLeading cognitive explanation: overweighting the self as the focal object in the comparison.
- Self-serving biassiblingAttributing successes to self and failures to circumstances; a motivationally related self-enhancement tendency.