Only accepted work is recordedAboutGuidesContactSign in
CitePepCitePep
The Dunbar Number: Understanding Social Circle Limits

The Dunbar Number: Understanding Social Circle Limits

Corrected by Emir Baycan · on When Notes Fly · 15 November 2026 · View published page ↗

Dunbar's number is contested: a 2021 reanalysis put the confidence interval from roughly 2 to over 500, leaving 150 statistically indistinguishable.

Factually incorrect

The exact change

Before
A 2016 study published in Royal Society Open Science analyzed 3.3 million Twitter users and found that the emotionally close network (the people with whom users had consistent, mutual, substantive exchanges) had a median size of approximately 100 to 200, fully consistent with the Dunbar prediction. | A 2020 study published in Proceedings of the National Academy of Sciences by Dunbar and colleagues analyzed mobile phone data from 35 million users across Europe. | People contacted their innermost circle (around 5 individuals) at rates roughly 10 times higher than their third layer (around 50 individuals). The frequency gradient matched the theoretical prediction from social circle theory remarkably well.
After
A 2011 study published in PLoS ONE analyzed the conversations of 1.7 million Twitter users and found that users could maintain a maximum of roughly 100 to 200 stable relationships, fully consistent with the Dunbar prediction. Dunbar's own 2016 study in Royal Society Open Science, surveying more than 3,300 British adults, found average Facebook friend counts of about 155 and 183 in its two samples. | A 2016 study published in Social Networks by Pádraig Mac Carron, Kimmo Kaski, and Dunbar analyzed mobile phone call records from about 35 million users in a European country. | The call-frequency data clustered into layers averaging about 4, 11, 30, and 129 people, close to the theoretical prediction from social circle theory.

Suggested change

Corrected 3 figures to the published source.

Why this is better

Each study and survey figure in the article was checked against the published source. Figures that differed from the source were corrected to the source's own numbers, dates and attribution.

More by Emir Baycan in Psychology Behavior

All of Emir Baycan's contributions →