Optimum group size for humans
In a provocative piece (titled equally provocatively — Your weren’t meant to have a boss — WoW!), Paul Graham makes several interesting points:
What’s so unnatural about working for a big company? The root of the problem is that humans weren’t meant to work in such large groups.
Humans also seem designed to work in groups, and what I’ve read about hunter-gatherers accords with research on organizations and my own experience to suggest roughly what the ideal size is: groups of 8 work well; by 20 they’re getting hard to manage; and a group of 50 is really unwieldy.
Though meant for programming groups and software companies, I think Graham’s piece might also be relevant to research groups, their size, and mentoring. Take a look!
Tags: group size
March 22, 2008 at 12:51 am
The Dunbar number for human beings is around 150, I have heard. Group sizes larger than this tend to break up. Different primates have different Dunbar numbers, linked to the size of the brain. Or, I don’t know if it works the other way, meaning, is the brain size dependant on the Dunbar number.
March 22, 2008 at 1:28 pm
Dear Raj,
Thanks for the pointer; I did not know about Dunbar number till you mentioned. However, since the Dunbar number is meant for social interactions and networking, the actual number for a group of people working together might be smaller than that. In that sense, Graham’s numbers make sense to me. Or, in other words, in a research Department for example, to be in a lab, there might be an optimum number, which Graham gives as 12 or so. On the other hands, for the interactions among different labs to be optimum, one may have to use the Dunbar number you quote, is what I think, which will restrict the total number of labs to be between 10 and 15.
Guru
March 23, 2008 at 10:36 am
Interesting.
In the same paper,(http://www.bbsonline.org/Preprints/OldArchive/bbs.dunbar.html) Dunbar refers to another finding on conversation group sizes ( university pantry, cocktail party,etc), that “the average number of people directly involved in a conversation (as speaker or attentive listener) reached an asymptotic value of about 3.4 (one speaker plus 2.4 listeners) and that groups tended to partition into new conversational cliques at multiples of about four individuals”
March 23, 2008 at 12:17 pm
Dear Raj,
That is interesting; in my own experience, I have found that intense technical discussions are only possible if the total number of (active) participants is at the max, four. However, a slightly larger group is OK provided everybody is not very keen on contributing to the discussions — some are there just to observe. So, Dunbar’s result does seem true. I think it might be worthwhile to check his paper out.
Guru