PsychBook Research

Collecting and analysing psychological research on the most popular social networking site in the world today.

Add/request/suggest – how many Facebook friends is best?

No more than twice the number archaeologically extrapolated from primates’ brains and but no less than would make a genetically unstable population! Otherwise people will think you are a narcissist or otherwise socially undesirable … which mightn’t be nice

One factor that makes social networking rather different from offline socialising – and what seems to have driven Facebook’s remarkable upward growth – is the facility to publicly advertise the number of ‘friends’ you have. We have all, to a more or less extent, been guilty of the Pokemon phenomenon, catching every person we could, hoping that our ‘network’ would spread around the world, pushing our ‘number of friends’ ever higher, towards the magical upperbound limit of 5,000.

But how many friends is optimal? I started thinking about this after I came across this research from the Facebook Data Team. Basically, they’ve crunched some numbers, and concluded that, whether you have 50, 150, or 500 ‘Facebook friends’, the number of relationships you maintain within that group – be it reciprocally or mutually – is a very small fraction of the total number of friendships. It’s generally around 10% – so if you have 80 Facebook friends, you’re probably using Facebook to keep in touch with about 8 or 9 of them: more if you’re a lassie, less if you’re a lad. Now the Facebook team, and Cameron Marlow have produced some interesting figures with this – but it’s sociological research, and as psychologist, I wanted something more. About the person, not the group. About you. Not me, obviously.

 So I came to research by Christine A. Kleck, Christen A. Reese, Dawn Ziergerer Behnken, and S. Shyam Sundar from 2007, entitled ‘The company you keep and the image you project: Putting your best face forward in online social networks’. This was a fairly interesting study, as the team presented participants with mock Facebook pages, varying elements such as video, picture or text only presentation, and the number of friends. The idea was to see how varying these factor would affect the perceived pleasantness, popularity, confidence, sexiness and attractiveness of the owner of the mock Facebook page. And what did they find?

On the one hand, varying text, picture and video presentation made no difference. But, on the other hand, there was a signifcant relationship for the other factor – “… you are perceived as being more popular, sexy, and attractive and have a higher level of self confidence, when your social network on your Facebook page includes a greater number of friends.”`(Kleck, Reese, Behnken, & Sundar, 2007, p. 21). Alright! Brilliant! So all I have to do is keep adding randoms and I’ll fake it til I make it? right on!

Well, not quite. As has been since pointed out, this research is best described as ‘exploratory’ (academic for ‘half-arsed’). The fact that it’s a conference paper, and not a published article, is the first hint. The main issue, which any ‘average user’ would see, is that the authors varied the number of friends as follows: low, 15 friends; medium, 82 friends; high, 261 friends. While they argue that these were ‘ballpark figures’ that they derived from “discussing the site with colleagues and acquaintances” (Kleck et al. 2007, p. 13, I kid you not), that type of jibberish really doesn’t cut the mustard. I mean, even if 261 was a high number of Facebook friends way back in 2007, you can’t do research with your most important numbers based on watercooler conversations. Hence, we are left with the incorrect conclusion that the number of Facebook friends you have will be directly related to how funky you will look to randoms viewing your profile. Shucks. I do hate to burst that bubble. But I did create it! and besides, there’s more ….

Much more solid research has since been published, in 2008, in a respectable journal, authored by Stephanie Tom Tong, Brandon Van Der Heide, Lindsey Langwell and Joseph B. Walther, the latter of whom is something of a ‘big wheel’ in computer-mediated communication research. Entitled ‘Too much of a good thing? The relationship between number of friends and interpersonal impressions on Facebook’ this article promised much from the outset. And kinda delivered …

In a similar way to the other paper, in this one the authors were also interested in the usual psychological jibber-jabber of ‘sociometric popularity’ and its relationship to social and physical attractiveness and also extraversion. And what did they find? Well, firstly there was no relationship found between number of friends and physical attractiveness – which isn’t really surprising. But what they did find was that there was a curvilinear relationship between number of friends and an individual’s social attractiveness. What this basically means, (as per the title of the article, duh) is that your social attractiveness begins to decline after a certain peak number of friends. And where is this peak?

Well that’s the problem with this research. The researchers used numbers of Facebook friends at 102, 302, 502, 702, and 902 – which were chosen ‘in order to reflect equal intervals amenable to trend analysis’ as well as ‘previous research’ and, of course, ‘informal discussions with Facebook users’ (Tong, Van Der Heide, Langwell & Walther, 2008, p. 540). Why, in the name of all that is holy, why couldn’t they randomise this number? These are computers after all! They can do maths!

But anyway, social attractiveness seems to build from 102 to 302 and then fall off. For some reason, people with upwards of that number are less socially attractive. While the increase up to that point concurs with Kleck et al. (2007), the subsequent drop-off is novel. This is probably a ‘spreading themselves too thin’ effect – the person appears unselective, undiscerning, indiscriminate: the Facebook whore exists! at above 300 friends and rising! In an attempt to get to the bottom of this, Tong et al. (2008) reported two other, minor findings.

With a smaller sample size, they did a similar test, but asked people to explain their impressions. Only a very small minority mentioned the number of friends in their explanation of a profile’s social attractiveness, though the same curvilinear relationship was found. This would seem to imply that, while we do judge a person’s social attractiveness from their number of friends, it is not a conscious process.

The other thing they looked at was a post-hoc analysis of whether or not people were simply rating a profile as socially attractive if its user had the same number of friends as themselves. This might seem obvious, given that 300 was the average number of friends of the participants used in the study. However, when analysed statistically, “… social attractiveness assessments attributable to the number of friends on a Facebook profile are not a significant function of the observer’s own friend count.” (Tong et al., 2008, p. 544). So the number of Facebook friends you see as best has nothing to do with your own number of friends.

So what are we to make of all of this? A bit of context might be helpful.

  • Firstly, from Facebook’s own statistics, the average number of Facebook friends is 130.
  • On a completely different note (just for fun) several years ago wacky human evolutionist R.I.M. Dunbar, (1993) in a paper examining neocortical complexity of primates and humans (honestly!) and also providing a table of ‘Topics of conversation in naturally formed groups in a university refectory’, reached some unusual conclusions. What he basically said was that the size of the brain has grown in tandem with the increase in social groups, which has necessitated the development of language. And by comparing the size of monkey’s brains and social groups to the size of our brains, he reckons that the optimal size of a human group is 148.
  • While I’m at it, I may as well go to outer space. In a paper presented to the American Association for the Advancement of Science, on February 15th, 2002, (though never published and of course no trace remains except the mountains of newspaper reports, the best of which is this) John H. Moore, an anthropologist at the University of Florida, spoke about ‘distant space travel’. What he reckoned was that, in the event of humanity setting off into the great unknown, because of the huge distances to be travelled, not to mention the likelihood of murder, the crews involved would be community-sized, rather than team-sized. Professor Moore’s chosen starting human population is 150-180.
  • Another way of looking at this is from an ecological perspective. Using mathematical modelling to interpret genetic and demographic  variables for the ‘Minimum viable population size in the presence of catastrophes’, Warren J. Ewens, P.J. Brockwell, J.M. Gani and S.I Resnick (1987)come up with a number of their own. I’m not sure that I understand the technical specifics of this one completely, but from what I can gather, all things being equal, their conclusion is that the minimum viable human population in the event of a disaster is 154.
  • And while I’m at it, H. Martin Wobst (1976), in examining ‘Locational relationships in Paleolithic society’, argued that in the Stone Age self-propagating populations had certain limits. Wobst’s (1976) parameter for Paleolithic self-sufficient bands of people is 175-475.

Is there any significance to any of this? Are we evolutionarily, physiologically, sociologically, demographically and genetically – naturally – suited to groups of somewhere between 150 and 180 people? And we’re hard-wired for this? And anything significantly beyond this is odd, artificial, or abnormal?

Well, that’s entirely debatable. Psychologically, it doesn’t seem plausible that we can manage groups of infinite size. At the same time, it’s curious that the average number of Facebook friends is now verging towards the numbers given above. If I was Mark Zuckerberg I would start to get very worried if that number rises much more: this would suggest a flooded market.

And if I were you? Would I be worried by any of this? From Tong et al. (2008) it does seem that a person will be viewed as more ‘socially attractive’ if they have around 300 Facebook friends, but more study is really needed to corroborate this (research grant please!).

But, having examined all of this in some detail, I believe the question should not be ‘how many friends should I have?’ but ‘how many of your Facebook friends are you actually communicating with?’

Categories: Opinion