Social Networks & Access to Opportunity

In my work on social networks, one thing that has been of particular interest to me is the way that social network structure seems to be correlated with access to opportunity — both for individuals, and (though I think this is less researched) for businesses. I’m not sure that we could go so far as to attribute causation to this link, but it’s pretty obvious that ‘who you know’ matters.

The immediate spur for this post is a recent article on The Guardian. The gist of the article is that, in spite of Oxford’s attempts to include people from low income backgrounds, there are a host of barriers that stand in the way of inclusion: it starts with local expectations (“… people from ‘x’ don’t go to Oxford…”), runs through the interview process (and the confidence that more expensively taught students exude), and culminates in a lack of “cultural and social capital — the family contacts, exposure to professional networks and carefully honed social skills…” O’Hara’s article concludes by linking outcomes (acceptance to, and success at, Oxford, her alma mater) back to the expectations that are put in place when high-achieving students in weaker state schools are just 14 or 15.

This much of the article is probably fairly obvious to anyone who has thought much about social capital, but there’s an underlying dynamic at work here that I think merits more consideration. A somewhat unwieldily-titled publication (“Behavioral Inference across Cultures: Using Telephones as a Cultural Lens“) by Nathan Eagle in IEEE Intelligent Systems (2008) looked at how phone data could shed light on the sociodemographic characteristics of populations.

What Nathan found in his analysis of data from British Telecommunications plc. is that there is a relationship (R = -0.75, p < 0.001) between the communicational ‘diversity’ of an area and the level of deprivation. Of course, there’s a devil in the detail of how we define ‘diversity of communication’, but the basic idea in the article (as I gather it) is that diversity is captured both spatially (how far do we call?) and, more importantly, socially (how many of the people I call know each other?).

It’s this latter part that is obviously a crucial link back to the issues discussed in the Guardian article: one of the markers of deprivation is not that people in deprived communities talk less or somehow have fewer meaningful relationships than people in less deprived communities; instead, it’s the problem originally identified (using much, much smaller samples) by Granovetter in his seminal paper: “The strength of weak ties” (1973, American Journal of Sociology).

In layman’s terms, what Eagle’s research seems to indicate is that people in deprived areas are very well connected to other people from their same area, while people from less deprived communities tend to have a broader net of social connections (not more, and not better, just wider) which grants them access to opportunities that their closest friends might not know about. It’s the classic cocktail party moment: “Oh, you just have to meet my friend X, they’re in the Y industry you know and were just talking about how they needed someone like you…”

Unfortunately, like a lot of really good research, Nathan’s work has taken him in a very different direction and his ‘preliminary’ analysis has never (to my knowledge) been carried forward into something more definitive. But what both of these articles highlight is the really complex nature of deprivation and how subtle the obstacles to opportunity are for people — and firms! — in these types of environments.

So although I tend to dismiss formal, government-led networking mechanisms in favour of internally-driven, ad hoc approaches, the fact remains that these may well play a crucial role (providing they don’t reinforce prevailing expectations about success, or the lack thereof) in enabling people and companies to overcome persistent ‘barriers to entry’ to the most exclusive, and most successful networks. However, I often wonder if there’s a real world tradeoff between the number of intensely meaningful relationships individuals have, and which seem to be correlated with happiness and contentment, and the number of weak relationships that they have, and which seem to be correlated with employment success and opportunity. Can we really have both?