Whose networks? Whose wealth?

A good book raises at least as many questions as it answers. By focusing on an issue that does not really get addressed by Yochai Benkler in the Wealth of Networks, I do not mean to suggest that the author should have written the book I would like to see written (in fact, I am working on it so it is just as well:). Rather, I would like to suggest some issues that are worth considering in the context of this discussion.

Before I present my comments, I would like to note that I very much enjoyed reading this book and think it will appeal to and be of value to a wide range of people. I like the mix of tackling large important questions while getting into the nitty-gritty of what is happening on the ground and illustrating points with very detailed and careful examples. The book serves as a helpful reference to various online discussions and events that have occurred in the past few years and it is useful to have them documented and linked together suggesting that they were more than isolated occurrences of how people are using digital technologies for political and social purposes. The other contributions of this seminar tackle the big questions addressed directly in the book. I have decided to address an aspect of the issue that is not addressed in detail yet I believe needs to be part of the conversation.

The point I want to focus on here is the unequal distribution of opportunities discussed in the book and what potential consequences this may have. That is, to what extent are the benefits of the possibilities raised by recent innovations distributed equally among different segments of the population and to the extent that they are not, what implications might that have?

To be sure, the author addresses this point in Chapter 7 as one of the expected critiques of the enthusiastic claims. Under the heading “Digital divide” (pp.236), the author acknowledges that distribution of digital media is not equal in society. He counters this concern by noting that (1) access differences in the US are much less significant than in the 1990s; (2) “growth rates among underrepresented groups are higher than the growth rate among the highly represented groups” (p.237); (3) “the democratizing effects of the Internet must be compared to democracy in the context of mass media, not in the context of an idealized utopia” (p.237). While these claims may all be true, I will address them here point-by-point, because I think they still leave room for discussion and concern.

First, while it is true that differences in access to the Internet by various population segments in the U.S. (and many other countries) have declined since the 1990s, we do not have much evidence showing continued strong growth. In fact, some data suggest that active home use is in decline in several nations, including the United States. While home use is not the only type of access to the Internet, it is not hard to argue that in most cases it is likely the most beneficial type of access as in many cases it will mean the highest level of autonomy in using the medium. That is, having to rely on connectivity at a library or school that is only accessible during limited hours, is located miles away, does not offer much privacy and may be limited by filters is a far cry from 24-hour access in the privacy of one’s home. I do not mean to exaggerate this point and do realize that for some people using the Internet at home may afford less privacy than otherwise, but those are likely the less frequent cases. Some work I have done suggests that home use is a statistically significant predictor of how knowledgeable people are about the Internet, which then predicts the diversity of uses to which they put the medium. All this implies that differences in access and access type continue to exist despite some changes over ten years.

Second, it is not saying too much that underrepresented groups are getting connected at higher rates than those who are already online in high numbers. After all, groups whose members are already online at 80+% only have so much further to go while groups whose members are connected at 15-20% have the potential to grow considerably. In many ways this is just a numbers game, and numbers can be interpreted in lots of ways. So while the rate of increase may be higher for an underrepresented group, that does not automatically result in high penetration levels or even a much better relative position. (For a detailed discussion of this point, take a look at Figures 5 & 6 and the accompanying discussion starting on p.15. of this paper.)

The third point regarding the relevant baseline comparison of the traditional media landscape versus some utopia seems less problematic, and it is certainly a point I like to call attention to as well when people challenge the limits of how digital media are being used today. That said, in the context of discussing inequalities, I think this point deserves a pause as well. The assumption seems to be that thanks to new opportunities, everybody will have more of a voice and input. However, is it not possible that inequalities may actually increase if some people are much more likely to take advantage of these opportunities than others? That is, if those who are already in more advantageous positions (e.g. higher socio-economic status) are more likely to use new tools and if they do so in ways that do not benefit everybody, then the disengaged may be left behind even more.

Instead of focusing simply on access figures, I have argued for years that we need to pay just as much attention to differences in uses and abilities. To this, we can add the notion of differences in participation. There are very different ways of being an Internet user and being counted in baseline statistics of the connected. However, large chunks of people who are online never use their connectivity for certain types of activities such as political engagement.

My work focuses on empirical studies of people’s Internet uses so the above comments are not hypotheticals about differences in abilities and types of uses. I have published several papers that discuss skill differences among users. I have also recently collected some data – not yet published – about college students’ Internet uses. This data set is based on surveys administered on 1,300 college students’ (mostly first-years) in February and March of this year. (Click here for some sample descriptives.) Students are a representative sample of those attending the University of Illinois Chicago campus.

It is hard, if not impossible, to come up with just one or two survey questions that would capture the extent to which young people – the most connected age group – are taking advantage of the medium in democratizing ways. Nonetheless, there are some revealing and helpful measures. For example, when asked how often they visit “blogs, discussion forums or other sites that allow you to interact with other participants”, the topic category of “politics, economics, law, policy” got the lowest rating among types of interactive sites students visit (the other options of sports, technology, arts & culture, and personal journal all ranked higher in popularity). Only five percent visit such sites daily and 63.5 percent never visit interactive sites on the topics of politics, economics, law or policy. When asked whether they have ever visited certain sites – undoubtedly the most conservative measure of any level of engagement with these blogs – only one percent say that they have ever visited Instapundit or Daily Kos. These participation rates are so low that it is not even possible to try to determine what types of students are most likely to engage in such online activities.

When looking at use of the Internet for different types of activities, it turns out that certain college students use the medium for statistically significantly fewer activities than others. For example, women, African-Americans, students whose parents have lower levels of education and those with lower levels of knowledge about the Internet do less online than others. So even if diffusion figures increase and access gaps are closing, use gaps remain and are not randomly distributed.

While Benkler is right to be enthusiastic about some of the opportunities made possible by new tools and services in the digital landscape and the possible implications of their implementations, it is important to put all this in the context of actual usage figures. There is no evidence to suggest that equal diffusion of all services to a broader public is simply a matter of time. So while it may be the case that a larger number of people get to participate in political and other social action, how much flattening can we expect if the already privileged continue to be the ones most likely to embrace and thus potentially benefit from the new opportunities?