David Huffaker Asks, Does (Group) Size Matter?

Huffaker, David A. “The Impact of Group Attributes on Communication Activity and Shared Language in Online Communities.” First Monday: Peer-Reviewed Journal on the Internet 16.4 (2011): n. pag. Web.

According to this study, the answer is no—not online, anyway. In this article, Huffaker discusses the effect of group size on community dynamics, testing some offline theories in an online space.

In 1987, R.E. Rice found that the vibrancy of a community increases with its size, but only to a point; when a group gets large enough, two problems arise. First, the members have much more trouble tracking all the input—this brings to mind the discussion of 4chan, where the number and frequency of posts make it impossible to actually know everything that is happening on the site. Second, the larger a group is, the harder it becomes for an outsider to figure out participation conventions or to integrate into the group. This one seems a bit counterintuitive to me, personally, since you would think that as a group gets larger its conventions would necessarily become more diffuse (doesn’t jargon arise when a small group of specialists need to talk about very specific things?), but perhaps it is related to the first point—that is, maybe the sheer number of posts is what makes it difficult for an outsider to pick out what is important.

Huffaker set out to see whether these conclusions hold water online as well as off-, measuring the presence of the following five characteristics (drawn from Heinz & Rice, 2009) within several Usenet groups:

  • Frequency of interaction (How often do people post, and how many posts are replies?)
  • Shared language (They mean this literally—how much overlap is there in the actual words people are using?)
  • Commitment (Is there a sense of group identity?)
  • Openness (How much are cooperation and sharing a part of the culture?)
  • Trust (Do people generally give others the benefit of the doubt I this environment?)

Looking at this list, it’s striking to me how carefully the Gawker commenting system (Jezebel especially) first cultivated these attributes and then, to mix metaphors, tore away at the foundation they had built. Users are constantly told that there isn’t a group identity to commit to, that their intentions are being questioned by the powers that be, that the shared ways of talking about certain topics are not longer legitimate. They really had created an effective space for community on the sites. Such a shame that they decided it was detrimental to their purposes.

Anyway, in the end, Huffaker found that while these five attributes are certainly found in online communities, unlike in offline communities there is no upper limit on group size—the larger the group, the more the characteristics are evident. “It is as if,” Huffaker writes, “more members mean more opportunities to contribute, respond, discuss, and spread ideas.” While this hardly seems surprising to me, I am happy to see him pointing to persistence as a possible factor in the difference between online and offline groups. When all previous conversations are available and searchable, it is much easier to join. Further, he says, these large groups may not suffer from the lack of cohesion that Rice observed precisely because of the core group of super-committed users who “maintain the group norms and institutional knowledge.”(As a side note, it’s interesting to me that Huffaker determined shared language by measuring the repetition of words or phrases throughout the thread. Under this method, a conversation in which several members request clarification about a particular term would show up as shared language—and the use of an in-joke wouldn’t register at all. A strike in favor of qualitative research!)

The core finding of the study, to me, is this:

Although commitment to the group can lead to its success, the findings here show that high member turnover does not negatively affect the ability of an individual to trigger a response, spark conversation or diffuse language. In light of the group size and participation equality observation, it could be that a subset of users might be making steady impact on the other users, even if they come and go.

In light of the emphasis on retention and commitment discussed by Barbara Ley (and others), these findings are perhaps surprising, especially considering that earlier in the article Huffaker cited several studies showing that heavy turnover crates a drop in participation. In this case, Huffaker points out, only 50 authors (of over 30,000) participated regularly over the study period. But apparently 50 was enough to keep the ball rolling—and to make sure it stays appropriately round.

This brings me again to Jezebel, where the editorial emphasis has turned from creating a tight-knit community of commenters to bringing new eyeballs to the site whenever possible. At a practical level, this has meant an influx of commenters who don’t understand (or don’t care about) the norms and conventions on the site, and many of the core group of frequent posters have either left the site or comment so infrequently that only someone with years of the experience on the site would recognize their names.

And yet, although the rules of decorum have changed, new practices have arisen to take their places, enforced by a different group of commenters. It is no longer verboten to post one’s measurements in the comments, a practice that used to be discouraged as a potential trigger for those suffering from eating disorders, but any post on naming practices has a bevy of commenters poised to debunk the urban legends of La—a, Lemonjello and Oranjello.

The establishment and maintenance of community norms is one of the things that separates the comments on Jezebel from the comments on, say, the Washington Post, in which people tend to post one-off comments or participate in flame wars but not express any investment in the site as a whole.

There may not have been an appreciable drop-off in the number of comments on the articles after Gawker shifted its focus, but I would guess that there are far more commenting accounts now with fewer than five comments than there were when the site began, when the focus was on training a cadre of wits and skeptics to create additional worthwhile content at the bottom of each post. And the open thread, the heart of the community? I wouldn’t call it a ghost town, exactly, but it’s certainly less populated than in the days of yore. Perhaps the threshold of regular users, that critical mass, is different depending on the kind of community or the function of the site.

Works Cited

Heinz, Marni, and Robert E. Rice. “An Integrated Model of Knowledge Sharing in Contemporary Communication Environments.” Communication Yearbook 33 (2009): 134-75. Print.

Rice, Robert E. “New Patterns of Social Structure on an Information Society.” In Competing Visions, Complex Realities: Social Aspects of the Information Society. Eds. Schement, J. and Lievrow, L. Norwood, NJ: Ablex, 1987. 107–120.

This entry was posted in Uncategorized and tagged , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s