I Wanna Be a Rock Star

Today's edition of The New York Times has an article on the Social Science Research Network, a free online repository of research abstracts and papers in, you guessed it, social science (economics, finance, marketing, accounting, and the like). Specifically, the article is on the site's tracking of most downloaded papers, and the academics' semi-obsessive checking of their work's rankings, in Amazon.com fashion. The SSRN's search page provides the following statistics: 189,040 abstracts, 151,824 full-text papers, 94,764 authors (who would have thought there were so many researchers in social science! it's about the population of Albany, New York!) The site received 20,241 papers in the last six months, and has a grand total of  21,633,245 downloads since its creation (yes, over twenty-one million), with over five million alone (5,351,717) occurring in the last twelve months and almost 600,000 (597,112) in the last thirty days. That is a daily average of 20,000. Not only are there a lot of authors, but there are a lot of readers as well, or so it seems.

Many of those will, of course, be professors and graduate students, eager to access the latest research in their field without having to email authors for a preprint, although the Times also points out some papers are of higher quality than others and, since there is no peer review process, the reader must judge the merits of the paper by himself. (Incidentally, a February 2008 article in the Times of Higher Education, focused on British students, starts with the headline: "Social science PhD students short on quantitative skills, study finds". Let's hope the SSRN website allows users to rate the papers.) The New York Times points out that "The research network raises the same big questions about what is lost and what is gained by removing the barriers to being heard in the public square," compared to users posting their music on MySpace and bloggers vs journalists. But the difference is that researchers do not post their papers online as a replacement to a more traditional career; instead, they aim at speeding the dissemination time of their work. No self-respecting researcher will post a half-finished paper that could be read by academics in his field, scholars he interacts with at conferences, and, in contrast with songwriting or blogging, it is unlikely that many laypeople are writing economic analyses in their spare time. Indeed, while there might be some "vanity press" effect in posting documents online, most of the documents I found by running searches on random keywords were written by university students and professors. (The Times mentions some papers are simply "musings", but gives no example. The most off-beat, and one of the most downloaded, posting appears to be a "a detailed examination of one particular curse word in the law." You can read the abstract here. Apparently, over 22,000 people have found this fascinating, and downloaded the paper to read further. I haven't, along with about 260,000 people who viewed the abstract but did not download. I guess there's a potential marketing study hidden in there, about which titles draw people to read the abstract and turn them away.)

While the website is touted as a major innovation, it simply does faster what had been done before with print journals: communicate information to other academics, while providing non-academics with the thrill of peeking into the world of research. What I find most interesting is the amount of data on readership that it provides. For instance, when you list the Top 50 most downloaded paper with "finance" in the title, abstract, or keywords, the crowd's favorite, by the University of Chicago's economist Eugene Fama, totals 65,246 downloads, but the 50th most downloaded paper, by Zvi Bodie at Boston University, was only downloaded 3,015 times, and that's already very high. Did 65,246 people truly read Fama's paper, or did they just browse through the first few pages and give up? Many papers are downloaded a more reasonable 30 or 50 times. The Times quotes a law scholar as saying: "If I could pick a certain 20 people to read my article, that would mean more to me than 20,000 others who read it." Downloads are certainly an imperfect proxy for influence, but one can imagine many more statistics, from paper ratings to "favorite this paper" (to use a term Technorati uses for blogs) and "favorite this author". SSRN will certainly add a few as its influence grows and some of its 550,000 users request more guidance. After all, Amazon.com lets users publish ratings, and blogs let readers post comments.