Want Readers? ResearchGate vs the Institutional Repository


Here’s a question I get at least a few times every month—I should really start keeping count … it goes something like this: “But I already have a ResearchGate profile, what’s the advantage of keeping other sites about my work up-to-date?” (Sometimes it’s “Academia.edu,” but less and less often on my campus.) It’s a hard question to answer. In part because it assumes so much—that RG is the baseline, that other sites have the same functions, that the advantages are comparable. It’s also a difficult question to answer because it’s often not the real question. I try to be accommodating--RG is a social network & yes, I have a profile too. I also try to remind folks of the unique benefits of a library-supported institutional repository: good metadata, stable URLs, and digital preservation from a trusted, non-commercial provider. I like to say, “RG is fine, but we do it right.” When I’m preaching to the choir (librarians and OA advocates), I get plenty of a-men’s. But when I’m selling the idea to faculty authors, I get a polite OK-but-I’m-busy look.

For the author who has already invested time in creating an RG or Academia.edu profile, the real questions are: What is the return on my investment? What are the rewards of adding an institutional repository (IR) to my dissemination strategy? Are they enough to make it worth my time?

Unfortunately, the last of these questions make this very difficult to answer … your time is your time. I don’t know what you do with it. But, let’s make this a dichotomy. Let’s say you really had to choose between using RG and using your IR.

For the RG fans at my campus, let’s start with the “good” news—if you’re looking for an academic social networking tool, RG beats our local IR, IUPUI ScholarWorks. Most IRs (including ScholarWorks) are not social networks. (On the plus side, this means, among other things, that the IR will not spam your co-authors.)

But if you’re looking for a tool that will help you increase your readership by sharing subscription-free access to full text versions of your works—you want to use the repository. Yes, you can upload your articles, presentations and book chapters to RG, but that doesn’t mean that readers will find them there. (By this point, authors in your RG network have either turned off notifications or they’ve learned to delete those daily emails anyway.) Your library-supported repository probably outperforms your RG profile for readers.

I say “probably” because RG makes its true performance difficult to measure. RG doesn’t provide transparent and easily managed data about how my works are used. I’m not even sure when I uploaded my works and maybe I didn’t … but when did my coauthors upload them. Nor can I look a full count of file downloads and abstract views across a time range. Instead, RG just lumps downloads and views together as “Reads” and (while giving me a cumulative count) only shows monthly counts back to January 2015. Not only does this make it difficult to compare RG performance with other sites, it pretty much disqualifies the use of RG “Reads” in my annual review and in my promotion and tenure dossier. (If I can’t explain it, I’m not going to rely on it.)

I do, however, have access to the complete usage data for my works in IUPUI ScholarWorks. So, here’s an evidence-based anecdote based on the six items that I have duplicated on both RG and IUPUI ScholarWorks. For this comparison, I limited usage data from ScholarWorks to the year 2015, but I let ResearchGate use its cumulative counts of “Reads.” I’m guessing that those counts began sometime in 2013 or 2014. Maybe an average of 18 months. But I know for a fact that they began before January 2015. But, again, who knows … RG doesn’t display an accession date. Because ScholarWorks doesn’t count “Reads,” but does count views (the number of times someone opened the landing page for the item) and downloads (the number of times someone opens the item file, typically a PDF)—I’ve created a “read” count for both sites in the table below. (By the way, I did the same count at Academia.edu … and it just wasn’t even worth talking about. For the same six works, my Academia.edu profile managed to generate a single download and a mere 20 views total.)

Table. ResearchGate "Reads" vs IUPUI ScholarWorks "Reads"

ResearchGate "Reads" RG Item Type of Work SW Item IUPUI ScholarWorks "Reads"
(views + downloads to date, about 2 years)       (reads = views + downloads in 2015)
82 Promotion and Tenure for Community-Engaged Research Journal Article Promotion and Tenure for Community-Engaged Research 59 = 43+16
66 Giving patients granular control of personal health information Journal Article Giving patients granular control of personal health information 73 = 41+32
24 Piloting a Nationally Disseminated, Interactive Human Subjects Protection Program for Community Partners Journal Article Piloting a Nationally Disseminated, Interactive Human Subjects Protection Program for Community Partners 94 = 24+70
24 When Informationists Get Involved Journal Article When Informationists Get Involved 37 = 22+15
19 Points to consider in ethically constructing patient-controlled electronic health records White Paper Points to consider in ethically constructing patient-controlled electronic health records 89 = 56+33
2 Report from the PredictER Expert Panel Meeting Report Report from the PredictER Expert Panel Meeting 48 = 26+22
217 (total)      

400 = 212+188 (total)

So, is it worth your time? I don’t know. Do you want readers or do you want to spam your co-authors? If you want readers, use your institutional repository. In my little anecdote, one year of IUPUI ScholarWorks outperforms all years of RG by 84%. I’d like to see what others have found, but I bet my anecdote will hold up. And if that’s so, don’t waste your time uploading works to RG and Academia.edu. Find your institutional repository and ask a librarian for help.

Jere Odell - Licensed under a Creative Commons Attribution 3.0 Unported License.

Creative Commons License