The emails appear with spammish regularity: “128 people recently read a paper you are mentioned in,” reads the subject line, followed by more clickbait: “A total of 63 papers on Academia mention your name.” The emails are from Academia.edu, the San Francisco-based social network for researchers. “Don’t miss a single citation,” the site warns. Follow the link, and you’re prompted to join the new, $99-a-year Premium service.
Launched a decade ago with venture-capital backing, Academia.edu now claims more than 57 million members and says it attracts more than 36 million unique visitors a month and hosts more than 19 million academic papers — the cornerstone of its campus-conquering strategy.
The service has been called “Facebook for academics,” and the analogy is fitting: Academia.edu mimics core social-media conventions, down to follower counts, curated profiles with pics, and a scrollable “News Feed” bulletin of followers’ uploads. But its reliance on user-facing metrics exceeds anything you’ll find on mainstream social media. It comes with a dedicated “Analytics” page resembling a gaming leaderboard or a corporation’s annual report.
The charts and graphs encourage incessant monitoring, which dovetails with university policies that seek to measure quantifiable “impact.” Scholars are caught in a “metric tide” imposed from above, and Academia.edu and ResearchGate, its rival, make administrators’ desires to quantify research seem achievable.
Scholars broadcast their intellectual status by an array of metrics ... promotional labor that requires time and energy.
The result? Scholars are internalizing an analytics mind-set. The academic reward system already incentivizes quantity over quality, leading to overproduction, p-hacking (mining data to present findings as statistically significant), and self-plagiarism. (As Goodhart’s Law states, “When a measure becomes a target, it ceases to be a good measure.”) The pressure to hoard citations has been imposed by funders and administrators. But now Academia.edu invites us to measure ourselves — to incorporate page views and download counts into our intellectual identities. With the audit culture lodged inside us, we don’t even need a warped incentive structure to overproduce papers and game the numbers.
The site has grown rapidly by preying on scholars’ hunger for visibility. The intellectual respect and influence that researchers crave, for a range of venal and ennobling reasons, are thwarted by paywalled obscurity. The typical journal article, by one estimate, is read in full by just 10 people. Uploading articles on the web, however, allows for broader dissemination.
“Academia is the easiest way to share papers with millions of people across the world for free,” proclaims the site. “A study published in PLOS ONE found that papers uploaded to Academia receive a 69% boost in citations over 5 years.” (What’s unstated on the company’s pitch is that the company’s CEO and six other employees are co-authors of the study.) Who would turn down a 69-percent increase in citations, especially in competitive disciplines where citation ranking comes with material benefits?
Academia.edu has taken a pair of professorial pain points — attention/citation scarcity and closed-access barriers to research — and harnessed one to resolve the other. Scholars upload their paywalled papers, which become Google-indexed PDFs; the company uses these to draw in new members; the audience increases, prompting more scholars to upload their work.
Academia.edu and ResearchGate are essentially peer-to-peer PDF-sharing repositories, akin to Napster circa 1994. In other words, Academia.edu is like Sci-Hub, but with venture backing (and a carefully written, liability-dodging copyright policy). Given the site’s unrelenting appeals for uploads, Academia.edu, and ResearchGate, would appear vulnerable to publisher lawsuits. (Since September, a coalition of science publishers has issued escalating threats to ResearchGate, which has responded by restricting access to nearly two million of its members’ papers; the publishers say they are not satisfied and promise a flurry of takedown notices.)
Academia.edu aims to quantify that gauziest of academic qualities: influence. An academic’s profile includes a “Total Views” tally — the higher the better, is the implication — and, for some, a “top” percentile designation (e.g., “top 5%"), complete with trophy glyph.
Prominent tallies of 30-day profile and paper views and of 30-day unique visitors appear on the Analytics page alongside a color-coded line graph that tracks the same metrics as they zigzag day by day. Granular “user activity” is recorded in a table, with time-stamped rows that log one-off article views by viewer geography, specific paper, and search engine. Users are even periodically alerted by email to the paper views, with subject lines like “Five people searched for you earlier on Google. …"
The profile page also features, in prime real estate, a single-digit number. This is your “AuthorRank,” the service’s algorithmically generated measure of overall influence. It’s the site’s final attempt to measure intellectual impact in digital relief.
At most universities, annual evaluations and tenure decisions hinge on numbers: article counts, journal impact factors, h-index scores, and, perhaps, standardized teaching evaluations. Academia.edu reflects that culture but brings it closer to home. The site’s unrelenting metrical bombardment, email clickbait, and algorithmic ranking invite a graphs-and-figures mind-set.
The academy’s reliance on standardized indexes of “impact” is inherited from the 20th century, when the tenure system evolved to demand, at many institutions, assessments of productivity. Though such bibliometrics are fraught with inconsistencies, the amped-up indexes of accountability demanded by university administrators mean that such metrics have taken on a new urgency.
While this has largely played out on a campus-by-campus basis, Academia.edu brings metrical tracking into a resolutely public arena. Scholars, in maintaining a profile, broadcast their intellectual status as measured by the site’s array of quantified reputation proxies. The visibility of the metrics compels users to tend to their online “brands” — promotional labor that requires time and energy.
Manicuring an intellectual profile takes time away from research and plays into social norms about acceptable self-promotion. Within and beyond the academy, men are more likely than women to highlight their accomplishments, while members of disenfranchised groups are less prone to self-aggrandizement. The implication of those disparities, played out on sites like Academia.edu, is that the status rewards that accompany self-branding are likely to be unevenly distributed.
Take the analogous case of citing oneself: Drawing on a vast data set of academic papers, the sociologist Molly King and her co-authors found that, over the past 20 years, men self-cite 70 percent more than women. Since citations tend to accumulate to the already well-cited, self-citations may help set this cumulative-advantage dynamic in motion. Academia.edu and other social networks have the potential to exacerbate these visibility gaps, especially as its self-feeding dynamics may act as a disparity multiplier.
There are other reasons to be wary of Academia.edu. Its provision of PDF downloads could undercut faculty pressure for genuine open access. Also, its users may soon become its product: Academia.edu has repeatedly hinted that it plans to sell user data to other companies.
But the main problem is what the site is doing to us. Academia.edu generates a data double for every scholar and then asks us to identify with it. In return, we get analytics anxiety and the passing thrill of quantified feedback. At some point, for some of us at least, the numbers are no longer proxies but ends in themselves. The academy has always been a hothouse of invidious comparison, but the site’s metrics encourage finely striated distinctions, drawn with spurious precision. We feel compelled to feed our data doppelgänger just to keep up, baited by the relentless email reminders. Academia.edu is a hard tug on the hamster wheel of university life.
The university is already beset by market pressures and the imperative to demonstrate measurable impact. Scholars around the world experience the market’s impingement on their work lives through enrollment-driven budgets, customer-service teaching, contingent-labor contracts, and mandatory performance assessments. Less obvious, perhaps, is our own internalization of the audit culture’s values, one AuthorRank at a time.
It doesn’t have to be this way. Yes, Academia.edu and ResearchGate moved first and won the network lock-in that scale bestows. But there’s still time to leave — to delete our accounts and go elsewhere. If we can — if we’re tenured and can take the hit in short-run visibility — we should decamp for nonprofit, scholar-run alternatives like the soon-to-launch ScholarlyHub (“For scholars, not profit”) or the Modern Language Association’s Humanities Commons. This isn’t just tilting at market-powered windmills: Scholarly communication really is up for grabs. If we want a service aligned with our knowledge-seeking values, it’s on us to vote with our uploads.
Jefferson Pooley is an associate professor of media and communication at Muhlenberg College. This essay is based on an article, written with Brooke Erin Duffy, an assistant professor of communication at Cornell University, that recently appeared in Social Media + Society.
