The research that has been featured in press or mentioned in policy documents.
Wikipedia's page on algorithmic bias pulls from "Who is a scientist? Gender and racial biases in Google vision AI" twice, once in the gender bias section and once in the racial bias section. Both times it's used to back up the same point: Google's Vision AI is far less likely to tag a woman, especially a woman of color, as a scientist than it is a man.
This Science piece draws on "The impact of geographic inequality in federal research funding," a longitudinal comparison of research output in EPSCoR versus non-EPSCoR states, to explore how location and funding geography shape a scientist's productivity over time.
Purdue's Institute for Information Literacy funded a project looking into how Ph.D. students actually judge AI-generated writing when they run into it in their coursework, and what that means for how AI literacy gets taught.
This opinion column argues that academic work needs the internet to stay relevant, framing social media platforms as a bridge between science and society, carrying scholarly work to people who'd otherwise never see it.
This LSE Impact blog post on gendered patterns in political science altmetrics points to "Academic information on Twitter: A user survey" to mark when "academic Twitter" really took off as a tool for sharing and discovering scholarly work, around 2013.
This interview digs into a Learned Publishing study whose headline finding is that most people sharing health and medical research on Facebook aren't academics at all. That has real implications for how we measure whether research is actually reaching anyone.
A European Commission report on open science indicators cites "Can Mendeley bookmarks reflect readership? A survey of user motivations" as part of its case study on Mendeley as a data source for tracking research readership across Europe.
This faculty research story looks at how social bots amplified conspiracy content and baiting language on Twitter after a mass shooting, drawing on an analysis of retweet strategies used to spread that content during the event.
A profile tracing a path from Iran to Wolverhampton to USC, and how that journey shaped both a research focus on scholarly communication and social media and an approach to teaching.
Coverage of a study finding that conspiracy theories following the Parkland shooting spread mainly through retweets by real human accounts, not automated bots, complicating the assumption that bots are the primary drivers of misinformation during breaking news events.
New Scientist's take on the same retweet-strategy analysis, highlighting that ordinary users, not bot networks, were responsible for most of the conspiracy content's reach in the days after the shooting.
Another European Commission report, this one on the future of research metrics, references "Assessing Non-standard Article Impact Using F1000 Labels" as an example of alternative, non-citation-based ways to evaluate the impact of published research.
A UK government report benchmarking the country's research output against other nations draws on "Mendeley readership altmetrics for the social sciences and humanities," using readership data as one measure of research evaluation and knowledge flow.
A European Commission report on how scholars build and showcase their reputation online draws on the lab's work on altmetrics and Mendeley readership data as part of its review of the social networking services researchers use to track and demonstrate impact.