A monthly overview of recent academic research about Wikipedia and other Wikimedia projects, also published as the Wikimedia Research Newsletter.

Reviewed by Czarking0 and Tilman Bayer

Times may be changing on our project. A recent publication in the Proceedings of the Royal Society highlights how representation of female tenure-track biologist has surpassed their male counterparts.^[1]

Using a recently generated census of 2104 female and 3721 male tenured and tenure-eligible faculty members affiliated with the biology departments of all United States R1 universities, we examined the extent to which biology faculty are covered on Wikipedia. Women in the census were significantly less likely to be featured on the platform than their male peers until 2018. However, the trend reversed in 2022, and women were 25% more likely than men to have a Wikipedia biography in 2024. In addition, women's biographies tend to be longer than those of men.

As we have noted here at various times before (not just regarding gender biases, but also e.g. regarding political biases), gaps in coverage are not necessarily due to biases on behalf of Wikipedia itself (in the sense of editors "putting the thumb on the scale"). In this case though, while trying to understand "the causes of this reversal", the authors plausibly point to changes on Wikipedia itself, namely the activities of the WikiProject Women in Red, which began counteracting the article gap in 2015, and related efforts:

Among the 163 women’s biographies created since 2015, 78 were created by Wikipedia editors that were affiliated with the Women in Red project (figure 5 [reproduced below]), 16 were created by editors affiliated with the Women Scientists project (however, 14 of these were created by individuals that were also part of the Women in Red project), and 13 were created by users registered as students enrolled in university courses that require/encourage contributing to Wikipedia.

The researchers also obtained various other variables as part of their census of professors, and analyzed their relation to the likelihood of having a Wikipedia biography:

Among faculty with Wikipedia biographies, women had significantly fewer publications and citations, lower h-indexes and shorter careers (i.e. they started publishing more recently [...]), suggesting that the research output required to be covered on Wikipedia is higher for men than for women. The university ranking was not significantly different between women and men with Wikipedia biographies

A few further statistical analyses controlling for these academic metrics indicated that

[...] gender has no effect on the number of views (p = 0.600), the number of edits (p = 0.488) or the number of edits per year (p = 0.485), indicating that women’s Wikipedia biographies are viewed and edited as often as those of men with the same academic rank and a similar affiliation, seniority and research output. However, gender does have an effect on biographies’ lengths (p = 2.66 × 10−5), and a separate linear regression analysis for women’s and men’s biographies indicates that, given any h-index, women’s biographies tend to be longer than those of men [...].

Reviewed by Tilman Bayer

Another recent study^[2] focused on the coverage of German scholars on the German Wikipedia and similarly found that

Surprisingly, gender had a positive and significant effect: female professors seem to have a higher chance of having a Wikipedia biography than their male counterparts (B = 0.340, p = 0.024).

This effect was determined to be concentrated in STEM disciplines:

Female professors in STEM fields had a higher predicted probability of having a Wikipedia biography than their male colleagues. Outside STEM, no gender effects were observed.

Similar to the above study of US biology professors on enwiki, the authors started with a systematic census of their target population, and also examined the (not necessarily causal) effects of other scholarly metrics on the likelihood of being covered on Wikipedia:

Our study is based on data from an online survey of 1839 professors in Germany. Four binary logistic regression models and predicted probabilities were calculated to test our hypotheses. The results indicated that the number of published articles (OR = 0.999; p = 0.110) and awards (OR = 1.007, p = 0.781) did not predict having a Wikipedia page, while press articles had a positive but small effect (OR = 1.005; p = 0.005). Among all variables aiming to capture notability, a professorship at a university of applied science (UAS) was most powerful and had a significant negative effect (OR = 0.176; p < 0.001). Individual activities, such as social media activity (OR = 1.248; p = 0.002) and networking (OR = 1.433; p = 0.012), had a positive effect.

Unlike the above US-focused Proceedings of the Royal Society paper, the authors here did not attempt to track changes in the German Wikipedia's coverage over time, or to assess the potential impact of editing campaigns that target gender gaps. Those are only mentioned in passing in a section titled: "What can scientists do to get a Wikipedia biography?":

[...] extended professional networks could also have a direct impact on Wikipedia biographies, as they increase the likelihood of being portrayed as part of Wikipedia editing campaigns (such as edit-a-thons). This may be especially helpful for female professors, as there are Wikipedia networks such as “FemNetz” and “WomenEdit” in Berlin as well as “Who writes his_tory?” in Switzerland, which are committed to writing biographies about women on Wikipedia [...].

Other recent publications that could not be covered in time for this issue include the items listed below. Contributions, whether reviewing or summarizing newly published research, are always welcome.

Compiled by Tilman Bayer

From the abstract:^[3]

"we quantitatively explore both (a) how scholarly biomedical research is cited in biomedically-focused Wikipedia articles, and (b) whether this reflects the topical patterns of those articles' readership in Wikipedia. Our methods identify quantitative avenues for improved alignment between editors' limited effort and readers' focus, with respect to both the biomedical topics themselves and the caliber of biomedical research cited. In other words, readers focus on topics that are very human-focused, but editors focus their citation work on basic science topics (e.g. molecular and cellular biology), leading to disproportionate citation rates of basic research publications for biomedical topics in Wikipedia."

See also our coverage of earlier research that similarly found a "Misalignment Between Supply and Demand" of Wikipedia's content

From the abstract:^[4]

"We explore the influence and interconnectivity of philosophical thinkers within the Wikipedia knowledge network. Using a dataset of 237 articles dedicated to philosophers across nine different language editions (Arabic, Chinese, English, French, German, Japanese, Portuguese, Russian, and Spanish), we apply the PageRank and CheiRank algorithms to analyze their relative ranking and influence in each linguistic context. Our findings suggest that while philosophers, and their associated schools of thought, do not serve as the foundational building blocks of the Wikipedia knowledge structure, they remain critical reference points within it. The top ten articles devoted to philosophers for each language edition is determined. We observe significant cultural variations in the ranking of these figures across language editions, with Western philosophers, in particular, being disproportionately represented among the most influential."

See also our earlier coverage of related research: "Wikipedia-based graphs visualize influences between thinkers, writers and musicians"

From the abstract:^[5]

"Wikidata [...] features a long-tail of items with limited data and a number of systematic gaps within the available content. In this paper, we present the results of a systematic literature review aimed to understand the state of these content gaps within Wikidata. We propose a typology of gaps based on prior research and contribute a theoretical framework intended to conceptualise gaps and support their measurement. We also describe the methods and metrics present used within the literature and classify them according to our framework to identify overlooked gaps that might occur in Wikidata. We then discuss the implications for collaboration and editor activity within Wikidata as well as future research directions. "