The Persistence of Retracted Papers on Wikipedia

TL;DR

This study examines how English Wikipedia handles citations to retracted papers and finds that a large share of such citations persist without reader-facing corrections for years. By linking Wikipedia revision histories with Retraction Watch, Crossref, Altmetric, and OpenAlex, the authors analyze 1,181 retracted citations and use survival analysis to model time to correction, revealing that 71.6% are problematic and a median correction time of 3.68 years. Corrections are faster when citations are Open Access, receive RetractionBot signals, or have high pre-retraction Twitter activity, and faster on pages with more categorical structure, but slower when the cited work has high academic citation counts. The work highlights a sociotechnical maintenance gap and proposes design interventions—such as integrating retraction signals into editing tools and leveraging structured metadata—to improve citation credibility at scale, with implications for CSCW and knowledge infrastructures.

Abstract

Wikipedia serves as a key infrastructure for public access to scientific knowledge, but it faces challenges in maintaining the credibility of cited sources--especially when scientific papers are retracted. This paper investigates how citations to retracted research are handled on English Wikipedia. We construct a novel dataset that integrates Wikipedia revision histories with metadata from Retraction Watch, Crossref, Altmetric, and OpenAlex, identifying 1,181 citations of retracted papers. We find that 71.6% of the citations were initially problematic and in need of reader-facing repair, defined as those added before the paper's retraction (51.5%) or introduced afterwards without proper warning (20.1%). While many are eventually corrected, our analysis reveals that these citations persist for a median of 3.68 years (1,344 days). Through survival analysis, we find that bot-mediated flagging (RetractionBot), open access availability, pre-existing online visibility (e.g., Twitter/X mention counts), and page-level organization (e.g., number of categories on a Wikipedia page) are associated with a higher hazard of correction. Conversely, a paper's established scholarly authority--a higher academic citation count--is associated with a slower time to correction. Our findings highlight how the Wikipedia community supports collaborative maintenance but leaves gaps in citation-level repair. We contribute to CSCW research by advancing our understanding of this sociotechnical vulnerability, which takes the form of a community coordination challenge, and by offering design directions to support citation credibility at scale.

Figures (8)

• Figure 1: Estimated cumulative and monthly page views of English Wikipedia articles that cite retracted papers without in-text mention of retraction status. Page view data were retrieved using the Wikimedia REST API based on the citation dataset detailed in Section \ref{['sec:data_collection']}. The figure only shows citations with no accompanying retraction-related keywords ("retract", "withdraw", or "invalid") in the main text. This keyword-based detection of uncorrected retraction provides a coarser approximation than the correction categorization used in the main analysis (see Section \ref{['sec:data_collection']} for details on data collection).
• Figure 2: Overview of the data collection and annotation process. The left panel shows the construction of the main dataset, including the extraction of 1,181 retracted paper–Wikipedia citation pairs from Crossref Event Data and corresponding revision histories from Wikipedia. One of the authors manually annotated whether each citation was cited with its retraction status noted at its first appearance and corrected in the latest revision (removal or retraction status noted on the page). The right panel displays the separate dataset of all Wikipedia-cited papers (N = 1,590,982), which is used to calculate citation ratios across academic domains. Retraction status is linked using Retraction Watch, and domains are categorized via OpenAlex.
• Figure 3: Share of problematic (orange) vs unproblematic (green) retracted-paper citations on Wikipedia.
• Figure 4: Comparison of the ratios of retracted and non-retracted papers cited on Wikipedia across academic domains. The retracted ratio represents the fraction of all retracted papers within a domain that are cited on Wikipedia (blue). In contrast, the non-retracted ratio is the fraction of non-retracted papers cited on Wikipedia (red).
• Figure 5: Distribution of corrections of problematic citations on Wikipedia.