Retracted Publications Related to Statistics: A Scopus-Based Bibliometric Analysis

Résumé fourni par la source

BACKGROUND: Statistics is a field that provides the methodological basis for modern scientific research, including data collection, analysis, and interpretation. The retraction procedure involves the official removal of a work from the literature due to substantial scientific concerns. Analysis of retracted publications is critical for identifying ethical violations and structural flaws in scientific publishing. METHODS: This study performed a bibliometric and descriptive analysis of publications labeled as "retracted" in the Scopus database, retrieved via a search utilizing the term "statistics" on August 18, 2025. The analysis of documents was conducted based on year, country, journal, keyword, and reasons for retraction. Collaboration networks of countries and authors were displayed with VOSviewer software, and reasons for retraction were confirmed using the Retraction Watch Database. Temporal trends were assessed by linear regression analysis. RESULTS: = 0.030). The most retracted publications came from China (n = 426), followed by India (n = 67), and the United States (n = 49). The Journal of Healthcare Engineering (n = 35) and Journal of Sensors (n = 25) stood out among the journals with the most retracted publications. The most common retraction reasons were investigation by journal/publisher (n = 297), unreliable results and/or conclusions (n = 189), and concerns/issues about referencing/attributions (n = 153). CONCLUSION: The number of retracted publications tagged with the keyword "statistics" has increased significantly over the years. This increase can be attributed to strengthening scientific oversight mechanisms and increased awareness of publication ethics. Increasing the involvement of statistical experts in the peer-review process, improving the statistical literacy of authors, and developing training and mentoring mechanisms to prevent ethical violations are crucial.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.