Retracted Citations and Self-citations in Retracted Publications: A Comparative Study of Plagiarism and Fake Peer Review

2025-02-02

AI Summary

This study addresses the propagation of erroneous citations stemming from retracted publications, systematically comparing citation diffusion patterns between two predominant types of academic misconduct—plagiarism and fraudulent peer review. Leveraging Scopus bibliometric data and Retraction Watch’s manually curated retraction rationale annotations, we employ statistical analysis and temporal trend modeling to quantify differential citation dynamics. Our findings reveal that plagiarized retracted articles accrue, on average, 2.5 times more annual citations than those retracted for fraudulent peer review and exert significantly longer-lasting influence (1.8× longer citation half-life). Conversely, fraudulent peer review cases are detected and retracted more rapidly; 53.6% of their post-retraction citations are directly attributable to the misconduct itself, and subsequently retracted papers exhibit markedly elevated self-citation rates. This work provides empirical evidence and a taxonomy-based governance framework to enable targeted interventions against erroneous citation propagation and enhance the integrity and reliability of the scientific literature.

Technology Category

• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Reasoning under Uncertainty: Causality

Application Category

• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Data transparency and provenance

Abstract

Retracted citations remain a significant concern in academia as they perpetuate misinformation and compromise the integrity of scientific literature despite their invalidation. To analyze the impact of retracted citations, we focused on two retraction categories: plagiarism and fake peer review. The data set was sourced from Scopus and the reasons for the retraction were mapped using the Retraction Watch database. The retraction trend shows a steady average growth in plagiarism cases of 1.2 times, while the fake peer review exhibits a fluctuating pattern with an average growth of 5.5 times. Although fewer papers are retracted in the plagiarism category compared to fake peer reviews, plagiarism-related papers receive 2.5 times more citations. Furthermore, the total number of retracted citations for plagiarized papers is 1.8 times higher than that for fake peer review papers. Within the plagiarism category, 46% of the retracted citations are due to plagiarism, while 53.6% of the retracted citations in the fake peer review category are attributed to the fake peer review. The results also suggest that fake peer review cases are identified and retracted more rapidly than plagiarism cases. Finally, self-citations constitute a small percentage of citations to retracted papers but are notably higher among citations that are later retracted in both the categories.

Problem

• Plagiarism
• Fake Peer Review
• Citation Integrity

Innovation

• Academic Integrity
• Plagiarism and Fake Peer Review
• Citation Analysis