Patterns of text reuse in a scientific corpus

Area of Science:

• Bibliometrics
• Scholarly Communication
• Research Integrity

Background:

• Text reuse, or plagiarism, is a concern in academic publishing.
• Understanding the prevalence and impact of text reuse is crucial for maintaining research integrity.

Purpose of the Study:

• To quantify the incidence and patterns of text reuse in scientific articles.
• To investigate the relationship between text reuse and research impact.
• To identify factors associated with high levels of text reuse.

Main Methods:

• Systematic pairwise comparison of text content for all articles on arXiv.org (1991-2012).
• Classification and measurement of three types of text reuse.
• Analysis of text reuse distribution among authors and countries of origin.

Main Results:

• Text reuse is globally prevalent, with varying frequencies across different classes.
• A subset of authors exhibits chronic text reuse.
• Higher amounts of reused text in an article show a negative correlation with its citation count.
• Geographic distribution of articles with substantial text reuse was analyzed.

Conclusions:

• The study establishes a baseline for text reuse in scientific literature, which appears permissive.
• Aberrant authors engaging in excessive text reuse were identified.
• Significant text reuse may negatively impact a research paper's influence and visibility.