Duplicate records are a common and often underestimated problem in academic research. Whether you are conducting bibliometric analysis, systematic reviews, or manuscript writing, unremoved duplicates can distort results and weaken methodological rigor.
This guide explains how duplicate articles are identified and removed in Mendeley, with practical steps and best practices that are useful across multiple research workflows.
Who should use this guide?
This workflow is useful for:
- PhD scholars and postgraduate students
- Public health and medical researchers
- Bibliometric and scientometric analysts
- Participants of research methodology and manuscript writing workshops
Why removing duplicate articles is important
When duplicate references remain in your dataset, they can:
- Inflate publication and citation counts
- Bias author productivity analysis
- Distort co-authorship and co-citation networks
- Create errors during reference export to tools like VOSviewer or R
For bibliometric and evidence-synthesis studies, deduplication is a mandatory preprocessing step, not an optional cleanup task.
How Mendeley detects duplicate articles
Mendeley identifies duplicates using metadata matching, not full-text comparison. The key fields it compares include:
- DOI (highest priority identifier)
- Article title
- Author names
- Journal name
- Publication year
- ISSN / PMID, when available
If two or more records show a high level of similarity across these fields, Mendeley flags them as potential duplicates.
Important note:
If metadata is incomplete or inconsistent (common with PDF imports), duplicates may not be detected automatically.
Step-by-step: Removing duplicates in Mendeley Desktop
The Desktop version remains the most reliable option for duplicate management.
Step 1: Open your library
Launch Mendeley Desktop and ensure all references from different databases (Scopus, Web of Science, PubMed, etc.) are fully synced.
Step 2: Check for duplicates
From the top menu, select:
Tools → Check for Duplicates
Step 3: Review duplicate records
Mendeley displays suspected duplicates side by side, allowing you to compare metadata such as title, authors, and DOI.
Step 4: Merge documents
Click Merge Documents to combine records into a single clean reference.
What happens during merging
- The most complete metadata is retained
- PDFs are merged under one record
- Notes and annotations are preserved
- Redundant entries are removed from the library
This results in a single, non-duplicated reference.
Removing duplicates in Mendeley Reference Manager (Web)
Mendeley’s web-based version also offers duplicate detection, but with limitations:
- Duplicates appear under the Duplicates section
- Each merge must be confirmed manually
- Matching accuracy is lower than Desktop
- Metadata control is limited
For large datasets or bibliometric studies, the Desktop version is strongly recommended.
Common duplicate scenarios in research datasets
Duplicates often arise due to:
- Importing the same article from multiple databases
- Missing DOI in one of the records
- Differences in title capitalization or punctuation
- Author initials versus full names
- Preprint and published versions of the same study
These variations can prevent automatic detection and require manual verification.
Limitations of Mendeley for deduplication
While Mendeley is useful, it is not a gold-standard deduplication tool, especially for advanced research synthesis.
Key limitations include:
- No fuzzy or probabilistic matching
- Limited handling of near-duplicates
- No transparency in matching rules
- Not designed specifically for systematic reviews
For large-scale bibliometric or SR/MA projects, Mendeley should be treated as a secondary deduplication step.
Mendeley provides a simple and effective method for basic duplicate removal, but it should be used thoughtfully and in combination with other tools for methodologically rigorous studies.
By following a structured deduplication workflow, researchers can ensure clean datasets, accurate analyses, and credible research outputs.