New Metrics for University Rankings
Author Information
Author(s): Ghaddar Ali, Thoumi Sergio, Saab Samer S.
Primary Institution: Lebanese American University
Hypothesis
Can journal citation-based metrics provide a more accurate evaluation of university research performance compared to traditional article citation metrics?
Conclusion
The proposed framework for university rankings offers a more equitable evaluation by focusing on journal-level citations and scholarly output, achieving high correlations with established ranking systems.
Supporting Evidence
- The new metrics showed significantly higher correlations with established rankings like QS and THE.
- The proposed metrics emphasize objective performance and mitigate citation biases.
- The study highlights the limitations of current ranking methodologies that rely on subjective reputation surveys.
Takeaway
This study suggests that universities should look at the overall quality of research, not just the number of citations, to improve their rankings.
Methodology
The study used correlation analysis to compare new metrics with existing university ranking systems and employed optimization techniques for feature selection.
Potential Biases
The reliance on citation counts can lead to manipulation and may not accurately reflect research impact.
Limitations
The study acknowledges potential biases in citation metrics and the limitations of data availability.
Participant Demographics
Top 500 universities featured in ARWU.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website