A quantitative approach to evaluating the performance and quality of modern academic research using statistical indices
Journal of Statistical Sciences and Computational Intelligence · 19.06.2026 · 2 (2) · 620-636
DOI: 10.64497/jssci.210 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Statistical Sciences and Computational Intelligence, online erschienen am 19.06.2026. Verfasst von Bright Emmanuel Udomette. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The growing reliance on statistical indices to evaluate academic performance has transformed research governance, funding allocation, and institutional ranking systems, and how researchers measure productivity and impact. This study empirically assesses the explanatory power of core bibliometric indicators (publication count, h-index, collaboration index (CI), degree of collaboration (DC), and citation count) in predicting overall research performance across STEM, Social Sciences, and Humanities disciplines. Using a cross-sectional dataset of 120 researchers and employing multiple regression and ANOVA techniques, the findings reveal that the integrated model explains 58% of the variance in Research Performance Score (R² = 0.58; Adjusted R² = 0.56; F(6,113) = 26.01, p < .001; f² = 1.38, very large effect). The h-index emerged as the strongest predictor (β = 0.42, p < .001), followed by publication count (β = 0.31, p < .001) and collaboration indicators (CI: β = 0.24, p = .004; DC: β = 0.18, p = .022). Significant disciplinary differences were also observed (η² up to 0.16, p < .001), confirming that metric behaviour is field-sensitive. The results demonstrate that research performance is a multidimensional construct best captured through composite, field-normalised evaluation frameworks rather than isolated metrics. The study contributes to scientometric literature by quantitatively validating the productivity–impact–collaboration triad within an integrated model and offers policy-relevant evidence for balanced, context-sensitive research assessment systems. It is therefore recommended that a multidimensional index system with multiple indices, benchmarking and evaluation frameworks should be field-normalised to ensure fairness across disciplines.
Zitieren
APA
Udomette, B. E. (2026). A quantitative approach to evaluating the performance and quality of modern academic research using statistical indices. Journal of Statistical Sciences and Computational Intelligence, 2(2), 620-636. https://doi.org/10.64497/jssci.210
Vancouver
Udomette BE. A quantitative approach to evaluating the performance and quality of modern academic research using statistical indices. Journal of Statistical Sciences and Computational Intelligence. 2026;2(2):620-636. doi:10.64497/jssci.210
BibTeX
@article{Udomette2026,
author = {Udomette, Bright Emmanuel},
title = {A quantitative approach to evaluating the performance and quality of modern academic research using statistical indices},
journal = {Journal of Statistical Sciences and Computational Intelligence},
year = {2026},
volume = {2},
number = {2},
pages = {620--636},
doi = {10.64497/jssci.210}
}
EUVIMED