Comparing standard, collaboration and fractional CNCI at the institutional level: Consequences for performance evaluation

Abstract The average Category Normalised Citation Impact (CNCI) of an institution’s publication output is a widely used indicator for research performance benchmarking. However, it combines all entity contributions, obscuring individual inputs and preventing clear insight sound policy recommendations if not correctly understood. Here, variations (Fractional Collaboration [Collab] CNCI)—which aim to address the obscurity problem—are compared Standard CNCI over 250 institutions, spread globally, covering ten-year period using Web Science data. Results demonstrate that both Fractional Collab methods produce lower index values than CNCI. results are often near-identical despite fundamentally different calculation approaches. Collab-CNCI, however, avoids assigning fractional credit (which potentially incorrect) relatively easy implement. As single metrics obscure inputs, institutional also deconstructed into five collaboration groups. These groups track increasing international trend, particularly highly multi-lateral studies decrease in publications authored by institutions. deconstruction shows increase with level collaboration. Collab-CNCI does necessarily follow this pattern thus enabling identification institutions where, example, their domestic articles best performing group. Comparing variants deconstructing portfolios type is, when understood correctly, essential tool interpreting informing making.