Using the Mann‐Whitney test on informetric data

Abstract

Abstract The fields of informetrics and scientometrics have suffered from the lack of a powerful test to detect the differences between two samples. We show that the Mann‐Whitney test is a good test on the publication productivity of journals and of authors. Its main limitation is a lack of power on small samples that have small differences. This is not the fault of the test, but rather reflects the fact that small, similar samples have little to distinguish between them.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.