Towards complexity-sensitive book metrics for scholarly monographs in national databases for research output

Abstract

Purpose This study investigates an approach to book metrics for research evaluation that takes into account the complexity of scholarly monographs. This approach is based on work sets – unique scholarly works and their within-work related bibliographic entities – for scholarly monographs in national databases for research output. Design/methodology/approach This study examines bibliographic records on scholarly monographs acquired from four European databases (VABB in Flanders, Belgium; CROSBI in Croatia; CRISTIN in Norway; COBISS in Slovenia). Following a data enrichment process using metadata from OCLC WorldCat and Amazon Goodreads, the authors identify work sets and the corresponding ISBNs. Next, on the basis of the number of ISBNs per work set and the presence in WorldCat, they design a typology of scholarly monographs: Globally visible single-expression works, Globally visible multi-expression works, Miscellaneous and Globally invisible works. Findings The findings show that the concept “work set” and the proposed typology can aid the identification of influential scholarly monographs in the social sciences and humanities (i.e. the Globally visible multi-expression works). Practical implications In light of the findings, the authors outline requirements for the bibliographic control of scholarly monographs in national databases for research output that facilitate the use of the approach proposed here. Originality/value The authors use insights from library and information science (LIS) to construct complexity-sensitive book metrics. In doing so, the authors, on the one hand, propose a solution to a problem in research evaluation and, on the other hand, bring to attention the need for a dialogue between LIS and neighbouring communities that work with bibliographic data.

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 — 7 citations exist upstream.

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