The bibliometric quotient (BQ), or how to measure a researcher’s performance capacity: A Bayesian Poisson Rasch model

Rüdiger Mutz and Hans-Dieter Daniel

Journal of Informetrics, 2018, vol. 12, issue 4, 1282-1295

Abstract: Regarding evaluation of individual researchers, the bibliometric indicators approach has been increasingly discussed recently, but there are some problems involved with it: construct definition, measurement errors, level of scale, dimensionality, normalization. Based on a psychometric model, the Rasch model, we developed a measuring scale for the theoretical construct ‘researcher’s performance capacity,’ defined as the competency of a researcher to write influential papers. The aim was a scale that is one-dimensional and continuous, is applicable to bibliometric count variables, and takes measurement errors into account. In this paper we present the psychometric model (Bayesian Poisson Rasch model, BPR) and its assumptions and examine the behavior of the model under various sampling conditions. For a sample of N = 254 researchers in a quantitative methodology section of an undisclosed German academic society for social sciences, using the BPR model we developed a scale that we named ‘Bibliometric Quotient’ (BQ, M = 100, SD = 15) (following the term ‘intelligence quotient’). The scale fulfills most of the test-theoretical requirements (e.g., high reliability αt = .96, no differential item functioning except for academic age and German states) and in addition allows researchers to be ranked. Women’s BQ scores were 8.3 points lower on the scale than men’s BQ scores.

Keywords: Rasch model; Rasch model for count data; Bayesian statistics; Psychometric model; Research evaluation (search for similar items in EconPapers)
Date: 2018
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S1751157718300798
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:infome:v:12:y:2018:i:4:p:1282-1295

DOI: 10.1016/j.joi.2018.10.006

Access Statistics for this article

Journal of Informetrics is currently edited by Leo Egghe

More articles in Journal of Informetrics from Elsevier
Bibliographic data for series maintained by Catherine Liu ().