The main aim of this study is to identify which measures or metrics are suited to supplement traditional impact metrics based on citations and to investigate their differences and robustness by... Show moreThe main aim of this study is to identify which measures or metrics are suited to supplement traditional impact metrics based on citations and to investigate their differences and robustness by using the parameter-free distribution-based method named “Characteristic Scores and Scales” (CSS). For this pilot study, we used all papers in two subfields of chemistry from the WoS Core Collection with a valid DOI number, which was the basis for the matching with PlumX data. We found significant correlations at different degrees among the most significant indicators covering three dimensions (captures, usage and citations). Three citation indicators of different platforms are highly correlated with each other, while Mendeley readership and WoS usage counts showed low and medium correlations. Finally, this case study also proves the benefits of the CSS method in order to identify metric-specific peculiarities as well as strong deviations and extreme values. The CSS analysis show similar distribution patterns of all the indicators in the two subfields. According to these results, abstract view at EBSCO showed the most skewed distribution. Show less
Glänzel, W.; Thijs, B.; Zhang, L.; Sivertsen, G. 2018
In the present study, we develop and apply a model for individual-level evaluation of research activity and citation impact. In particular, we analyse publication activity and citation impact on... Show moreIn the present study, we develop and apply a model for individual-level evaluation of research activity and citation impact. In particular, we analyse publication activity and citation impact on the basis of a unique cleaned national dataset, where the most commonly known issue in individual-level bibliometrics, namely the problem with large-scale author identification is eliminated. The full individual validation of the set also helps gain clarity regarding the final limitations of the subject, profile and seniority factor in productivity studies. The study focuses on the following questions. Firstly, can we extend the model of Characteristic Scores and Scales (CSS) to publication activity and what are the caveats? Secondly, in how far does co-authorship fractionation matter affect the indictors? Finally, how can we combine publication and citation CSS models? In order to answer these questions, a 10-year publication dataset of the Norwegian Cristin database was used in combination with citations from the Clarivate Analytics Web of Science Core Collection. Show less
The main purpose of this article is to reveal the effect of self-archiving on the citation impact of preprints in Information Science & Library Science (LIS). The arXiv and WoS database are... Show moreThe main purpose of this article is to reveal the effect of self-archiving on the citation impact of preprints in Information Science & Library Science (LIS). The arXiv and WoS database are used and all citations from WoS are collected to address two key questions—one is the share of papers published in the core LIS journals that also deposited in arXiv and time lags in publication process, the other is the differences in aging characteristics and citation impact of journal articles deposited in arXiv compared to those not deposited—through statistical and bibliometric analysis. The results show that the proportion of journal articles deposited in arXiv is low but there is an overall increasing trend, nearly 80% of all arXiv papers are preprints and authors are more likely to post the paper to arXiv when it is received or accepted by a journal, the preprints enjoy a boost in early citations and the citation advantage decrease with the increasing author productivity, the citations to arXiv version papers indicate an OA effect of arXiv. Show less
