Photo of measuring instrument

Bibliometrics – Publication patterns and impact

Bibliometrics is a collective name for various quantitative methods used to map and analyze research communication. The units primarily used are publications and citations, which can provide indications of scientific productivity, collaboration patterns and impact. These indicators are in turn often used in research evaluations, but can also be used as a tool in one’s own research process.

The University Library offers:

Information and advice on bibliometrics

Activities include presentations, participation in meetings and other educational settings to increase knowledge of bibliometrics. For example, we can inform about its context in research policy, different indexes, how to plan strategically when publishing, and how it is used in different types of evaluations. Contact researchsupport@lnu.se for group or individual advice. 

Delivery of publication statistics and execution of bibliometric analyses

The University Library's mission primarily includes commissions from the university management and faculty managements, where the largest task is to produce publication and citation statistics for the university’s annual report at the end of each year. Currently, DiVA is mainly used to produce publication statistics, Scopus' analysis tool SciVal is used to perform publication and citation analyses, and Altmetric Explorer and Overton are used to analyze dissemination in society. 

When is bibliometrics used? 

Bibliometrics has become more common in the academic world in recent years because it is a quick and easy way to evaluate research production and communication. These evaluations can, for example, form the basis for strategic decisions at the university level about funding allocations or investments in specific subject areas, but also for individual researchers' publishing strategies for maximum knowledge dissemination and visibility. 

Bibliometrics is best suited for evaluations of large numbers of publications, where peer review rarely is realistically feasible. However, in smaller evaluations, bibliometrics should be used as a complement to peer review, since peer review can always provide a more nuanced and contextualized picture of different publication patterns than what quantitative methods can. Bibliometric data and indices are also sometimes used as proxy indicators of quality. One example is the assumption that the more citations, the greater (positive) influence the article has had on science. 

The Swedish government's research bill 2008/09:50 presented a new model for distributing research grants to universities. With this competition-focused model, they wanted to raise the quality of Swedish research and stimulate researchers to publish in international journals. The distribution model is based on number of publications and citations in Web of Science, as well as the ability to attract external research funding. This continues to apply in the latest bill, 2024/25:60, where the bibliometric quality indicator is to be based on the top 10% most cited articles globally per subject area. 

For individual researchers, bibliometric indices can be helpful in designing a publication strategy. In order to get the best possible dissemination of their research, and thereby increase the chances of citation and impact, information that can be of value is, for example, highly cited research topics or what is being discussed a lot in the media. 

When is bibliometrics not used? 

The University Library does not conduct publication or citation analyses of individual researchers or research groups commissioned by third parties. This is based, among other things, on the fact that the Linnaeus University has signed the declarations for DORA and CoARA on responsible research assessment. Previously, for example, publication channel and/or Journal Impact Factor have been used as proxy indicators of the quality of publications and researchers' competence, but since these indicators are journal-based, they are now considered to be poorly suited for individual publications and researchers. At this level, qualitative evaluation is therefore advocated instead of quantitative evaluation. Another reason is that for individuals or small research groups there are rarely enough publications to constitute a robust statistical basis. This also applies to larger units, such as faculties, if the publications are too few. 

Altmetrics 

Altmetrics (alternative metrics) is the collective name for new and alternative ways of measuring the impact of research, largely outside of academic settings. These measure, for example, the number of views and downloads of scientific publications, whether they are saved in various bookmarking and database services such as Mendeley, and whether they have been discussed in news media, blogs and social media. Recently, citations of scientific publications in policy documents and patents have received more attention, as they can indicate societal impact. Societal impact plays a prominent role in the latest research bill (2024/25:60). 

For compilations of this type of impact, Linnaeus University has access to the databases Altmetric Explorer (including social media and patents) and Overton (policy documents). They can be used as a basis for, for example, analyzing the visibility of publications and planning communication strategies outside of the scientific community. 

Examples of bibliometric indices

 

Citations: The number of times a publication is found in the reference list of other publications.

Field Weighted Citation Impact: The number of citations for a publication in relation to the expected average for publications of the same type and year of publication in the same subject area.

h-index: An author with at least h articles, each of which has been cited at least h times.

Journal Impact Factor: The average number of citations during one year of all articles published in the previous two years in a journal.

Top X%: The proportion of publications among the X% most cited publications in a given field.

Contact us

Do you have questions about bibliometrics? Please contact researchsupport@lnu.se.