A Quick Guide to Common Journal & Article Metrics

July 1, 2017 Chris Swann

Do you know the nuances of each different metric? This article will provide you with a high-level cheat sheet, and then refer to the following article for a wider discussion on the pros and cons of each type.

Journal Citation Metrics

5-Year Impact Factor
Data source: Web of Science
Published in the annual Journal Citation Reports. Average citations in the JCR year to substantive papers (articles, proceedings papers, reviews) published in the previous 5 years.

CiteScore
Data source: Scopus
Average citations in the CiteScore year to papers published in the previous 3 years. Very similar to the Impact Factor, but with a longer citation window and no control for document types (except for meeting abstracts, which are not indexed in Scopus). Scopus tends to index a higher proportion of social science and humanities content than Web of Science, so titles in these areas will tend to perform better in the CiteScore rankings than in the JCR rankings. Titles publishing a significant number of editorial-style papers will tend to perform less well.

Eigenfactor
Data source: Web of Science
Published in the annual Journal Citation Reports. Based on weighted citations in the JCR year (excluding journal self-citations) to papers published within the previous 5 years. Citations are weighted according to the prestige of the citing journal (i.e. citations from top journals ‘mean more’ citations from lesser journals). The mathematics of the calculation are akin to the PageRank calculations that Google uses in its ranking algorithms.

Google Scholar Metrics
Data source: Google Scholar
These are ‘rolling metrics,’ i.e. based on a continually changing dataset. The main Google Scholar journal metric is the H5 index. This is very similar to the H-Index (explained below) but limited to papers published within the past 5 years.

Immediacy Index
Data source: Web of Science
Published in the annual Journal Citation Reports. Average citations in the JCR year to substantive papers published in the same year. This is really an indication of how rapidly research is cited. Journals with a high Immediacy Index will usually be journals representing a fast-paced research environment.

Impact Factor
Data source: Web of Science
Published in the annual Journal Citation Reports. Average citations in the JCR year to substantive papers published in the previous two years.

SJR
Data source: Scopus
Calculated by Scimago. The Scimago Journal Rank (SJR) is based on weighted citations in Year X to papers published in the previous 3 years. Citations are weighted by the ‘prestige’ of the citing journal, so that a citation from a top journal will ‘mean more’ than a citation from a low-ranked journal. As with the Eigenfactor, the calculation is broadly similar to the Google PageRank algorithm.

SNIP
Data source: Scopus
Calculated by the CWTS at the University of Leiden. The Source Normalized Impact per Paper (SNIP) measures average citations in Year X to papers published in the previous 3 years. Citations are weighted by the ‘citation potential’ of the journal’s subject category, thereby making the metric more comparable across different disciplines.

“Alternative” (Article or Author) Metrics

Altmetrics
Article-level metrics based on a broad spectrum of indicators, such as such as tweets, blog mentions, social bookmarking, etc. Each of these sources are weighted as to the ‘ease’ of being mentioned, and the prestige score of the mentioner. Scores cannot be averaged, but an article can be ranked with respect to the scores attained by other articles within the same journal.

Relative Citation Ratio
The Relative Citation Ratio (RCR) was launched in October 2015 by a team of researchers from the National Institutes of Health. Like the SNIP, the RCR is a field-weighted citation metric, yet calculated at the article rather than the journal level. The novelty is the way in which the article’s field is defined—i.e. according to its co-citation network, rather than a more generic journal-level classification. Unlike the network of papers citing the article (which can be heavily skewed by the addition of a relatively small number of citations), the co-citation network (other papers cited concurrently with the target paper) attracts a larger citation pool, and therefore renders the calculation less vulnerable to fluctuation. The metric is calculated via the PubMed database.

H-index
Data source: Any
A citation metric designed to evaluate individual authors, but which can be extended to any dataset. The H-index indicates the number of papers, H, that have been cited at least H times, e.g. an H-index of 15 means that 15 papers have been cited at least 15 times each. This metric does not control for the age of documents or citations, and can be calculated from any citation database. Caution is advised, as the same group of articles will yield a different H-Index in different databases.

Usage Metrics
Usage metrics (such as full text downloads or page accesses) are collected at an article level, and often subsequently aggregated to a journal level. Their key benefit is that they are more ‘immediate’ than citation metrics (which can take several years to accumulate), yet they remain vulnerable to sudden skews—the result of ‘going viral’, of changes in accessibility (i.e. open access) or of un- tracked and emergent web crawlers. These factors can then show a rapid and short-lived upturn in the number of visits to an article within the journal website, with corresponding increase in downloads. Critically, dramatic changes within usage statistics at an article and journal level may be due to many factors, but rarely are they due to an article being of exceptionally ‘high quality’. As such, usage metrics measure academic attention, but not necessarily value to an academic discipline.

Previous Article
New Tools and Classic Techniques: Top Tips for Finding Quality Reviewers
New Tools and Classic Techniques: Top Tips for Finding Quality Reviewers

Finding rigorous reviewers can be a challenge. Read on to learn some top techniques that make it easier.

Next Article
Journal and Article Metrics: An Overview
Journal and Article Metrics: An Overview

A history and evaluation of Journal Impact Factors and the metrics that add depth and nuance to them.