The purpose of article-level metrics is to establish the impact of an article. The most common way of evaluating this is to count the number of times an article has been cited in other articles. Citation counts usually vary by database. This is because different databases index different sets of journals. Follow the tabs at the top of this box to learn about article metrics in the following databases:
Keep in mind that it is usually lag a between the time an article is published and when authors start citing it in their publications.
Follow the tabs at the top of this box to learn about article metrics in Web of Science, Google Scholar, Einstein Research Profiles, and on journal websites.
PubMed's citation counts are based on how many times an article is cited by articles in PubMed Central.
Web of Science measures citation counts based on the journals listed in its Master Journal List (MJL). The Web of Science Core Collection currently indexes 9,529 journals.
In addition to a citation counts, Web of Science also provides usage counts (circled in green below), which show the number of times other Web of Science users have viewed a citation .
Citation counts tend to be higher in search engine such as Google Scholar than traditional bibliographic databases such as PubMed and Web of Science. In addition to journals, Google Scholars web crawlers pull citation data from other sources, such as books, government publications, preprint servers, and content stored in institutional repositories. Unlike traditional databases, Google Scholar does not provide a list of journals or resources that it covers.
Einstein Research Profiles pulls citation counts from Elsevier's Scopus database.
Publishers have different ways of calculating citation counts, and it's not always obvious what's being counted. The American Society for Clinical Investigation calculates citation counts for is articles based on data from Crossref. Other publishers might use different sources.
JCI includes the following disclaimer on their website, "This citation data is accumulated from CrossRef, which receives citation information from participating publishers, including this journal. Not all publishers participate in CrossRef, so this information is not comprehensive. Additionally, data may not reflect the most current citations to this article, and the data may differ from citation information available from other sources (for example, Google Scholar, Web of Science, and Scopus)."
Author-level metrics attempt to quantify impact by analyzing citations arising from an individual author's publications.
Advantages: These metrics can give a more holistic idea of author impact by including a wide range of journals.
Disadvantages:These metrics are biased toward more prolific and more established authors. They also are not generalizable across disciplines.
One of the most common author metrics is the H-Index (or Hirsch-index), which measures the impact of a particular scientist rather than a journal. It takes into account the number of papers published and the number of citations received by these papers resulting in a single number rating. For example, a scholar with an h-index of 5 has published 5 papers, each of which has been cited by others at least 5 times.
Note that an individual's h-index will probably vary by database. This is because the databases index different journals and cover different years. For instance, Web of Science calculates an h-index using the years 1985-present. Google Scholar Citations covers a different set of years and journals. Authors' h-indices will also vary by discipline. Different fields of research have different citation rates.
Follow the tabs at the top of this box to learn about author metrics in Web of Science, Google Scholar, and Einstein Research Profiles.
Web of Science provides Citation Reports for authors covering the years 1985-present in the Web of Science Core Collection. Citation reports include:
The numbers in Google Scholar's Author Profile tend to be higher than resources like Web of Science and Scopus due to the large number of journals and websites that it crawls. Google does not disclose how frequently data is updated or how far back in time it searches.
Authors can edit their Profiles to ensure that relevant publications are included and that their contact information is correct. Information in a Google Scholar Author Profile includes:
Einstein Research Profiles uses Scopus data to provide citation countsm h-index, and graph of research output by year. In addition to publications, Einstein Research Profiles provides a more complete picture of an author by including data on grants data on grants and a research "Fingerprint" providing keywords that describes an authors area of interest.
Journal-level metrics attempt to quantify a journal's impact by analyzing the citations arising from the articles it publishes.
Advantages: Journal metrics can give a sense of which journals are popular and/or respected within a specific field.
Disadvantages: These metrics effectively average the impact of a journal's articles and authors, so they hide variations among articles and authors. Journal metrics also are not generalizable across disciplines.
Caveat: As with other metrics, a journal impact factor by itself is just a number and needs context to be meaningful. When looking at a journal's impact factor be sure to compare it with other journals in the same field, and pay attention to how it changes over time.
Follow the tabs at the top of this box to learn about journal metrics in JCI Impact Factors, Eigenfactor, SCImago, Google Scholar, and PubsHub.
The Eigenfactor, like the Impact Factor, starts with the citation data from Journal Citation Reports but has a more complicated algorithm. Journals are considered to be more influential if they are cited often by other influential journals. For example, citations from Nature or Cell are valued more highly than citations from journals with a narrower readership. Eigenfactor scores are also adjusted for differences in citation patterns across disciplines. They rely on data from five years, as compared to two for the Impact Factor.
Eigenfactor scores are scaled so that the sum of the Eigenfactor scores of all journals listed in Journal Citation Reports (JCR) is 100. In 2012, the journal Nature had the highest Eigenfactor score, with a score of 1.56539. In 2020, Nature fell to #4 with a score of 1.21714. PloS One had the highest Eigenfactor with a score of 1.38933.
The SCImago Journal & Country Rank is a portal that includes the journals and country scientific indicators developed from the information contained in Elevier's Scopus database.
Here are a few screen shots of the SCImago report for the journal Blood.
To locate Google Scholar's journal metrics, click on the menu icon at the top, left-hand corener, and select Metrics.
Here is the metrics page for the journal Blood.
The term altmetrics refers to new, alternative ways of assessing the impact of authors and publications. It presumes that value can be assessed by counting online shares, saves, reviews, adaptations and mentions. Altmetrics take into account a variety of research products, including gray literature; research blogs; data sets; citation managers, like Mendeley; and social media, such as Facebook and Twitter
Sources for Altmetrics
Web of Science has tools for analyzing search results and viewing citation trends. Using the Analyze Results option will allow you identify trends within a set of results:
From the Results page, click Analyze Results.
Select the criteria you want to analyze from the dropdown menu. Visualization options include treemap and bar charts. Visualizations can be downloaded in JPG format. Data tables can be downloaded in TXT format.