Your goal is to scan some abstracts, read 2-3 articles, and then move on. What do you do, then, if your keyword search turns up 10,000 search results? If you just need a few good articles, one strategy is to take advantage of sorting and relevance ranking features – to try to move “good” articles closer to the top of the list. A challenge is that each database defines “good” (relevance) differently, and none of them can read your mind (yet).
Here’s an overview of sorting and relevance ranking features in PubMed, Google Scholar, and Web of Science, followed by a search example showing how they compare. (Click HERE for screen shots showing how to change sorting options.)
PubMed: The “Best Match” option sorts search results according to where your search terms appear and how frequently they appear in the reference.
Google Scholar: Automatically ranks search results according to a proprietary algorithm which considers how frequently an article has been cited, where your search terms appear, and other factors.
Web of Science: Offers multiple sorting options. Can sort search results by relevance, times cited, or usage count. Like PubMed, “relevance” is based on where the search terms appear and how frequently they appear.
| Resource | Content Searched | Factors Considered in Relevance Ranking | Sorting Options |
| PubMed | Titles, abstracts, author keywords, & MeSH terms – of journal articles | Location and frequency of search terms in the reference | Publication Date (default); Best Match; First Author; Last Author; Journal; Title |
| Google Scholar | Full text of journal articles, dissertations, books, & other scholarly publications. | Numbers of times cited; location of search terms in reference; journal; “other”. | No options. Ranking & sorting are automatic. |
| Web of Science | Titles, abstracts, & keywords of journal articles, conference abstracts, & book chapters. | Location and frequency of search terms in the reference. | Date (default); Relevance; Times Cited; Usage Count; First Author; Journal Title |
Search Example: The relationship of obesity to the microbiome.
Search Terms: microbiome obesity
| Resource | Interpretation of Search Terms |
Number of Results (June2018) | Article from the Top 5 Results Ranked by Relevance (or Best Match) |
| PubMed | (microbiome OR microbiota) AND obesity | 2,926 | “Looking for a signal in the noise: revisiting obesity and the microbiome” – MBio, 2016. |
| Google Scholar | microbiome AND obesity | “About 46,900” | “An obesity-associated gut microbiome with increased capacity for energy harvest” – Nature, 2006. |
| Web of Science | microbiome AND obesity | 1,609 | “Obesity: Gut microbiome and serum metabolome changes” – editorial – Nature Reviews Endocrinology, 2017. |
Still Not Finding What You Need?
What you consider “good” might not be a highly cited article or an article with your search terms in the title! Consider looking for recent review articles, articles by a known expert in the field, or contact your librarian for assistance!