Skip to main content

Doing the literature review: Selecting databases

Selecting databases

You have to decide which databases you will use in your literature search. To limit location bias, you have to use more than one database. Make an informed choice! In this module we list some database features you can take into account.

There are different types of literature databases:

  • Abstracting & indexing databases (A&I) provide metadata and abstracts. The metadata includes the title, author(s), date of publication, journal title, volume and issue, page numbers, keywords, DOI, etc.
    • Discipline specific databases, such as PsycINFO, Philosopher's Index, Sociologial Abstracts and Business Source Premier, offer extra search tools, for example a thesaurus. The journals (some databases even select at the article level!) indexed in these databases are selected carefully, based on selection criteria.
      These databases may be available via different vendors/platforms.
    • Multidisciplinary databases, such as Scopus and Web of Science, index academic journals from all disciplines, ranging from astronomy to zoology. These will help you find relevant articles in journals outside your own discipline, but non-relevant results are hard to avoid. Both Scopus and Web of Science are citation databases, which means that they track citations: you can see if an article is cited in other papers in the database and by which authors.
      JSTOR also offers access to journals from different disciplines, but be aware that this database has an archive function - for most journals you can’t access or search the most recently published volumes.
  • Publishers databases, such as ScienceDirect (Elsevier), SpringerLink and Wiley Online Library, are limited to publications from a particular publisher. Therefore they are not suitable for a literature search for a review. But of course, you will use these databases to retrieve the full text of articles.

The Checklist Selecting Databases provides an overview of features you have to take into account when choosing the databases to use for your literature review.

What about Google Scholar?

Google Scholar indexes websites with scholarly articles – including websites of academic publishers, university repositories and personal websites of researchers. A major difference between Google Scholar and A&I databases is that Google Scholar doesn’t provide information about the indexed websites or journals. It’s hard to check if a particular journal is indexed cover-to-cover in Google Scholar. Google Scholar gives no definition of ‘scholarly’. Amongst the scholarly results you might get results from predatory publishers and papers written by students.

When you use Google Scholar for a search for your literature review, be aware that it can be hard to perform a structured, repeatable search:

  • the advanced search options and filter options are limited
  • you can't combine two search queries afterwards
  • you can enter up to 256 characters in the search box of Google Scholar - if you want to include synonyms of search terms you often need more characters
  • Google Scholar limits the results of any search query to 1000 papers, irrespectible of the number on top of the search results
  • the information shown in the search results of Google Scholar can be very limited
  • the metadata of articles (for example the publication year) can be incorrect, due to parsing errors
  • specific journals or publishers might not be indexed by Google Scholar, due to technical reasons.

Google Scholar is a great tool for locating articles you know the titles of. In the Scholar settings you can add a Library link to the Erasmus University Library (see the module Get the most out of Google Scholar) then follow the FULL-TEXT @ EUR links to the published version of an article in the EUR Library collection. 

TIP: Publish or Perish (also called PoP) is software used for citation analysis, based on Google Scholar citation data. The general citation search in Publish or Perish allows you to perform an Advanced Scholar Search query and analyse its results. The advantage is the presentation of the results: you can sort by author, year, times cited, publication and publisher. The abstract is not shown. It's possible to export the output, for example to Excel.