Why Scopus is often the first choice
Scopus is a multidisciplinary abstract and citation database operated by Elsevier, covering journals, conference proceedings and books from a very large number of publishers. Compared with many alternatives it tends to offer wider coverage in the social sciences, engineering and journals published outside North America and Western Europe. That makes it a natural starting point when your field has substantial output from Asia or from specialised regional journals.
Each Scopus record carries fields that matter for bibliometrics: author names with identifiers, affiliations, author keywords, indexed keywords, abstracts, funding information and the full reference list. Using these fields properly, rather than just counting records, is what separates a credible Scopus bibliometric analysis from a superficial one.
Scopus is not automatically the right choice, however. In some natural science fields the long citation history of Web of Science matters more, and some journals in your area may be indexed in one database but not the other. We check this with you during scoping, and the Advanced package allows both sources to be merged when neither alone is sufficient.
Building a Scopus query that holds up
We work in Advanced search, combining field codes, Boolean operators and filters. Here is an illustrative query for a hypothetical project on the circular economy:
TITLE-ABS-KEY ( "circular economy" OR "circular business model*" )
AND PUBYEAR > 2009 AND PUBYEAR < 2026
AND ( LIMIT-TO ( DOCTYPE , "ar" ) OR LIMIT-TO ( DOCTYPE , "re" ) )
AND ( LIMIT-TO ( LANGUAGE , "English" ) )
The query is refined over several rounds. We check that a benchmark set of papers you know to be relevant is retrieved, read a random sample of hits to estimate the share of irrelevant records, and then add synonyms or exclusion terms as needed. Each version and its hit count is logged so the final choice can be defended and reported.
Exporting in the right format
| Format | Best for | Notes |
|---|---|---|
| CSV | VOSviewer, Bibliometrix, spreadsheets | Select all available information, including abstracts, keywords and references |
| BibTeX | Bibliometrix, reference managers | Convenient when merging with other sources in R |
| RIS | Reference managers and screening tools | Rarely used directly for network analysis |
How many records can be exported at once depends on the format and on your institution's subscription. For large result sets we split the export by publication year, recombine the parts and verify that the total matches the original hit count with no gaps or duplicates. Our step-by-step guide on exporting Scopus data covers the details if you prefer to export yourself.
Project workflow
Confirm access
Where the licence requires it, records are exported through your institution's legitimate access. If you export them yourself, we send precise instructions covering which fields to tick, which format to choose and how to name the files so that batches can be recombined without errors.
Finalise the query
Versions are tested and logged with hit counts for your appendix.
Clean
Errata, records without authors and malformed entries are removed; affiliation and keyword variants are harmonised.
Analyse
Descriptive statistics on output, citations, authors, sources and countries, plus four VOSviewer maps in the Basic package.
Deliver
Processed files, figures and a technical report within five working days on the Basic package, with free revisions for 30 days within the package scope.
Scopus-specific details we handle
- Author ID: separates authors who share a name, though one person occasionally has more than one profile. We check highly productive authors in particular.
- Affiliations: a single university may appear under several spellings, especially when non-English names are transliterated with or without diacritics.
- Document types: whether to keep conference papers, book chapters and articles in press depends on your aims and is decided with you.
- Indexed keywords: assigned from controlled vocabularies; useful, but different in nature from author keywords, so the choice is documented.
- Secondary documents: cited works that are not themselves indexed appear only inside reference strings, which affects co-citation results and is noted in the report.
Two further checks are easy to miss. Scopus occasionally contains near-duplicate records, for instance an article in press and its final version, which we detect and merge. Citation counts also keep changing after export, so we record the export date and avoid mixing counts retrieved on different days within one analysis.
Reporting Scopus as your data source
State the database, the exact query, the retrieval date, all filters, and the number of records before and after cleaning. Explain why Scopus was chosen and acknowledge coverage limits, for example national journals that are not indexed. Our technical report contains every element you need, and you then write the rationale and limitations in your own words.
A hypothetical example of the record flow: 2,184 records retrieved on the export date, 37 removed as errata or records without authors, 21 near-duplicates merged, leaving 2,126 documents for analysis. Numbers like these feed directly into a PRISMA 2020 style flow diagram if your journal expects one, which is available from the Advanced package upward.
Licence compliance and integrity
See pricing or start an order.