Export, import and merge database records

Different analyses need different fields, and the default export settings of most databases leave out the most important one: the cited references. A few extra clicks at export time decide which analyses you can run later, so it is worth getting right.

Which fields each analysis needs

Scopus

1. Select all results and click Export, then choose CSV (BibTeX and RIS also work).
2. Select every field group: citation information, bibliographical information, abstract and keywords, funding details and References.
3. Export. For more than 20,000 records, split the search by year and export each part.

Web of Science

1. Search the Core Collection (other databases on the platform export fewer fields).
2. Click Export and choose Plain text file (tab-delimited also works).
3. Set the record content to Full Record and Cited References.
4. Export in batches of 1,000 records (1–1,000, 1,001–2,000 and so on).

PubMed and other sources

In PubMed, choose Save, all results, format PubMed (a MEDLINE-tagged text file with MeSH terms and affiliations). Dimensions, Lens.org and OpenAlex export CSV files; Lens and OpenAlex give references as identifiers, which supports local citations and coupling but not reference labels. Reference managers (Zotero, Mendeley, EndNote) and databases such as IEEE Xplore, EBSCO and ProQuest export RIS or BibTeX, usually without cited references.

Importing and merging

Drop every export file onto the Import data page at once: several Web of Science batches, or a Scopus file and a Web of Science file together. The Studio detects each format automatically, reads the records into one collection and removes duplicates. Matching uses the DOI first, then the normalised title together with the publication year for records without a DOI. When a document is found in two databases, the more complete record is kept and its gaps are filled from the other (for example, Web of Science cited references with Scopus keywords). The import summary reports how many records each file contributed and how many duplicates were removed; these numbers go into the PRISMA diagram.

A warning about mixed reference formats

Scopus and Web of Science write cited references in different formats (Scopus gives full author lists and titles, Web of Science a short “first author, year, journal, volume, page” label). After merging, the same cited work can appear under two labels, which splits its co-citation counts. If co-citation analysis is central to your study, run it on a single database, or report that references were not harmonised across databases.

Your data stays with you

All parsing happens in your browser; the files are not uploaded. That matters when you work with licensed database content. Use Save project in the Studio, or keep the raw export files, so you can rebuild the collection at any time.

Common export problems

• No references in co-citation analysis: the export did not include cited references. Re-export with the full record.
• Missing affiliations or countries: some document types and older records lack them; check the Data completeness page.
• Garbled characters: export with UTF-8 encoding where offered; the Studio reads UTF-8 and most common encodings.
• Too many records: split by year and drop all parts together.