Convert DOC to RIS

What the DOC format is

.doc is Microsoft's older binary Word format, used for manuscripts, reports, and reference lists. Microsoft Word, older versions of LibreOffice Writer, and other word processors can create it; current Word versions normally use .docx. A DOC file may contain ordinary typed references, Word citation fields, or a bibliography generated by a reference-management add-in.

What the RIS format is

RIS is a plain-text exchange format for bibliographic records. A record uses tagged lines such as TY - JOUR for the reference type, AU - for an author, TI - for a title, JO - for a journal, and ER - to mark the record's end. Reference managers and library databases use RIS to transfer authors, titles, dates, journal names, identifiers, and other citation fields.

RIS is useful because it is portable, editable as text, and supported by Zotero, EndNote, RefWorks, Citavi, and many academic databases. It is not a document format, so it cannot preserve Word formatting or non-bibliographic content.

How to convert DOC to RIS

There is no reliable universal one-click conversion when a DOC contains a formatted reference list. The practical workflow is to extract the references, parse them into structured records, verify the metadata, and then export RIS.

Use the original reference library when available. If the bibliography was produced by Zotero, EndNote, RefWorks, or another citation add-in, open that application's library instead of parsing the DOC. Select the relevant records and use its export command, such as File → Export, then choose RIS. The library contains structured fields that may be lost when Word displays the bibliography.

For an ordinary typed reference list:

1. Make a copy of the DOC and open it in Microsoft Word or LibreOffice Writer. Remove headings, page numbers, footnotes, surrounding prose, and other material that is not part of the references.
2. Save only the reference list as plain text with File → Save As → Text. If the application cannot read the old DOC correctly, save a working copy as .docx first, then export that copy as text.
3. Paste the text into AnyStyle, a browser-based reference parser. Review its detected authors, titles, years, journals, volume and issue numbers, pages, DOI values, and reference types before exporting.
4. Export the parsed records as BibTeX and import that file into Zotero with File → Import.
5. Select the imported items in Zotero and use File → Export Items or File → Export Library. Choose RIS and save the resulting .ris file.

AnyStyle is convenient for non-sensitive bibliographies. Do not upload confidential manuscripts, unpublished references, or personal data to an online parser. For a desktop-only workflow, use LibreOffice to extract text, then enter records manually in Zotero or EndNote, or use a local parsing tool that you have tested.

LibreOffice can extract text from the command line if it is installed and the output directory already exists:

soffice --headless --convert-to txt:Text --outdir extracted "references.doc"

Use the generated text file as parser input. The command extracts visible text; it does not recover citation fields from proprietary Word add-ins.

For references with identifiers, retrieve them instead of parsing them. DOI values can be looked up through Crossref, a publisher, a university library, or Zotero's Add Item by Identifier. PMID values can be retrieved from PubMed, and ISBNs from a library catalogue or book database. Export the resulting records directly as RIS. Identifier-based records generally contain more accurate journal, author, page, and publication-type metadata than a parser working from visual formatting.

For only a few references, enter them manually. Create each item in Zotero or EndNote, verify it against the original publication, and export the selected records as RIS. Manual entry is often quicker than correcting parser errors in a short, inconsistently formatted list.

Quality and compatibility limits

A displayed Word citation may be ordinary text, a Word field, or content generated by a proprietary add-in. A DOC file therefore does not necessarily contain a complete, portable bibliography database. Parsing can misidentify initials, corporate authors, subtitles, article numbers, page ranges, accented names, and multiple works in one paragraph.

Open the resulting .ris file in a text editor and check that every record has a plausible TY - line and ends with ER -. Import it into the destination reference manager and inspect a sample, including author order, DOI, journal title, year, volume, issue, pages, and reference type. Some programs interpret RIS tags differently, so correct fields in the destination application after import.

Do not treat a PDF or DOC conversion service that promises automatic RIS output as authoritative: formatting alone cannot reliably identify every bibliographic field. Keep the original DOC and the verified reference library or exported RIS file as separate records; RIS will not preserve the source document's layout, images, tracked changes, or bibliography styling.