Photo: mikemacmarketing / Wikimedia Commons / CC BY 2.0
JSTOR and PubMed are top academic search engines. They are strong for different kinds of research. JSTOR is broad and useful across the humanities and social sciences. PubMed is built for health and life science research. Their value comes from clear coverage and focused search tools.
I use “top” here in a narrow sense. It does not mean that either database contains every useful article. It means that each is a major starting point within its main subject area. The right database depends on the question, the field, and the type of source needed.
JSTOR brings together academic journals, books, images, and primary sources. A primary source is material from the time or place being studied. It may be a letter, report, photograph, pamphlet, or other original record. This mix makes JSTOR useful for research that crosses articles, books, and historical material.
JSTOR also supports full-text search. Full text means the search can look inside the words of an item, not only its title or short summary. Its search tools include phrase searching, Boolean operators, fields, filters, and limits by content type. Boolean operators are words such as AND, OR, and NOT. They help connect or remove search terms.
This matters when a topic has many names. A search for “public memory” may produce a broad set of results. A search using a phrase, an author field, a date filter, or a subject limit can make the results easier to review. The search page also lets researchers look for content they can access, rather than showing every record in the database.
That last point deserves care. JSTOR search results may include material that a reader cannot open through a free account. JSTOR offers open content, and access may also come through a subscribing library or another permitted route. Search visibility and reading access are separate matters.
PubMed has a different purpose. It is a free search service for biomedical and life science literature. Its records include citations and summaries for research articles, along with links to full text when available. PubMed is managed by the U.S. National Library of Medicine.
PubMed is especially useful because it uses structured subject terms. These terms are part of a controlled vocabulary. A controlled vocabulary is a set of approved labels used to describe topics in a steady way. In PubMed, these labels help connect searches that use different everyday words for the same idea.
For example, a researcher may search for a disease by its common name. PubMed may also connect that term with related subject headings, older names, or narrower concepts. This can improve recall. Recall means finding more of the relevant material. It can also bring results that need careful review.
PubMed has limits that are easy to miss. It is not a general database for all academic fields. A strong PubMed result does not show that a topic has been fully covered in history, literature, law, or another humanities field. It also does not mean that every record has free full text.
The difference between the two databases is therefore more useful than a simple ranking. JSTOR is a broad research platform with strong value for journals, books, images, and primary sources. PubMed is a focused index for biomedical and life science literature. Each database organizes knowledge according to its own purpose.
I would begin with the database that matches the subject. A study of medical history may need both. JSTOR may help locate work in history, cultural studies, or ethics. PubMed may help locate medical research and the terms used within that field. The two result sets may support different parts of the same project.
Search terms also need care. A keyword is a word entered into a search box. It may appear in a title, subject label, abstract, or full text. A database can miss useful work when the search uses only one spelling, one time period, or one term for a broad idea.
A sound search often begins with two or three main concepts. Those concepts can then be tested with related words. In JSTOR, phrase marks and filters can narrow a large result set. In PubMed, subject headings and field searches can show how the database has described the topic.
The search results still require judgment. A database record is a guide to a source. It is not proof that the source is sound, complete, or useful for the exact claim being studied. Titles can sound relevant while the article treats the topic only briefly.
There is also a wider limit. Database coverage changes. Publishers add content, records are revised, and search systems may change their rules or tools. Public descriptions explain the intended coverage, but they do not remove every question about omissions, indexing choices, or older records.
That uncertainty does not make JSTOR or PubMed weak. It sets the proper boundary for their use. They are discovery systems. They help researchers locate material, learn the terms of a field, and trace related work. They do not replace close reading or source evaluation.
For an article database for research, the main choice is often clear. JSTOR is a strong first stop for broad humanities and social science work. PubMed is a strong first stop for biomedical and life science work. When a topic crosses fields, using both may reveal different parts of the record.
The practical lesson is simple. Match the database to the subject, then match the search method to the database. Check whether a result is a citation, a summary, or full text. Check whether the item is available through an open route or a library connection. Treat the result list as a beginning, not a final answer.
That is the kind of information The Source List is meant to keep in view: one digital source worth knowing, one search tip, and one honest limitation. For JSTOR and PubMed, the source choice is clear, the search tools are useful, and the limits should stay visible.