Finding the full text of a research article

A search for a scientific paper usually turns up a title, a list of authors and a short summary. The paper itself, with its methods, data and figures, can be much harder to reach. This page explains what the different pieces are, where legitimately free versions live, and the ordinary routes readers use when a paper is not freely available. It offers no files and links to no papers.

Abstract, record and full paper

Three different things tend to get confused. The first is the bibliographic record: the title, authors, journal, year and identifiers that tell you a paper exists and where it was published. The second is the abstract, a short summary written by the authors. Wikipedia's article on IMRaD notes that a brief abstract is usually required for publication, and that it should be composed to function as an autonomous text. The third is the full paper, which contains the introduction, the methods, the results with their figures and tables, the discussion, and the reference list.

The difference matters because an abstract is, by design, a compression. It reports the question, the main result and the authors' conclusion in a few sentences, but it leaves out almost everything a careful reader needs to judge the work: how many samples there were, how measurements were made, what controls were used and what the limitations were. Reading the abstract tells you what a paper claims; only the full paper lets you see how well the claim is supported. A good habit is to treat any abstract as an invitation to read further, never as the last word.

What PubMed is, and what it is not

For biomedical and life-science literature, PubMed is often the first stop. Its own About page describes it as a free resource supporting the search and retrieval of biomedical and life sciences literature, containing more than 40 million citations and abstracts. The same page is clear about a point many newcomers miss: PubMed does not include full text journal articles, although links to the full text are often present when available from other sources, such as the publisher's website or PubMed Central.

The About page explains that PubMed has been available to the public online since 1996 and is developed and maintained by the National Center for Biotechnology Information at the National Library of Medicine, part of the National Institutes of Health. It describes three components: MEDLINE, the largest, consisting primarily of citations from journals selected for it and indexed with Medical Subject Headings; citations for PubMed Central articles, the second largest component, where PubMed Central is a repository of complete articles from journals reviewed and selected for inclusion and of individual articles deposited to comply with funder policies; and citations for books and chapters available on Bookshelf. So a PubMed record is a signpost, and the full paper, when it is freely readable, usually lives somewhere else.

Open access

Much of the research literature can now be read without charge. Wikipedia's article on open access defines it as a set of principles and practices through which publications are delivered to readers free of access charges or other barriers, and notes that under the strict definition barriers to copying and reuse are also reduced by an open licence. It explains that open access types are commonly described with a colour system, the most familiar names being green, gold and hybrid. In broad terms, gold open access means the publisher makes the paper free on its own site, while green open access means the author posts a version elsewhere, in a repository run by an institution or an independent service, where readers can reach it without paying.

This has a practical consequence. A paper that sits behind a publisher's paywall may still have a legitimate free version: an author's accepted manuscript in a university repository, or a copy in PubMed Central deposited under a funder's policy. The wording of such versions can differ slightly from the final published text, and careful readers note which version they have read.

Where open-access journals are listed

For open-access journals themselves, a major directory is DOAJ. Its About page describes DOAJ as a unique and extensive index of diverse open access journals from around the world, committed to ensuring quality content is openly available online, with a mission to increase the visibility, accessibility, reputation, usage and impact of quality, peer-reviewed, open access scholarly research journals, regardless of discipline, geography or language. It states that DOAJ is run by the DOAJ Foundation, a non-profit, community-governed foundation based in Denmark. A reader who wants to know whether a journal presents itself as open access, and whether it has been accepted into a curated list, can look for it there.

When a paper is not free

If no free version turns up, there are still ordinary, legitimate routes.

Libraries

University and research libraries pay for subscriptions so that their members can read paywalled journals, and many public libraries offer access to some databases as well. Where a library does not hold a journal, interlibrary loan services can often obtain a copy of a single article for a registered reader. Librarians are also expert searchers and can often find a version that a quick search missed.

The authors

Researchers generally want their work read. Most are willing to share a copy of their own paper with a reader who asks politely, and many list their publications, sometimes with accepted manuscripts, on institutional web pages. The contact details for the corresponding author are normally printed with the paper itself.

The publisher

Publishers usually sell access to individual articles and sometimes offer short-term rental. This is the most expensive route, and it is worth trying the others first.

Checking what you have found

However a paper is reached, it is worth confirming that it is the right one. The title, authors, year and journal should match the record where you first met it. Be cautious of sites that offer large numbers of papers without the publisher's or authors' permission: apart from the copyright question, such copies may be incomplete, out of date or altered. A copy from the publisher, a recognised repository, a library or the authors themselves is the one to trust.