03 · What You Need to Know
Research and Publication Are Connected but Distinct
The research exists before the journal article
A research project begins with a question or problem and proceeds through design, evidence generation or identification, analysis, interpretation, and conclusions. These are the activities through which the research contribution is produced.
Publication occurs later, when some representation of that work is communicated to an audience.
This follows from the meaning and essential characteristics of research. Journal acceptance is not one of the characteristics that transforms an investigation into research.
Publication is one form of research dissemination
A peer-reviewed journal article is an important scholarly output, but it is not the only way research can be communicated.
Research may appear in books, conference proceedings, theses, dissertations, institutional repositories, technical reports, government reports, working papers, preprints, datasets, registered reports, monographs, policy reports, or other disciplinary formats.
Different fields have different communication traditions. In some areas of computer science, for example, peer-reviewed conference proceedings can be major scholarly outputs. In other disciplines, journal articles or books play a larger role.
The appropriate dissemination route depends on the field, research output, audience, and applicable requirements.
Peer review and publication are not the same thing
Peer review is a process through which other specialists evaluate scholarly work. Publication is the act or process of making the work available through a publication venue.
Many scholarly journals use peer review before publication, but the concepts should remain separate. A preprint may be publicly available before formal peer review. Some reports undergo expert review outside journals. A thesis may be examined by a committee. Editorial formats also differ among journals.
Research
The systematic investigation through which a question is examined and a knowledge contribution is developed or evaluated.
Peer review
Evaluation of scholarly work by relevant experts or peers according to the procedures of a journal, conference, funder, institution, or other scholarly process.
Publication
The communication and dissemination of research or other scholarly work through a publication venue or platform.
Peer review does not certify that research is correct
Publication in a peer-reviewed journal can provide useful quality control. Reviewers may identify methodological weaknesses, unclear reasoning, missing literature, unsupported conclusions, or reporting problems.
But peer review is not an experimental verification service.
Reviewers usually evaluate a manuscript and the materials available to them. They may not independently reproduce every analysis, inspect every raw observation, verify every citation, or detect every methodological problem.
Published research can later be corrected, challenged, contradicted, or retracted. The National Academies emphasizes that scientific knowledge develops through continued scrutiny, confirmation, replication, and correction rather than through treating publication as final validation.
Watch Out
"Peer reviewed" means that work passed through a particular review process. It does not mean every claim has been independently verified or that the article is permanently correct. Evaluate the actual methods and evidence.
Unpublished research can still be rigorous
A study may remain unpublished for reasons unrelated to methodological quality. The research team may lack time to prepare a manuscript. A student may complete a thesis and leave academia. A journal may reject a paper because it does not fit its scope. A sponsor may produce an internal report. The findings may be considered insufficiently novel for a particular journal.
None of these circumstances changes how the original study was conducted.
Likewise, publication cannot rescue poor research. A methodologically weak study remains weak after publication.
Research quality and publication status are therefore related only indirectly.
Publication decisions involve more than research validity
Editors consider whether a manuscript fits the journal's scope and audience, whether the contribution is sufficiently important or interesting for the venue, whether reporting meets journal requirements, and whether the article competes successfully for limited editorial resources.
A rejection therefore does not necessarily mean that the underlying investigation was not research or that its findings are invalid.
Conversely, acceptance does not establish that every conclusion is correct.
Researchers should distinguish methodological evaluation from editorial selection.
Publication matters because research is cumulative
If research does not need publication to exist, why publish it at all?
Because knowledge develops collectively.
Other researchers need access to findings to evaluate claims, conduct replications, avoid unnecessary duplication, refine theories, design new studies, synthesize evidence, and identify unresolved questions.
A study known only to the research team contributes much less to this collective process than one that is responsibly communicated.
Dissemination therefore does not define research, but it strongly affects whether research can contribute to the broader knowledge system.
Nonpublication can distort the visible evidence base
One of the most consequential problems occurs when the probability that research becomes visible depends on what the study found.
If studies reporting statistically significant, positive, or striking findings are more likely to be published than studies reporting null, negative, or inconclusive results, the published literature can provide a distorted picture of the underlying evidence.
This phenomenon is commonly discussed as publication bias.
The Cochrane Handbook treats non-reporting biases as an important threat to evidence synthesis because studies or results may be unavailable depending on the nature or direction of their findings.
Publication bias therefore affects more than the career of an individual researcher. It can alter what later researchers, practitioners, and policymakers believe the evidence shows.
“Negative results” can still be valuable results
A well-designed study does not become worthless because it fails to support the researcher's preferred hypothesis.
A result showing little evidence of an expected effect can constrain theory, prevent unnecessary investment in an ineffective intervention, motivate better measurement, or inform future study design.
The interpretation must remain statistically and methodologically appropriate. Failure to reject a null hypothesis, for example, should not automatically be translated into proof that no meaningful effect exists.
The broader point is that the value of research should not depend solely on whether its findings are exciting.
This is closely connected to why confirmatory and replication research remains valuable even when results reinforce rather than overturn existing knowledge.
Selective reporting can occur even when a study is published
Publication alone does not guarantee that the complete research record has become visible.
A study may measure several outcomes but report only favorable ones. Researchers may conduct many analyses and present only those producing interesting results. A multi-study project may publish some experiments but not others.
Transparency therefore concerns not only whether a paper exists but whether the report accurately represents what was done and found.
Prospective registration, protocols, reporting guidelines, data and code sharing where appropriate, and explicit identification of exploratory analyses can help address different aspects of this problem.
Preprints make research visible before formal journal publication
A preprint is a scholarly manuscript made publicly available before formal journal publication, typically before or during peer review.
Preprints can accelerate communication, establish a public record of work, invite feedback, and make findings accessible while journal review proceeds.
They should not be mistaken for peer-reviewed journal articles unless they have subsequently undergone that process. Readers need to evaluate their status and evidence accordingly.
The existence of preprints makes the distinction between research, dissemination, peer review, and publication particularly visible. The research can exist first, become publicly disseminated as a preprint, undergo peer review later, and eventually appear in a journal.
A thesis or dissertation can communicate research without a journal article
Student research provides another clear example. A doctoral dissertation may contain substantial original research and be deposited in a university repository without every study being converted into a journal article.
The research does not wait for journal acceptance before becoming research.
Whether a degree requires publication is a separate institutional question. Some programs impose publication requirements or expectations; others assess the thesis or dissertation directly.
This distinction connects directly to when student work counts as research.
Research funders may require dissemination regardless of journal publication
Researchers may have obligations to report results to funders, registries, participants, institutions, regulators, or the public. Requirements vary by research type, jurisdiction, sponsor, and funding agreement.
For example, NIH's Public Access Policy requires covered manuscripts resulting from NIH-funded research to be submitted to PubMed Central and made publicly available under the applicable policy requirements.
Clinical trials can also be subject to registration and results-reporting requirements independent of whether investigators later publish journal articles.
Researchers should therefore distinguish "not published in a journal" from "no reporting obligation exists."
Ethical obligations can favor making results accessible
Research involving participants asks people to contribute time, information, biological materials, or accept research-related risks partly so that knowledge can be generated.
Suppressing findings without adequate reason can therefore raise ethical concerns, particularly when nonreporting wastes participant contributions or distorts the evidence base.
The World Medical Association's Declaration of Helsinki states that researchers, authors, sponsors, editors, and publishers have ethical obligations concerning publication and dissemination of research results and that negative and inconclusive as well as positive results should be made publicly available.
The Declaration applies specifically to medical research involving human participants, so its requirements should not be generalized automatically to every discipline. Its underlying concern illustrates why responsible dissemination can have an ethical dimension as well as an academic one.
Publication should not be confused with impact
A paper can be published and never read. Another research output may influence policy, professional practice, software, education, public understanding, or later research without accumulating many citations.
Publication makes dissemination possible but does not guarantee visibility, use, citation, or societal impact.
These are separate stages and outcomes in the life of research.
Research can be complete methodologically while dissemination remains incomplete
A useful distinction is to ask two questions:
Has the investigation been conducted sufficiently to support its research conclusions?
Have those conclusions been communicated sufficiently to contribute to the research record?
A project can receive different answers to those questions. Research may be completed but poorly disseminated. A manuscript may be published while the underlying study remains methodologically weak.
Separating these dimensions prevents publication status from becoming a substitute for evaluating research quality.
