A manuscript in research is a structured, written document that reports the results of original research, experiments, or observations in a formal format designed for submission to a peer-reviewed journal. The term technically refers to the document before it’s published. Once experts in the field have evaluated it and a journal accepts it, it becomes a “published article” or “paper.” In practice, though, the words manuscript, article, and paper are used interchangeably in academic settings.
How a Manuscript Differs From a Published Paper
The distinction is really about stage. A manuscript is the working version of a research document: the draft that authors prepare, format, and submit. It goes through editorial checks, peer review, and usually multiple rounds of revision. Only after a journal formally accepts and publishes it does it become a journal article. Think of a manuscript as the pre-acceptance version and a published paper as the post-acceptance version of the same work.
There’s also a middle ground called a preprint. A preprint is a complete draft of a manuscript that the authors post publicly before formal peer review. Preprints get a cursory screening from the hosting server but no expert evaluation. Major preprint servers include arXiv for physics and math, bioRxiv for biology, medRxiv for medical research, and PsyArXiv for psychology. The key distinction is certification: a preprint has not been vetted through peer review, while a published article has.
Types of Research Manuscripts
Not every manuscript reports a new experiment. There are three broad categories:
- Original research articles describe new experiments, observations, or analyses. These make up the bulk of what journals publish and follow the most standardized structure.
- Reviews synthesize existing research on a topic. These range from narrative reviews (including editorials and commentaries) to systematic reviews, which follow a strict methodology, and meta-analyses, which combine data from multiple studies to reach statistical conclusions.
- Case reports document unusual or noteworthy clinical cases in medicine. They’re typically shorter and formatted differently than full research articles.
The Standard Structure: IMRAD
Most original research manuscripts follow what’s called IMRAD format: Introduction, Methods, Results, and Discussion. This structure became the dominant format in scientific publishing over the past several decades and is now expected by most journals. A manuscript is considered to use IMRAD when it includes clear headings (or synonyms) for methods, results, and discussion. The introduction is always present but doesn’t always carry a formal heading.
The introduction explains the research question and why it matters. The methods section describes exactly what the researchers did, in enough detail that someone else could repeat the experiment. Results present the findings, often with tables and figures. The discussion interprets those findings, addresses limitations, and places the work in the context of existing knowledge. Longer manuscripts sometimes use subheadings within the results and discussion sections to keep things organized.
What You Submit Alongside the Manuscript
A manuscript file alone isn’t enough. Journals require several supporting documents at submission. A cover letter introduces the work to the editor, explains why it’s a good fit for the journal, and highlights what makes the research significant. A title page lists the title, all authors, and their institutional affiliations. Most journals also require a financial disclosure statement detailing how the research was funded, along with a competing interests declaration noting any potential conflicts of interest.
Increasingly, journals require disclosure of artificial intelligence use during manuscript preparation. The International Committee of Medical Journal Editors (ICMJE) mandates that any AI involvement be clearly described, including the purpose and extent of use. Major journals like Nature, The Lancet, JAMA, and BMJ all require authors to name the specific AI tools used and explain what content they helped generate. The general framework: if AI helped establish hypotheses, write portions of the text, analyze data, or generate figures, that must be disclosed. If it was used only for grammar correction or sentence restructuring, disclosure is optional.
What Happens After Submission
The journey from submitted manuscript to published article follows a predictable sequence. First, an editorial production assistant runs quality checks on the submission, sometimes requesting additional information or corrections. If those checks pass, the manuscript is assigned to an editor who evaluates whether it fits the journal’s scope and quality standards. Many manuscripts are rejected at this stage without ever reaching peer review. JAMA Internal Medicine, for example, reported that 78% of manuscripts received in 2017 were rejected without review.
Manuscripts that clear the editor’s desk are sent to external expert reviewers. The editorial team works to secure at least two reviewers, often contacting several before enough agree to participate. Reviewers typically have two weeks to evaluate the manuscript and submit their assessment. Based on those reviews, the editor decides whether to accept the manuscript, request revisions, or reject it.
The odds vary dramatically by journal. Top-tier journals like the New England Journal of Medicine and The Lancet accept fewer than 10% of submissions. Mid-tier journals accept roughly 20% to 30%, and regional or specialized journals tend to have higher acceptance rates.
Why Manuscripts Get Rejected
The most straightforward reason for rejection is submitting to the wrong journal. If your work falls outside a journal’s scope, it won’t be sent for review regardless of quality. Beyond that, common reasons cluster around a few themes.
Study design problems are a frequent issue: a sample size too small to support meaningful conclusions, lack of proper controls, or using outdated methods when more robust approaches exist. Incomplete reporting is another major factor. Editors and reviewers need to fully understand and, in theory, replicate your experiments. If methodological details or data are missing, the manuscript won’t move forward.
Presentation matters more than many researchers expect. Manuscripts that lack proper structure, ignore the journal’s formatting guidelines, or include low-quality figures (pixelated or blurry images) can be rejected on those grounds alone. Poor English writing is a significant barrier for non-native speakers, particularly those without access to institutional editing services. And plagiarism, which journals now check automatically at submission, results in immediate rejection.
Perhaps the most avoidable mistake is failing to confirm that your research question is genuinely novel. Before writing, a thorough literature search across multiple databases can reveal whether your hypothesis has already been tested, saving months of work on a manuscript that would be rejected for lack of originality.