The Format of Thesis Writing vs Journal Articles: Key Structural and Style Differences

The two are not interchangeable documents. A thesis is examined and carries its methodological detail in the body; a journal article is peer-reviewed, follows IMRaD, moves that detail into supplementary material, cites only directly pertinent references, and reports a single claim.

The thesis-to-journal conversion kit at the end has the comparison matrix, a one-claim worksheet and a pre-submission checklist, in Word and PDF.

Is the format of thesis writing the same as in a journal article?

No, and the gap is structural rather than cosmetic. Students often assume the format of thesis writing is the same as in a published paper, and the assumption survives until a supervisor writes "this reads like a journal submission" in the margin of chapter three. The comment is not about style. It is about what each document has to prove.

A thesis is assessed. Examiners are deciding whether you have demonstrated the competence a degree certifies, which means they need to see the reasoning, the instrument you built, the pilot that failed, and the analysis you rejected. A journal article is reviewed. Referees decide whether one claim is sound and worth the space, and everything failing to serve that claim is a cost to them.

That difference explains almost every structural rule below. Get it, and the rest stops feeling arbitrary.

Depth vs brevity: why your thesis requires methodological exhaustion

The shared standard is worth stating first, because students often think journals are simply more relaxed. They are not. The ICMJE Recommendations say the methods section should be detailed enough that others with access to the data would be able to reproduce the results. ICMJE writes for medical journals, and its house rules do not bind an engineering or humanities title, so read what follows as the clearest published statement of a shared convention rather than as your journal's law. Reproducibility is the floor in both documents.

What differs is where that detail sits.

A journal has a page budget and a reader who wants the argument. ICMJE's guidance handles the overflow explicitly: extra material and technical details can go into an appendix, or be published solely in the electronic version. So the calibration curve, the full instrument and the sensitivity analysis do not disappear from a paper. They move out of the running text.

A thesis has no such escape valve, and it has that second job. The pilot study that went wrong belongs in a thesis because it evidences judgement. It almost never belongs in a paper, because referees are not assessing your judgement, they are assessing a result.

Several further rules follow from the same logic, and each catches students who learned to write by imitating papers.

Methods are written from the protocol, not from hindsight. ICMJE states that the section should generally include only information available when the plan or protocol was being written, and that everything discovered during the study belongs in the results. Theses often blur this, narrating discovery in the order it happened. Examiners tolerate that; referees read it as muddled.

An introduction, meanwhile, is not a literature review in miniature. ICMJE asks authors to cite only directly pertinent references and to keep data and conclusions from the present work out of it. Chapter two of a thesis is expected to be comprehensive, because part of what is being examined is whether you have read the field.

Reference lists run the same way, against the instinct most students have. The guidance observes that fewer references to key original papers often serve as well as more exhaustive lists, whereas in a thesis bibliography exhaustiveness is closer to the point.

One piece of vocabulary separates people who have published from people who have not: reporting guidelines. ICMJE encourages journals to ask authors to write against the checklist for their study design, and names CONSORT for randomised trials, STROBE for observational studies, PRISMA for systematic reviews and meta-analyses, and STARD for diagnostic accuracy studies, with the EQUATOR Network as the place to find others. Find your checklist before you draft, not after a reviewer asks for it.

The seven-dimension comparison matrix

Print that table and mark the column that applies to whatever you are writing this week. Most of the confusion this article addresses comes from applying one column's rules inside the other document.

The table leaves out one element that causes more trouble than any row in it, because you have already written it. An abstract is not the same object in the two documents. ICMJE requires structured abstracts for original research, systematic reviews and meta-analyses, and expects them to state the study's purpose, basic procedures, main findings with specific effect sizes where possible, and principal conclusions, while noting important limitations and avoiding overinterpretation. Journals set their own headings and word ceiling for this, and the format varies between them. A thesis abstract is usually a single unstructured passage written to your institution's limit, and pasting it into a submission portal is a visible sign the manuscript has not been prepared for that journal. Rewrite it against the target journal's instructions, last, once the paper has settled.

Converting a 100-page thesis into an 8-page journal paper

Strategic condensation is not compression. You are not shrinking the thesis; you are extracting one publishable claim from it and rebuilding a document around that claim. Work in this order.

Choose the claim before you open the thesis file. Write one sentence stating what the field did not know before your work and now does. If you cannot, the problem is not the writing. Most theses hold one or two such sentences, occasionally three, never twelve.

Rebuild the introduction from scratch. Do not cut chapters one and two down. Draft afresh from the claim, citing only what a reader needs to understand why the question mattered, and keeping your own results out of it.

Sort the methods into two piles. Anything required to reproduce the result stays. Everything else becomes supplementary material, which ICMJE explicitly contemplates and most journals accept. Instrument validation, ethics documentation and the full protocol usually belong in the second pile.

Report primary and secondary outcomes, and nothing else. ICMJE asks for absolute numbers alongside derived figures such as percentages, and warns against duplicating the same data in both a table and a graph. A thesis results chapter that shows everything becomes a results section that shows what the claim rests on.

Write a discussion that does not repeat. Summarise the main findings briefly, state the limitations, and avoid conclusions the data will not carry. Repetition across sections is the single most common reason a converted thesis chapter reads badly as a paper.

Settle authorship before you write, not after

This is where supervisions turn sour, and the criteria are public. ICMJE recommends authorship rest on four conditions, all of which must be met: substantial contribution to conception, design, data acquisition, analysis or interpretation; drafting or critically revising the work; final approval of the version to be published; and agreement to be accountable for all aspects of it.

Read the corollary carefully, because it is the part students do not expect. ICMJE lists activities that do not by themselves qualify someone as an author, and the list includes acquisition of funding, general supervision of a research group, general administrative support, and writing or language editing. A supervisor who only supervised is acknowledged, not credited. Anyone meeting all four criteria must be listed, which cuts the other way too: the lab member who collected half your data and reviewed the draft is an author whether or not it is convenient.

Your deposited thesis and the prior publication question

Where a thesis is deposited in an institutional or national repository, as doctoral theses commonly are, the work becomes publicly accessible before any paper is submitted. ICMJE defines duplicate publication as substantial overlap with something already published without clear reference to it, and notes that prior publication may include release of information in the public domain.

Be precise about what that does and does not mean. The ICMJE page does not single out theses, and journal policies on this vary widely, so nobody can tell you from outside what your target journal will do. What the guidance does require is disclosure: where work has already been reported in large part elsewhere, the submission letter should say so and the authors should supply copies of the related material. Say it in the cover letter, attach the thesis chapter, and let the editor decide. Editors dislike surprises far more than they dislike theses.

If artificial intelligence tools were used at any stage, that too is a disclosure matter. ICMJE requires authors to describe such use, places writing assistance in the acknowledgements and study use in the methods, and states that chatbots should not be listed as authors because they cannot take responsibility for the work.

Where Sovi.AI fits

Condensation is reading work, and the reading is the slow part. Opening a 100-page file and deciding which paragraphs carry the claim takes a full day the first time you do it.

Sovi.AI Smart Writing accepts file attachments, and the file picker takes PDF, Markdown and plain text, several at once, so the thesis itself can be the input rather than pasted fragments. The prompt box is capped, which suits this task: the instruction stays short, the source document does the work. An Academic tone preset sits in the tone list alongside the everyday registers, and the second step on the page lets you supply a sample of your own earlier writing, so the output tracks your voice instead of drifting into a generic register.

Plan around two limits. Output arrives as text you edit and copy, with no file export, so budget a step to move it into your manuscript template. More importantly, condensation is exactly where numbers get rounded, hedges get dropped and a qualified finding turns into a flat claim. Check every figure, every citation and every limitation against the thesis before the draft goes near a submission portal. The formatting rules your institution sets for the thesis itself are covered separately in the guide to thesis format, font size and spacing.

The thesis-to-journal conversion kit

• thesis-to-journal-conversion-kit.docx for typing into.

Inside:

1. The seven-dimension matrix from this article, with an empty column for your target journal's author instructions.
2. A one-claim worksheet, which forces the claim sentence, the outcomes supporting it, and the material moving to supplementary, onto a single page.
3. A pre-submission checklist covering authorship against the four criteria, the reporting guideline for your design, the prior-publication disclosure, and any AI disclosure.

Complete Part 3 before you write the cover letter, since most of its items change what the letter has to say.

Frequently asked questions

1. Can I submit a thesis chapter to a journal without rewriting it?

Rarely, and the tell is the introduction, which in a thesis is written to survey a field and in a paper is written to justify one question. Expect to rebuild the framing sections even when the methods and results survive largely intact.

2. How many papers should come out of one thesis?

As many as there are defensible claims, which for most theses means one or two. Slicing a single contribution across several submissions to raise a count is recognised by editors and damages the work's standing.

3. Does my supervisor automatically become a co-author?

Not automatically. The ICMJE criteria turn on contribution, not on position, and general supervision by itself is listed among the activities that do not qualify. In practice most supervisors do meet the criteria, but the conversation should happen before drafting.

4. What goes into supplementary material rather than the paper?

Anything a reader needs only if they intend to replicate or audit the work: full protocols, instrument text, extended tables, code. Anything a reader needs to follow the argument stays in the paper.

5. My thesis is already in the university repository. Is that a problem?

It is a disclosure issue, not an automatic bar. Tell the editor in the cover letter, supply the related material, and check the journal's own policy on prior deposit before choosing where to submit.

6. Which referencing style should the journal version use?

The target journal's, without exception, and it is often not the one your institution required. Convert the references as the last step before submission, after the text has stopped changing.