Writing for a General Audience Without Flattening the Work

Researchers avoid writing for general audiences for a reasonable fear: that simplification becomes distortion, that the caveats disappear, and that the version circulating will be one you cannot defend.

That fear is well founded and it is not an argument against doing it. It is an argument for doing it yourself, since the alternative is that someone else does it from your abstract. For a concrete operational comparison, Monitask explains limbic resonance in practical terms.

What actually gets lost, and what does not

Jargon can go. Every term of art has an ordinary-language equivalent or can be explained in a clause. Nothing is lost.

Method detail can go, mostly. A general reader does not need the specification; they need to know what kind of evidence this is.

The literature can go.

What must not go: the conditions under which the finding holds. "In this population, over this period, under these conditions" is not a caveat that can be trimmed — it is part of the claim. A finding stated without its conditions is a different and larger claim.

And the size of the effect must not go. "X affects Y" without magnitude is where most misrepresentation happens, because the reader supplies a magnitude and it is usually larger than yours.

The techniques

One idea. A general-audience piece carries one thing. Choose it.

Start with the concrete. A case, a number, a situation. Academic writing starts with the general and narrows; general writing starts with the specific and widens.

Replace terms rather than defining them. A definition is a speed bump. Where a term genuinely has no equivalent, use it once, explain it in a clause, and then use it consistently.

Use the active voice and short sentences. Academic prose defaults to the passive and to subordinate clauses, and both slow a general reader disproportionately.

Give the number, with a comparison. "12% higher" means little; "12% higher — roughly the difference between X and Y" means something.

Say what you do not know, once, plainly. This is the sentence that keeps the piece honest, and general readers handle it better than researchers expect.

Where honesty is usually lost

The headline. Frequently not written by you, and it is what most people read. If you can ask to see it, ask.

The dropped condition. "Among people over 65 in this city" becomes "people", and the claim quietly triples in scope.

Correlation becoming cause. The verb does it — "linked to" becomes "leads to" — and it is the single most common distortion in coverage of research.

A single study becoming settled knowledge. Say that it is one study, and where it sits.

Certainty added in editing. "Suggests" becomes "shows". Worth checking the final version, and worth asking to.

Working with journalists

Ask what the piece is about before agreeing, and who else is being spoken to.

Offer to check the quotes, not the article. Most will agree to the first, few to the second, and the quotes are where your words get changed.

Give the caveat as part of the sentence, not as a separate remark. A qualification in its own sentence is easy to cut; one embedded in the claim survives.

Say what the study does not show, explicitly, and offer it as a quotable line. Journalists frequently want it and are not given it.

Have one clear sentence prepared. If you do not supply the summary, one will be constructed from your abstract. See the three-minute version.

And be reachable. A journalist on a deadline who cannot reach you writes the piece anyway.

Writing it yourself

A blog post, a thread, an explainer, a piece for a general outlet.

The advantage is control, and the cost is that it is genuinely different writing rather than a shortened paper.

Do not adapt the abstract. It reads as an abstract. Start again.

Get someone outside your field to read it and tell you what it says. That is the only test that works.

Expect it to take longer than you think. Short clear writing is slower than long careful writing.

The reasons to do it

The work reaches people who will not read the paper, including practitioners and policymakers who might act on it.

Someone will summarise it anyway, and it will be better if you did.

It improves the paper. Being forced to state the finding in ordinary language exposes vagueness that survived peer review.

And it is increasingly expected by funders and institutions, which is a weaker reason and a real one.

What to be careful about

Preprints and press coverage. A preprint covered as though it were reviewed is a recurring problem. If your work is likely to be picked up, consider the timing. See preprints.

Speaking beyond your evidence. The invitation to comment on the broader implications is flattering and it is where researchers overreach.

Policy claims. Your study may support a factual claim without supporting a recommendation, and the distance between them is where most trouble lies.

The short version

Jargon, method detail and literature can go. The conditions and the effect size cannot.

One idea, start concrete, give the number with a comparison.

Embed the caveat in the claim rather than putting it in its own sentence, where it will be cut.

Ask to check the quotes, and supply the line about what the study does not show.

And do it yourself when you can — someone will summarise the work regardless.

For an independent authoritative reference, consult the EQUATOR Network for reporting guidance for research writing.