Scientific Writing Sample: An Annotated Research Paper Example

Draft:

Cover crops reveal a key mechanism for protecting water quality in agricultural landscapes

Revised:

Winter rye cover crops lowered drainage nitrate in maize plots over two winters

What a reviewer reads: "Reveal" and "mechanism" promise more than a field comparison can show. The study measured an outcome; it did not test how rye takes up nitrate. The revision names the crop, the outcome, the setting and the time frame, and claims nothing the data do not show. Our review flags a title that claims a mechanism the paper never tests, because the editor reads the title before anything else.

PLOS ONE, to take one journal, asks that the abstract "summarize the most important results and their significance" and "not exceed 300 words". Your target journal may set a different limit or a structured format.

Draft:

Nitrate pollution is a major global problem. In this study, we investigated cover crops. Results showed that cover crops significantly reduced nitrate. These findings have important implications for sustainable agriculture.

Revised:

Nitrate lost from cropland in winter drainage can push nearby surface water above drinking-water limits. We compared winter rye with bare soil on 24 maize plots (12 per treatment) over two winters. Mean drainage nitrate was lower under rye in both winters, and the difference was larger in the wetter winter. A cover crop is a practical way to cut winter nitrate loss on drained maize land, though plot-scale results still need testing at field and catchment scale.

What a reviewer reads: The draft has no numbers, no design and no limit, so a reader cannot judge it. "Significantly" without a test or an effect size is the vague-quantifier habit that the ICMJE recommendations single out: they ask authors to avoid nontechnical uses of statistical terms such as "significant". In a real abstract, put the actual means and intervals where the revision says "lower".

Most of an Introduction is context. The paragraph that decides whether a reviewer keeps reading is the last one, where the question appears.

Draft:

Therefore, more research is needed on cover crops. This study aims to fill this gap in the literature.

Revised:

Cover crops are known to take up residual soil nitrate, but most plot trials have measured soil nitrate rather than the drainage water that reaches streams. We therefore asked whether a rye cover crop lowers nitrate in tile drainage from maize plots, and whether the effect holds across a dry and a wet winter.

What a reviewer reads: "Fill this gap" names no gap. The revision says what earlier work measured, what it missed, and the exact question this paper answers. A reviewer can now test the rest of the paper against that sentence.

Draft:

Samples were collected regularly and analyzed using standard methods.

Revised:

Drainage water was sampled from each plot's tile outlet after every flow event between November and March, and nitrate-N was measured by ion chromatography; the analysis plan, including the comparison of winters, was set before the second winter began.

What a reviewer reads: "Regularly" and "standard methods" cannot be repeated by anyone. The revision gives the sampling rule, the season and the assay, and says which comparison was planned. The ICMJE recommendations ask authors to "distinguish prespecified from exploratory analyses, including subgroup analyses", and the wet-versus-dry comparison is exactly that kind of analysis.

Draft:

As shown in Figure 2, cover crops dramatically reduced nitrate levels (p < 0.05), demonstrating their effectiveness.

Revised:

Mean drainage nitrate-N was lower under rye than under bare soil in both winters (Figure 2; Table 1). The difference was larger in the wet winter, when drainage volume was also higher. Table 1 gives plot-level means with 95% confidence intervals and the number of sampled flow events per plot.

What a reviewer reads: Three problems sit in the draft's one sentence. "Dramatically" is an opinion. A bare p-value hides the size of the effect. "Demonstrating their effectiveness" belongs in the Discussion, if anywhere. The ICMJE recommendations also ask for absolute numbers alongside percentages, which is why the revision points to a table of means and counts. With 12 plots per arm, our review would also ask whether any subgroup split (by soil type, say) has enough plots to carry a claim, and would suggest labeling a thin split as exploratory.

The ICMJE recommendations suggest opening the Discussion by "briefly summarizing the main findings".

Draft:

Our results are consistent with previous studies [4-9] and prove that cover crops are the best solution for nitrate pollution.

Revised:

Over two winters, rye lowered drainage nitrate from maize plots, with a larger effect in the wetter winter. That pattern matches trials that measured soil nitrate [4, 5], and it extends them to the water that actually leaves the field.

What a reviewer reads: "Prove" and "the best solution" fail on any data, and a citation block of six papers hides which one says what. Our Full Review checks numbered citations against each cited paper's abstract (a capped sample on a long reference list) and marks each one checked as supported, partially supported or not supported. A sentence that leans on [4-9] for "prove" is where an unsupported citation usually turns up.

Draft:

This study has some limitations. Future research should address these.

Revised:

Two winters cannot separate a weather effect from year-to-year variation, and plot-scale drainage may not scale to a whole catchment, where groundwater flow paths differ. We did not measure rye nitrogen uptake, so we cannot say whether lower leaching came from uptake, from reduced drainage volume, or from both.

What a reviewer reads: The draft admits nothing, which reviewers read as either naive or evasive. The revision names three specific limits and, in the last sentence, says plainly which mechanism the study did not test. Stating the limit yourself takes the reviewer's strongest objection off the table. PLOS makes the same point in its guide to discussions and conclusions: naming your limitations "shows reviewers and readers that you have considered the weaknesses of your research."

The habits in the left column are the ones our Full Review is built to catch, because they are the ones reviewers punish. It flags figure and title wording that claims causation when the data show correlation, and words like "demonstrate" or "reveal" when the evidence is indirect. It treats groups smaller than ten per arm as exploratory unless the paper justifies them, and asks for the missing controls and alternative explanations a reviewer would raise. It checks whether the cited papers say what the sentences claim.

None of that is visible from inside your own draft, which is the point of a sample like this one. You can see the gap between the two versions here because the study is not yours.

The annotations on this page are Manusights judgment, not a prediction of any editorial decision. A paper written like the revised column can still be rejected for its science or its fit; one written like the draft column gives the reviewer easy reasons before the science is even weighed.

Run a free readiness scan on your manuscript to get a submission-risk signal and the single top blocker in roughly 2-3 minutes, with no card required. For the full treatment shown on this page, the $39 Full Review analyzes your paper one section at a time, checks citations against live scholarly sources, adds figure, methods and reviewer-risk feedback, and closes on a prioritized revision plan.

If you want structured teaching rather than feedback on one paper, our comparison of scientific writing courses covers the free and paid options.