The errors that cost points are usually not the ones that get noticed
Fluent second-language writers often lose points on a small set of patterns: articles, collocation and register. None of them impede understanding, which is why they survive every proofread, and all of them are visible to a grader.
Articles, and the question that resolves most cases
The choice between a, the and no article depends on what the reader knows at that point in the sentence.
Use a or an when introducing something for the first time, or when any one of a class would do. Use the when the reader can identify which one you mean, either because it was mentioned already, because it is unique, or because the sentence itself specifies it. Use no article for plural or uncountable nouns in general statements.
All three, in sequence
We conducted a survey of rural nurses. The survey included 412 responses. Surveys of this kind are common in workforce research.
First mention takes a. Second mention takes the, because now the reader knows which survey. The general statement takes no article, because it is about surveys in general.
The pattern worth memorizing: a noun specified by a following phrase almost always takes the. The results of the study. The effect of scheduling autonomy. If the sentence goes noun plus of, it is nearly always the.
Collocation
These are the pairings English happens to use. They are not derivable from rules and they are what most reliably marks writing as non-native.
| Not conventional | Conventional |
|---|---|
| make research | conduct research, carry out research |
| take a conclusion | draw a conclusion, reach a conclusion |
| do an experiment | conduct an experiment, run an experiment |
| give attention on | pay attention to |
| discuss about | discuss |
| according to me | in my view |
| in the other hand | on the other hand |
| researchs | research (uncountable) |
| informations, evidences, literatures | information, evidence, literature |
That last row causes persistent trouble. Research, information, evidence, literature and feedback are uncountable in English: they take no plural s and no a. Write a piece of research, or several studies.
Sentence length and nominalization
Two habits often travel together in second-language academic writing: long sentences and heavy noun phrases.
Nominalized and long
The implementation of the intervention resulted in an improvement in the retention rates of the staff members who were working in the rural facilities.
Verbs doing the work
After the intervention, retention improved among rural staff.
The first version turns actions into nouns: implementation, improvement. English academic prose reads better when verbs carry the action. It is also shorter, which graders notice.
Hedging conventions
English academic writing hedges more than the academic conventions of several other languages, and less than some. Getting it wrong in either direction is noticeable.
Direct assertion where English expects a hedge reads as overclaiming. Conversely, stacking hedges reads as evasive.
Unhedged
The intervention improves retention.
Overhedged
It may perhaps be possible to suggest that the intervention might potentially have some effect.
Conventional
The intervention appears to improve retention, although the design cannot rule out selection effects.
One hedge, then the specific reason for it. That is the pattern.
Building a personal error list
Most writers repeat a small number of patterns. After each piece of feedback, record the corrections in a list grouped by type: articles, collocation, prepositions, tense.
Within a few assignments the list stops growing and the top three items will account for most of your corrections. Checking a draft against those three specifically is far more effective than rereading the whole document hoping to notice something.
Frequently Asked Questions
Because many languages do not have them, and because the rules depend on what the reader already knows rather than on the noun itself. The same noun takes a, the or nothing depending on whether it has been mentioned, whether it is unique, and whether it is countable in that use.
Which words conventionally go together. You conduct research rather than make research, and draw a conclusion rather than take one. Both versions are grammatical and only one sounds native. Collocation errors are the single most common thing that makes correct writing read as non-native.
Possibly. Writers working in a second language often overcorrect toward formality, producing long Latinate words and heavy nominalizations. English academic prose is more direct than many writers expect, and shorter words are frequently the better choice.
As a first pass, yes. It reliably catches agreement and some article errors. It is weak on collocation, on register and on the meaning-level problems that matter most, and it sometimes suggests changes that are wrong in an academic context.
It should not. Language editing corrects accuracy and register while keeping your argument, your structure and your voice. If an edit rewrites what you meant rather than how you said it, that is a problem with the editing rather than a necessary cost.
Read academic writing in your own field with attention to phrasing rather than content, and keep a personal list of the corrections you receive most often. Most writers have five or six recurring patterns, and knowing which are yours is more useful than general study.
Want the language checked without losing your voice?
Send the draft and we will return it with the language corrected and the reason beside each change, so the next one needs fewer.
Get My Free Quote