Non-native English writers often produce manuscripts with correct grammar but word choices that make reviewers doubt the conclusions. The problem is not that the research is weak. The problem is that certain words signal imprecision, overconfidence, or logical errors that undermine the credibility of otherwise solid work.
These five categories of misused words appear across manuscripts at every career stage. Each section includes a direct replacement strategy and a before-and-after rewrite.
1. Overcertainty Words: prove, confirm, demonstrate
Words like prove, confirm, and demonstrate assert that a hypothesis has been established as fact. In most experimental contexts, this claim is not justified. A single study supports or suggests a conclusion; it does not prove it. Reviewers trained in scientific reasoning react immediately to these words, often noting that the authors appear unaware of the limits of their own evidence.
Demonstrate belongs in a separate sub-category: it is acceptable for describing what a method, figure, or procedure shows, but not for characterizing a statistical result or a research conclusion. Writing “Figure 2 demonstrates the distribution of responses” is fine. Writing “these results demonstrate that the drug is effective” overstates what the data can support. The word show is the standard replacement in the latter context.
Overclaims:
These results prove that the intervention reduces inflammatory markers in patients with type 2 diabetes.
The data confirm that early intervention improves prognosis.
Revised:
These results suggest that the intervention may reduce inflammatory markers in patients with type 2 diabetes.
The data support the view that early intervention improves prognosis.
Before/After rewrite:
Before:
Our findings confirm that higher vitamin D levels lead to better surgical outcomes.
After:
Our findings indicate that higher vitamin D levels are associated with better surgical outcomes.
2. Causality Words: cause, lead to, result in
Causal language states that one variable directly produces another. This claim requires evidence from a randomized controlled trial or a well-designed mechanistic experiment. Observational studies, cohort studies, and cross-sectional analyses can establish association, not causation. Using cause, lead to, or result in in an observational context will be flagged by reviewers, sometimes as a basis for rejection.
Overclaims causation:
Increased sedentary time causes higher fasting glucose levels in middle-aged adults.
Stress leads to elevated cortisol secretion in this patient population.
Revised (observational context):
Increased sedentary time is associated with higher fasting glucose levels in middle-aged adults.
Stress is linked to elevated cortisol secretion in this patient population.
Before/After rewrite:
Before:
The mutation leads to impaired mitochondrial function and subsequent cell death.
After:
The mutation is linked to impaired mitochondrial function and subsequent cell death.
Note: if the study design supports a causal inference (e.g., a mechanistic experiment with sufficient evidence), causal language may be appropriate. Apply this replacement primarily in observational research.
3. Statistical Words: significant, insignificant
Significant carries two meanings that are consistently conflated. In everyday English, it means important or large. In statistics, it means the result crossed a predefined threshold (typically p < 0.05). Using significant without qualification leaves readers uncertain which meaning is intended.
Insignificant is the more dangerous word. It implies that the finding has no value, which is a scientific judgment, not a statistical one. A result that does not reach statistical significance may still be clinically meaningful, hypothesis-generating, or informative for future meta-analyses. The correct phrasing describes the statistical outcome without passing judgment on the scientific value.
Ambiguous or incorrect:
The difference in mortality rate was insignificant between the two groups.
We observed a significant improvement in quality of life scores.
Revised:
The difference in mortality rate did not reach statistical significance between the two groups (p = 0.12).
We observed a statistically significant improvement in quality of life scores (p = 0.03), with a mean increase of 8.2 points on the SF-36 scale.
Before/After rewrite:
Before:
The effect of treatment on biomarker levels was significant and the placebo effect was insignificant.
After:
The effect of treatment on biomarker levels was statistically significant (p = 0.01), while the placebo group showed no statistically significant change (p = 0.43).
4. Comparison Words: compare to, similar, different
Three related errors cluster around comparison language.
First, compare to versus compare with. In scientific writing, compare with is standard for examining differences between two things. Compare to is a rhetorical construction used to draw a likeness. Most scientific comparisons require compare with.
Second, similar and different without prepositions. Both words require a preposition to be grammatically complete: similar to, different from. Writing “the results were similar” or “the expression was different” leaves the comparison incomplete and reads as an error in formal academic writing.
Incorrect:
We compared the outcomes of technique A to technique B.
The expression levels in the treatment group were similar and different compared to the control group.
Revised:
We compared the outcomes of technique A with those of technique B.
The expression levels in the treatment group were similar to those in the control group at baseline but different from those at week 12.
Before/After rewrite:
Before:
Survival rates were similar in both groups, but complication rates were different.
After:
Survival rates were similar between both groups (82.1% vs. 83.4%), but complication rates were different between the laparoscopic group (14.2%) and the open group (6.8%).
5. Degree Adverbs: dramatically, extremely, very, highly
Degree adverbs amplify claims in ways that reduce rather than increase credibility. Dramatically carries a theatrical quality inappropriate for scientific reporting. Extremely and very are informal intensifiers that belong in conversation, not manuscripts. Highly is overused: phrases like “highly significant” and “highly effective” have been repeated so often in the literature that they no longer communicate precisely.
The most effective strategy is to replace adverbs with specific data. When a quantitative comparison is not available, neutral alternatives such as markedly, substantially, and considerably are appropriate.
Informal or imprecise:
The treatment dramatically improved survival rates in the experimental group.
Protein expression was very high in tumor samples.
Revised:
The treatment markedly improved survival rates in the experimental group (hazard ratio 0.42, 95% CI 0.31-0.57).
Protein expression was substantially higher in tumor samples than in normal tissue (mean fold change: 4.7).
Before/After rewrite:
Before:
Inflammation markers were extremely elevated in the treatment group, and the drug response was very rapid.
After:
Inflammation markers were substantially elevated in the treatment group (mean CRP 48.3 mg/L vs. 12.1 mg/L in controls), and the drug response was rapid (median time to response: 3 days).
Checklist Before Submitting Your Manuscript
- Have you used prove, confirm, or demonstrate to describe your results? Replace with support, suggest, or indicate.
- Does your study design support causal claims? If observational, replace cause, lead to, and result in with associated with or linked to.
- Is every use of significant qualified as statistically significant? Have you replaced insignificant with did not reach statistical significance?
- Does every compare use with, not to? Do similar and different have prepositions (to and from)?
- Are there any dramatically, extremely, or very that could be replaced with specific data or a more precise adverb?
Word choice at the level described here is one of the things professional manuscript editors catch that authors miss precisely because they know what they meant to say. If you want a second set of eyes on your manuscript before submission, ScholarMemory provides professional editing for medical and life science researchers. Contact us at contact@scholarmemory.com.