Writing Science: Common Errors
A Note on Significance
The word significant is used in everyday language to convey a meaning of importance. In scientific writing, however, significant should be reserved for reference to statistically significant results. We must be aware of two factors when interpreting statistical results:
- If a result is not statistically significant, there is no difference. Statistics are used to mathematically determine if differences among numbers are real and are not confounded by variation, or other abnormalities, in the data. Therefore, it is important to recognize when a seemingly large difference in means, as an example, is not statistically significant.
- A statistically significant result may not be meaningful in practice. When a statistically significant result is found, it must be critically assessed to determine its applicability. For example, you might find a statistically significant difference between two numbers, but the difference might be too small to be meaningful to the study.
Finally, avoid terms such as marginally significant or slightly significant; a result is either significant, or not, as defined by the alpha-level established in your methods.
Back to Writing Science