ABSTRACT

Effective technical writing requires precision in the use of terminology. The word ‘significant’ is frequently misused as meaning only ‘important’ or ‘noteworthy’. This article demonstrates how the term should properly be used to indicate ‘statistical significance’. The process of hypothesis testing is necessary in order to determine statistical significance.

Introduction

Many words or phrases convey both an ordinary, everyday meaning and a precise meaning that is often highly specific to a particular profession or area of academic or technical study. For this reason, it is necessary to operationalize the concepts we use in academic and professional reports, i.e. define them so that they can be measured and/or expressed quantitatively or qualitatively. In sum, effective writing for academic and professional purposes requires a level of precision above commonplace usage.

Does ‘significant’ mean important?

One word that is frequently used in scientific/technical reports without the necessary specificity is significant. The everyday meaning of significant is something like, “sufficiently great or important to be worthy of attention” (Oxford Dictionaries). Now, consider the following statements drawn from a student’s case study:

“Zachary has significant difficulties with syntax.”

 “His comprehension shows a significant delay.”

Are we to interpret these accounts as merely indicating that Zachary has a noticeable difficulty with syntax, and a noteworthy delay in comprehension development? The difficulty in interpretation is because, in scientific/technical reports, the word significant has a precise meaning:  it refers to statistical significance.

Statistically significant

Statistical significance is a measure of whether or not the result of an assessment (or experiment, or correlation between variables, and so on) could have occurred by chance. If it can be shown that the result is highly unlikely to have occurred by chance alone then the result is said to be statistically significant. Statistical tests are performed to determine whether or not results are statistically significant and they are an integral part of hypothesis testing.

Does ‘statistically significant’ mean important?

Suppose we found a statistically significant difference of one point in the average standard scores for reading comprehension between a group of 11-year-old children with dyspraxia and a group of 11-year-old children with no speech-language difficulties. Whilst this result may be of theoretical interest it really has no practical importance. In most classroom situations it is highly unlikely that a teacher would recognise any difference between the two groups, as the difference is so small as to be detectable only through statistical testing.

Effect of sample size

As a general rule, the larger the sample sizes the smaller the differences that can be detected and any differences shown to be statistically significant. Equally, a small difference between groups may be statistically significant within a large sample size of, for example, 200 but the same difference may not be statistically significant with a smaller sample size of 40. For large sample sizes then, small differences between groups may be detectable and shown to be statistically significant whereas, for small sample sizes, even large differences may not.

So, detecting a statistically significant difference between groups does not necessarily mean that either the difference is large or that it has practical importance.

Recommendation

Effective writing in the social sciences – whether technical reports, student assessments, academic essays, and so on – requires a level of specificity above that encountered in everyday, commonplace communications such as emails, letters to friends, notes to colleagues, and similar. As well as appropriate editorial control of the material, it is necessary to give due consideration to the terminology used and to ensure that this is as precise as possible. This article has highlighted the use of the term significant – commonly used in technical texts. In everyday, informal use, significant is typically interpreted as meaning important or noteworthy. However, in scientific/technical writing it has a highly specific meaning, referring to statistical significance.

In order to avoid confusion, therefore, I would recommend that the word significant is only used in scientific/technical texts when one needs to indicate that appropriate statistical tests have been performed in order to determine whether or not the results of assessments, experiments, and so on, are statistically significant. In this way, the informed reader will understand that the author has carried out the process of testing a hypothesis and arrived at reasoned, and appropriately supported, conclusions.