Identifying and Avoiding Bias in Research Papers
The words used in scholarly writing should always be chosen with care and sensitivity. Dehumanising language, for instance, should be avoided when writing about human beings, and words that acknowledge the presence or role of human beings in a study should not be omitted. There is a tendency, however, for participants to be reduced through a kind of shorthand to the condition they represent in a study: ‘diabetes and non-diabetes’ might be used, for instance, instead of the more humanising ‘participants with and without diabetes’ or ‘participants with and without a diagnosis of diabetes.’ While shorthand language is sometimes necessary to convey results efficiently, especially in tabular form, it should be avoided as much as possible and certainly not used when first introducing the people involved in a study.

Some publishers and journals may even frown upon the use of ‘subjects’ instead of ‘individuals’ or ‘people’ simply because it is too impersonal, and most will want the age of human participants and other people to be referred to accurately. Young men and women, for instance, should not be called ‘boys’ and ‘girls,’ which, as a general rule, should be used only of children 12 years of age and under. It is also important to keep the particular context in mind, however, and use common sense while considering each term. For example, while referring to a 25-year-old man as a ‘boy’ is inappropriate in most cases, referring to a 40-year-old prostitute as a ‘working girl’ may not be if that is what the prostitute calls herself and the term is used in quotations marks along with clear explanation.

Appropriate word use of this kind is a matter of achieving precision and avoiding bias. If, for instance, an author refers to a 30-year-old man as a ‘man,’ but refers to a woman of the same age as a ‘girl,’ or uses the masculine pronoun ‘he’ when writing of doctors and the feminine pronoun ‘she’ when writing of nurses without specifying a context and details that justify this treatment, it may not be a deliberate distinction, but it will come across as both inaccurate and biassed. Bias can occur in terms of race, nationality, sex/gender, class, education, age and so on, and can involve arbitrarily prioritising one group of people over another or stereotyping any particular group of people. Some readers might extend this to historical times and figures (the idea, for example, that any one time is better than another or the common notion that people now are more intelligent or more imaginative than people were in the past) as well as to animals and other creatures (with the prioritisation of people over animals or the environment, for instance, smacking of anthropocentrism).

Avoiding gender bias is particularly important in western (including English-speaking) societies of the twenty-first century, so it is essential to reflect on any instances in your writing in which you mention men or women alone. If women are the only subjects of the study or if women alone are relevant for a particular statement (only women can actually bear children, for instance), using ‘women’ alone is appropriate, but if both men and women are involved (both men and women can be parents, for example), both should be mentioned or an alternative that implies both (such as ‘parents,’ ‘people’ or ‘participants’) should be used.