Variables in health research

If you do have a chance to do some research in your village, then you will need to know about what sorts of variables might be suitable for study. A variable is something that can take on different values. A variable may be in the form of numbers (such as weight or age), or non-numerical characteristics (such as occupation, ethnicity, education level or gender). It can be a characteristic of a person, object, place, event or phenomenon. For instance, a water filter is an object that exists in several different types (or variables), so you could study the effectiveness of different water filters as the variable in a research study on prevention of diarrhoeal diseases. The type of house that people live in is a variable 'place' that affects health, and an accidental injury is an example of a variable 'event or phenomenon' that you could study.

Variables may take numerical (quantitative) or non-numerical (qualitative) values. Measurements such as weight expressed in kilograms or pounds, height expressed in metres or centimetres, income expressed in dollars or birr, and number of children per family are examples of numerical or quantitative variables.

Qualitative variables are attributes described in terms of the presence or absence of certain characteristics, such as diseased or non-diseased, smoker or non-smoker, married or unmarried. Qualitative variables may also be descriptive, such as occupation or political views. In the question above, items A (marital status) and E (family support) are qualitative variables.

Knowing about variables is very important, because you will not be able to undertake useful research or discuss your research findings unless you understand which variables you are dealing with. The identification of variables as non-numerical or numerical determines whether the data is qualitative or quantitative, and this in turn determines the method of analysing the data.