This chapter builds upon and further develops the concepts and strategies described in Ch.6 of
Mother and Child Health: Research methods.
We have so far considered situations where the outcome variable is numeric continuous and
the explanatory variables are a mix of numeric and categorical ones. But in clinical work we
often wish to evaluate the effects of multiple explanatory variables on a binary outcome
variable. For example, the effects of a number of risk factors on the development or otherwise
of a disease. A patient may be cured of a disease or not; a risk factor may be present or not; a
person may be immunised or not, and so on. The situation now is different from multiple
regression analysis which requires that the outcome variable be measured on a continuous
scale. For example, in the data file on Crime in the Unites States, the outcome variable ‘Crime
Rate’ was continuous, and among the explanatory variables one (North/South) was
categorical, the rest being all numerical. In the data file on Scores, the outcome variable
‘Score’ was continuous, and both the independent variables (Sex and Teaching Method) were
categorical. When the outcome variable is binary, and one wishes to measure the effects of
several independent variables on it the method of analysis to use is Logistic Regression. The
binary outcome variable is coded 0 and1. The convention is to associate 1 with ‘success’ (e.g.
patient survived; risk factor is present; correct answer is given, and so on), and 0 with
‘failure’.