4.2 Descriptive analysis

Descriptive analyses are conducted for three key purposes:

1. to detect errors and clean data
2. to understand the distribution of the data and the relationships between variables
3. to generate hypotheses.

They are generally the first step in any data analysis and consist of characterising and visualising the dataset of interest. In this section, we will look at how to describe a variable depending on its type and how to describe the relationship between two or more variables. Aspects related to data visualisation will be covered in more detail in the module Summarising and presenting data.