Descriptive vs. Inferential Statistics

When analysing data, we can apply a descriptive and/or inferential approach. The key is to understand the purpose of each so that we can apply them appropriately.

Descriptive Statistics

Descriptive statistics is interested in (as the name implies) describing and summarising a dataset. It does not allow us to make conclusions beyond the data that we have analysed.

Generally, there are two types of statistic that can be used to describe the data.

• Measures of central tendency (mean, mode, median)
• Measures of dispersion/spread (standard deviation, variance)

The summary of the data is usually presented as a table, graph or statistical commentary.

Inferential Statistics

We do not often have the luxury of access to the entire population - it can be extremely costly or impractical to survey a large number of people. Inferential statistics allows us to make generalisations about the population that a sample was drawn from, assuming that the sample is representative of the population.

This technique is invaluable and has multiple applications, ranging from predicting election results to measuring the effects of various medical treatments.

The methods of inferential statistics are:

1. the estimation of parameter(s) and
2. testing of statistical hypothesis.