9.1 Types of data

To interpret and work with data you need to understand the different kinds of data. Certain statistical measurements can only be applied to specific data types. Broadly, data can be classified as qualitative (categorical) or quantitative (numerical) and within each of these types there are separate subgroups.

A chart showing the hierarchical nature of data. Data can be qualitative or quantitative. Qualitative data can be further characterised as nominal (for example blood type or ethnicity) or ordinal (for example pain on a scale of 1 to 10). Quantitative data can be characterised as discrete (for example a product of counting such as heart rate or number of offspring) or continuous (for example measured with infinite values such as height or weight).
Figure 9.1: Categorizing the different forms of data.

Qualitative (categorical) data

Qualitative data includes such things as survey data. This kind of data can be further classified as nominal and ordinal. Nominal data are just labels and have no order; that is, if you assemble nominal data as a list you would not change the meaning of the values. Examples of nominal data could be patient ethnicity, country of origin or blood type. A further subtype of nominal data is binary data where the value can be one thing or another such as biological sex. Ordinal data on the other hand has a hierarchy and could include variables such as satisfaction ratings, socio-economic status, or perception of pain on a categorial scale.

Quantitative (numerical) data

Quantitative data involves numbers and can be grouped into discrete or continuous data. Discrete data can be counted, like heart rate in beats per minute or number of patients in a clinical trial. Continuous data is measured rather than counted. A characteristic of continuous data is that it can be divided (at least in theory) into infinitely finer parts. Examples of continuous data could be patient weight or serum sodium concentration.

Share This Book