Data measurement scale

Introduction

Data can be of a qualitative or quantitative nature. Qualitative data is also called nominal data, which exists as discrete, named values without a natural order amongst the values. Examples are different languages (e.g. English, Swahili, Dutch), different soil types (e.g. sand, clay, peat) or different land use categories (e.g. arable land, pasture). In the map, qualitative data are classified according to disciplinary insights, such as a soil classification system represented as basic geographic units: homogeneous areas associated with a single soil type, recognizable by the soil classification.

Quantitative data can be measured, either along an interval or ratio scale. For data measured on an interval scale, the exact distance between values is known, but there is no absolute zero on the scale. Temperature is an example: 40 ◦C is not twice as hot as 20 ◦C, and 0 ◦C is not an absolute zero.

Quantitative data with a ratio scale do have a known absolute zero. An example is income: someone earning $100 earns twice as much as someone with an income of $50. In order to generate maps, quantitative data are often classified into categories according to some mathematical method.

In between qualitative and quantitative data, one can distinguish ordinal data. These data are measured along a relative scale and are as such based on hierarchy. For instance, one knows that a particular value is “more” than another value, such as “warm” versus “cool”. Another example is a hierarchy of road types: “highway”, “main road”, “secondary road” and “track”. The different types of data are summarized in Table.

Table: Differences in the nature of data and their measurement scales.
Measurement scale Nature of data
Nominal, categorical Data of different nature / identity of things
(qualitative)
Ordinal Data with a clear element of order, though
not quantitatively determined (ordered)
Interval Quantitative information with arbitrary
zero
Ratio Quantitative data with absolute zero

Outgoing relations

Incoming relations

Learning paths