Washington University School of Medicine

Measurements in clinical phenomena can be categorized into four types of data: nominal, ordinal, discrete, and continuous. Nominal data have no inherent order, while ordinal data have some ranking but no specified intervals. Discrete data are whole numbers, and continuous data can take on any value. It is important to classify data correctly in order to choose the appropriate statistical test for analysis. Measurements of clinical phenomena yield four kinds of data, nominal, ordinal, discrete, and continuous. This figure demonstrates those different types of data. There are two types of categorical or qualitative data, known as nominal and ordinal, and there are two types of numerical or quantitative data, often referred to as discrete or continuous data. Nominal data occur in categories without any inherent order. Examples of nominal data are shown here. ABO blood types, for example, are dramatic and discrete events, such as death, dialysis, and surgery. The data can be placed in categories without much concern about misclassification. Nominal data that are divided into two categories, for example, present or absent, yes or no, or in this case, pregnant or not pregnant, are referred to as dichotomous data. Ordinal data possess some inherent ordering or rank, such as small to large or good to bad, but the size of the intervals between categories is not specified. Some clinical examples would include, as shown here, tonsil size, where tonsil size can range from 1 to 4, being the largest. Discrete data can take on whole integer values, such as number of pregnancies or number of seizures, but does not allow for increments of less than whole integer. Continuous data can take on any value in a continuum, regardless of whether or not they are reported that way. Examples include most serum chemistries, weight, blood pressure, partial pressure of oxygen in arterial blood, as some examples. This figure shows the types of reporting of data as categories, rank order, equal spacing, and true value according to types of clinical phenomena measurements. An important reason, and perhaps the most important reason to classify data, is to direct the selection of statistical tests. This figure shows how the type of predictor variable and outcome variable directs the type of statistical test, as shown here. So, in order to pick and select the correct statistical test, you need to understand the type of variables you're analyzing.