# Types of Data

Many different types of data can be collected for statistical analysis. Understanding the differences between different types of data is important because the types of statistical analyses one can perform, and the claims that can be made on the basis of these analyses, differ depending on the type of data in question.

The most important distinction is between categorical and numerical data. Categorical data do not have numbers associated with them, but instead have only category labels. For example, in a survey asking which political party respondents were planning to vote for, each response would take the form of a name of a political party, and thus the resulting data would be categorical. One could calculate the percentage of respondents planning to vote for each party, but it would not be possible to calculate the ‘average vote’, since there is no way to take an average of different political parties.

Contrasting with categorical data is numerical data, where the data takes the form of numbers. For example, measuring the height of a sample of people, or the amount of time an animal spends grazing in a particular location, would result in numerical data, since each value would be a number associated with a single measurement. Mean, variance, and other such values can be calculated for numerical data which cannot be computed for categorical data.

Note that sometimes it is possible to collect and present data in either categorical or numerical form. For example, a survey could ask households to report their annual income as a dollar amount, yielding numerical data, or it could ask them which income bracket they fit into, resulting in categorical data. Although the underlying information being collected is similar, the different forms that the data can take has important implications for the sorts of calculations that can be done and statistical tests that can be conducted.