Descriptive statistics is the process of summarizing data in both a quantitative and visual manner. In a quantitative sense, typically this involves highlighting the main elements of a collection of data, for example, calculating the mean, median and mode. It is through the construction of figures, such as histograms and pie charts that data can be visually represented. Essentially, the overarching goal with descriptive statistics is to characterize data so that the basic quantitative features can be properly understood.

In descriptive statistics, the method which is utilized to express data depends on the type of data which is being analyzed. Simply put, data can be separated into two general groups: categorical data and numerical data. When dealing with categorical data, the variables which are measured are usually either nominal or ordinal in nature. A nominal variable is one representative of a category, with no associated numerical value. For example, data collected on the different colours of hummingbirds in a sample would represent nominal data. However, ordinal variables do have associated intrinsic values, usually classified as levels or rankings. For example, pretend that in the hummingbird scenario only yellow hummingbirds were measured in terms of the intensity of yellow which they possessed, either being low, medium or high. Therefore, these three levels of intensity would represent ordinal variables.

In comparison to categorical data, numerical data classifies variables as being either continuous or discrete. Discrete variables are whole numbers, whereas, continuous variables can resemble any number within a range.

Descriptive statistics is an integral part to the analysis of data because it allows data to be represented in a visual way, in which patterns and outliers can be identified. The use of descriptive statistics is useful for any study which aims to describe the characteristics of the different measures being collected. Additionally, it is critical to understand that descriptive statistics does not aim to formulate inferences which extend beyond the sample data being studied to the population at large. Rather, it functions to solely evaluate and visualize the immediate data collected.