A clustered bar graph, also called a grouped bar graph, is visually very similar to a “regular” bar graph, with one important difference in the data being presented: a second categorical variable is used to further break down the existing categories, thus aiding in the representation of data.
A simple example of a clustered bar graph is shown to the right. Here, salaries for several different career types are graphed. Within each original category (career type), the data is categorized further based on a second categorical variable (gender). These two “gender” categories are created within each of the three “career type” categories, and each of the six categories-within-a-category has its own bar (with one bar missing, in this case, due to an apparent lack of data regarding female custodial workers). While the original categories are distinguished using the original labeling system along the x-axis, the second categorical variable is determined using a color scheme.
More examples of clustered and stacked bar graphs illustrate several different ways to represent data with more than one categorical variable. Note, above all, that a clustered bar graph is not always an appropriate way to represent two categorical variables. With that said, a clustered bar graph is often a very elegant solution when used properly: this discussion briefly presents, as one of its examples, a clustered bar graph as a solution to a complex data set. Also note its use of a second labeling system as a possible alternative to complementing the first labeling system with color.