What type of data are you using?
When deciding on the right chart you need to think about the type of data you are using. There are two types of data; categorical and numerical.
What is categorical data?
Categorical data as the name implies is data that has been grouped into categories. For example you could group peoples gender or race and then compare them with this type of data. It is also possible to use numerical values although they should not have a mathematical meaning. Pie and Bar charts are recommended for this data. Categorical data can be broken down even further into nominal and ordinal.
- Nominal data - This is used for labelling variables that do not have any quantitative value. A good example of nominal data is: What is your gender? M- Male or F-Female.
- Ordinal data - It is the order of the data that is significant here. A good example of ordinal data is to measure a concept like satisfaction, rating it on a scale from very poor to very good.
What is numerical data?
Numerical data is information that can be measured. A good example would be the number of people that attended the gym over the course of a month. To identify numerical data test if the data can be added together and also if the answer can be in decimal/fraction form. If the information has to be grouped into categories then this would be considered as categorical data. Line and scatter charts are recommended for this data. Numerical data can be broken down into discrete and continuous.
- Discrete data - This is data that represents units of measurement that cannot be split up. A good example of discrete data would be number of students in a class. This type of data is counted.
- Continuous data - This data is not restricted like discrete data, and can take any value within a range. A good example of continuous data would be a person’s height, as this could be any value. This type of data is measured.
Categories for your data
Before choosing your chart you need to first understand what message you are trying to convey with your data. Generally you are trying to show one of the four things below:
- Distribution - Shows a collection of related/unrelated information to see how it correlates, if at all. The best charts to use for this type of data are Scatter, Column, and Line charts.
- Composition - Collecting different types of data that make up a whole and displaying them together. The best charts to use for this type of data are Pie, Donut, Stacked area, Stacked column and Stacked bar charts.
- Comparison - Sets variables apart from each other, and shows how they interact. The best charts to use for this type of data are Line, Bar and column charts.
- Relationship - This type of data tries to show a correlation between two or more variables. The best chart to use for this type of data is a Scatter chart.