Data can be divided into two broad categories:
Qualitative
Qualitative data are categorical rather than numeric. However, be careful because it is fairly common practice to code data using numbers to represent categories, for example 1 = male and 2 = female. Nominal data are categories with no meaningful order, such as individual students on a class roster. Ordinal data do have a meaningful order, such as students' level of satisfaction with their instructor: Very Unsatisfied, Unsatisfied, Neutral, Satisfied, or Very Satisfied. With both types of qualitative data, the degree of difference between data points cannot be determined. In other words, we have no way of actually measuring the difference between Satisfied and Very Satisfied.
Quantitative
Quantitative data are numeric and measured on a continuous scale. Interval data have an arbitrary zero value and the degree of difference between data points can be measured. For example, a temperature of 0°C is not the absence of temperature, its just cold, and the difference between 0°C and 10°C can be calculated as 10 units. Ratio data do have a meaningful zero value and the difference between two data points can be calculated as a ratio. For example, if there are 0 dollars in a bank account, there is an absence of money, and if Person A has 5 dollars and Person B has 10 dollars, we can accurately say that Person B has twice as much money as Person A.
The type of data will determine the best method for visual enconding, see Best Practices.
