Wading Through the Data Swamp:
Program Evaluation 201
Levels of Measurement

Variables are measured in different ways. How they are measured determines which statistical tests can be used.
Below are the different levels:
Nominal-level variables categorize qualitative differences. There is no quantitative meaning. For example, drug use can be a nominal variable. You may ask the participants in your program if they have used drugs in the past 30 days, "yes" or "no." You cannot say that yes is twice as much as no.
Have you used marijuana in the last 30 days?
(yes or no)
Ordinal-level variables are attributes that may be logically rank ordered. Each attribute represents more or less of the variable. For example, client satisfaction would logically be ordered from lesser satisfaction to greater satisfaction.
How satisfied are you with the program?
<----------------------------------------------------->
Not at all Somewhat Very
Interval-level variables have points at equal intervals. For example, a 1-year difference is always the same whether you are 10 or 50 years old. However, with interval-level variables, there is no true zero.
What is your age?
__13
__14
__15
__16
Ratio-level variables have equal intervals between the attributes and are based on a true zero point.
How many siblings do you have? __________








