Statistics – measurement level

 In Statistics & methods

What is one of the most important thing to know, before you start your statistical analysis? My answer is : measurement level of your variables. If you define your variables into “Nominal”, “Ordinal” or “Scale” before you start analyze your data, then you get automatic help in some of the commands like: graphs and Custom Tables.

You can see the definition in the column “Measures” in the Variable View for example:

The most important difference between these measures is that “Nominal” and “Ordinal” are categorical variables, and “Scale” is numeric.

So look at the picture above: Salary is typically “Scale” with numeric values that can be used to calculate numeric information like mean values. But you cannot do any mean values of job categories, only frequencies.

To define “Nominal” and “Ordinal” correct, you must know the meanings of the codes, so have a look in the “Variables-tool”:


If you have a look on the variable “Job_category”. You have 7 codes, but the codes have not any order because you cannot say that code 5 is more than 4, for example. So then it’s defined “Nominal”:

Then if  you have a look on the “Agegroup”-variable. You have 3 codes, but these codes have an order, because you can say that 3 is more than 2, and 2 is more than 1. So then it’s defined as “Ordinal”:

When you have defined the measurement levels then you will get a lot of help in some of the commands, like the “Chart Builder” command.

Greetings Gunilla Rudander


error: Content is protected !!