Analysing two ordinal variables
If you have two ordinal variables you can either check if there is a relationship between them, or if both have identical scales to see if for example people changed their mind on something. This section of the website will deal with the first option; a relationship. If you are interested in change then look at the unpaired section for two ordinal variables found here.
To analyse a potential relationship between two ordinal variables, the following steps are done.
1) Get an impression of the sample data by creating a cross table
2) Visualise the sample data by using a heat map.
3) Test if the relation might still be present in the population, and how strong this effect is by using Goodman-Kruskal Gamma.
4) Write up the report.
For each of the described steps above a separate page is created. Let's begin with getting an impression from the data on the first page.