Analysing a nominal and ordinal variable
Part 2: Visualisation of the sample data (compound bar-chart)
We got a first impression from the sample data on the previous page, but a visualisation might be easier for others in our report. My preference for visualising a nominal and ordinal variable in one chart, is a compound bar-chart (also known as a stacked bar-chart). The advantage of an ordinal variable is that there is a logical order to the categories in this variable (e.g. fully disagree to fully agree), and can visualise this by stacking those. In Figure 1 the compound bar-chart from the sample data.
Figure 1. Location vs. Motivated by teacher
Click here to see how you can create a stacked/compound bar-chart as above, with SPSS, with R, or with Excel.
using Chart builder
using Legacy dialogs
converting a cross table
From the chart we see the same things as we already established from the table. In the report I recommend using a ‘Introduce – Show – Tell’ approach. So when reporting this graph, it could be for example like this:
The teacher for the class was different at each location. Students were asked how motivational they found the teacher at each location. The results are shown in Figure 1.
From the sample it appears there are some differences between the three locations, but will there also be present in the population. This will be the topic for the next page.
Nominal vs. Ordinal