Analyzing 3 or more paired scale variables
Impression
We are curious to see if the scores on multiple scale variables are different from each other. Although usually an impression of the data is done with a frequency table, or cross-table, with scale variables this might not be so effective (the table would get too large, and no longer give a good impression).
Instead some statistical measures might be more helpful. The mean (average) and standard deviation of each scale variable can be helpful. The mean is a measure of central tendency, which most people are familiar with. The standard deviation is a measure of dispersion. It indicates how much variety there was in the data. For more information on the mean and standard deviation see the 'Center & Dispersion' section at the one variable scale part of this site.
From the sample data an overview of the mean and standard deviation per category is shown in Table 1.
| category | n | mean | standard deviation |
|---|---|---|---|
| thriller | 10 | 7.00 | 2.36 |
| horror | 10 | 4.20 | 2.62 |
| comedy | 10 | 7.20 | 1.40 |
| adventure | 10 | 7.50 | 1.18 |
Click here to see how to determine the mean and standard deviation with SPSS, with R, or with Excel.
with SPSS
There are a two different ways to determine multiple means and standard deviations with SPSS.
using Frequencies
using Descriptives
with R
with Excel
From Table 1 we can see that 'adventure' has the highest average grade, and is also where most respondents agreed with each other (low standard deviation). For 'horror' this is the exact opposite.
However, numbers can also be deceiving. Therefor it is always a good idea to also inspect the data using a visualisation. Which visualisation is discussed in the next section.
3+ Scale variables
Google adds