# Two-way multivariate analysis of ordinal data

• Sep 10th 2013, 04:10 AM
quizzical
Two-way multivariate analysis of ordinal data
Hi, I am wondering if you guys could help me with a problem I'm having?

I have collected some ordinal data (non-parametric) from an experiment. The results come from a questionnaire completed by participants who answered using 5 discrete categories (essentially 1-5).

Originally I collected the data over a period of four days under one condition and was planning to do a Friedman test to determine any statistical significance.

However, the data I collected raised another question. So using the same participants as before I decided to conduct the experiment again under a different condition. An example of the data I collected is:

 Day 1 Day 2 Day 3 Day 4 Condition 1 1 2 4 3 Condition 2 3 4 5 4

(Results are provided as median values, I would much rather work with mean values if possible(Speechless))

And this is where I'm stuck, what statistical test do I use to determine any statistical significance between the two conditions?

The internet doesn't seem to give me a straight answer here and I'm confused. If this was interval data I would typically use a two-way ANOVA, but I don't know what the ordinal equivalent is. Please help.

Thanks