Complete newbie needs advice. Can this dissertation data have a statistical test applied to it?

Hi. When I set out on this dissertation, stats were the last thing on my mind. I don't think it will harm my overall marks, but I'd hate to miss out on an opportunity to try out a bit of statistical analysis. It's certainly well out of my comfort zone, however. Below is a picture of some of the quantitative data I collected. It's just a bog standard Likert questionnaire. Is there any test for significance I could apply to it?

Many thanks


Data for forum.png


Fortran must die
If you mean a phd type dissertation than changes late in the day would make me wonder. I would stick to what I had agreed on with my committee and do the rest after you graduated :p That is purely a personal opinion.

What is your dependent variable and your very general theory? It determines answers to your question.
Hi. It's a Masters dissertation and there's no expectation that I have to use a statistical analysis. To be honest, most of the analysis will come from thematic analysis of qualitative data. There isn't a hypothesis as such. The study is looking at the impact of a school's curriculum on the self-determination (SD) levels of six autistic children. I have some 'before and after data which could maybe be analysed with a chi-squared perhaps?

It's anonymous so I've included the names. SD is their level of self-determination as a percentage.

Data for forum 2.png

I've also got raw data that shows their capacity to develop SD and opportunities to develop SD.

Data for forum 3.png
If you think I can do anything with these please let me know, and thanks for taking the time to reply.



TS Contributor
You can compare the respective dependent variables between time points using the Wilcoxon signed rank test. It will tell you whether (e.g.) capacity 2019 tends to be higher (or lower) than capacity 2018. Since your sample size is very small, only large changes will produce a statistically significant result, though.

With kind regards