Dissertation test results. Help!!!

#1
Hi Guys,

I was just wondering if you could help me out with some statistics stuff as I'm having a little trouble with interpreting my dissertation test results in a mathematical way.

Basically, I have done some testing on reported presence scores when changing different audio representations in virtual reality. The presence and audio realism scores vary from 1-7 for each of the possible scenarios in the virtual world (Mono, Stereo or Binaural, each with and without ambiance). In total that is six scenarios.

I want to test how well each subjects presence scores and realism scores match the complexity of the audio representation. I was thinking of using linear regression to do that, so I could put the scenarios in some order of complexity (and maybe analyze ambient scenes and non ambient separately). So i could represent mono as a 1, stereo as a 2 and binaural as a 3 and then see if there is a positive gradient with reported presence?

I have used a two way Anova on my presence data to test the significance of the effect of spatialisation and ambiance on presence and I think I have got some fairly suitable statistics but I would like to see how well each individuals data correlates with the increase in complexity. This is so that i can see how well each persons listening experience correlates with how well their presence scores correlate with increased audio spatialisation (if that makes sense). I'm not totally sure how to go about that.

My apologies for a badly written question, I'm in a bit of a panic.

Would you be able to give me some advice?

Many thanks,

Ross