Independant samples T-Test

Question

I gathered data using Likert scale questions 1,2,3,4,5, which I then inputted into SPSS and carried out an independent samples T-Test analysis. The T-Test results I then have to discuss, when it shows they are significantly different in variance and mean I need to discuss the mean as they are significantly different. However the means are e.g. 3.22 1.98 and I was just wondering if these can be round up or down to single values to make more sense of the rating scale?

Answer 1

T-tests really are meant to be used with interval data not the means of likert responses. You may want to look into doing a Mann-Whitney test for comparing likert means.