I am doing an action research experiment with one grade 5 class as the experimental group and the other grade 5 class as the control group. My dependent variable is performance and my independent variable is new program. I cannot randomly assign the students to either group as they are already in set classrooms and I see them at different times of the day. I am giving both groups a pre-survey based on a six point likert scale and the same survey as a post-survey. I want to examine the differences in attitudes within each group and then compare them to each other. What kind of tests would I need to do? Z score? T of independent means, chi? :confused:

Actually, I have narrowed the test down to either a t-test of independent means or a ANCOVA. Anybody know which one would be the best?