Hello Rob, basically, I will have two groups both are control groups and I am conducting an experiment in which two corpus tools similar to each other in some respect are introduced to improve students’ English grammar. We are planning to give pre-test and post-test to both groups. Also, groups are nonrandomized. Actually, it seems to fit to quasi-experimental the matching only pre-post test control group design. We avoided from not training one group on the corpus and other without corpus.
Thanks for your reply