Over 1800 courses were transitioned from f2f to distance due to COVID-19. The power calculation provided, I need a sample of 413 courses. My null hypotheses is there is no difference in pass rates from spring 2019 and spring 2020. It doesn't make sense to compare the means of pass rates. Can this be a paired t-test?
Evaluate the normality and variance of the data to determine if not to use u de mann withney, if you start with the previous hypothesis that there is a difference, this can be quantified to reinforce your results by means of the bayes factor for comparative t tests, now there are free programs like JASP or JAMOVI, that can perform these analyses in a simple way and reinforce your statistical conclusions.
Christian is on to something. You would want to use the Mann-Whitney-U test if the distribution turned out to be noticably non-normal. So check the variance and disyribution and if they are out of limits you should not use the parametric t-test. Good luck.
