Two different samples of students, different intervention, pre and post

#1
Hello again. I wanted to test if a course is effective by comparing precourse scores and post course scores. They told me I have to use a control group. This gives me less time and will make samples of students smaller. Can someone please point me to the right direction? When you compare medians you can use Mann Whitney, but when you compare pre and post differences what can you use? Thank you in advance.
 

Karabiner

TS Contributor
#2
It depends on the measurement level of your dependent variable.
If it is interval scaled, you can calculate individual differences pre-post
and compare them between groups using U-test or t-test. Or you
perform a repeated-measures analysis of variance with group as
between-subjects factor and time of measurement (pre, post) as
within-subjects factor.

With kind regards

Karabiner
 

hlsmith

Less is more. Stay pure. Stay poor.
#4
More so lately, I am partial to using quantile regression. This would allow you to compare the differences in medians across groups (i.e., intervention vs. control).