Quasi-experimetal Research is it necessary a control group? A short-term entrepreneurship course is gonna be implemented in one city in 10 schools. I would like to stablish a quasi-experimental research in order to measure the impact of this course on skills, entrepreneurship intentions, risk taking...
The course will last about two months with 8 lectures. There will be a precourse survey and post course survey .However, is it necessary to take a control group given that the course is not going to last too much, just one month? 
The experiment can not  be random given that they have to sign on it however it is expected that the most of the students will take the course. The expected sample will be about more than 400 children.
Thanks in advance
 A: There are two distinct issues here. The first is the meaning of the term "quasi-experiment". Quasi-experiments, at least in the way the term was used by Don Campbell who created it, includes numerous research designs, not all of which have a control group. The second issue is whether in this situation you need a control group for the study to be meaningful. A single pre-post design such as this is generally considered a very weak research design in the abstract (Campbell called it uninterpretable). However, there are contexts where it can provide some information regarding the effect of an intervention if you can assume (and it is an untested assumption!) that there would be little to no change in the absence of the intervention (i.e., that is, you are making an inference about the counterfactual or essentially assuming that if you did have a control group, the mean change for that group would be roughly zero). That is a difficult assumption to justify when your participants are children who are constantly changing.
