The ideal group work maybe has people learn better.
I've never had a group project in my life where we all learned that well. Usually, a few of us did most or all the work and even when a functional group did happen, the work was distributed so each person had a part of the whole picture and was missing out on the other parts. It only works if people share and want to learn--- when the group finishes the task, hardly anybody is interested in picking up on what parts they were not exposed to. Perhaps one could facilitate the ideal conditions but that is never done and I'd not be surprised if we had little on how to properly facilitate the desired outcome (and I'm not talking policy solutions which is all people ever discuss, having students rank each other etc are policy BS that is extremely limited.) For example, an alternative approach could be to have a series of group assignments which force rotate their roles; or even better... you have them rank each other or you rank them... then next step you purposely put all the weak ones in the worst positions and grade them as a whole... that would force them to help each other!
I wonder how they study the benefits of group work because the studies seem to always back up the theory. If you measure it wrong you'll end up with the same results and it's entirely possible the common techniques used produce biased results until the day somebody proves they have better techniques to study such things... Before that happens, most people will continue to adhere to the conventions they learned in school and even religiously stick to them even after they begin to be dis-proven.
I'm skeptical group work is so great; also, previous experience tells me that even a valid result gets overly generalized and over applied. It's like telling somebody to reboot their computer when they have a problem so then every time their ISP goes down they reboot their computer until the internet works again. I had a client who called me every time (for years) their ISP was down because their website didn't work! I'd have to prove it each time by having Microsoft.com not work either. ("oh, well if microsoft doesn't work either it must be a big problem..." heard that a few times too!) Now imagine having somebody that thick headed in a group project...
The last group study I read was teaching programming. Showing that pairs of students do better. not 1, 3,4,5 but 2 people did best. I didn't feel confident in the results because I wasn't given any idea what they tested them on. Depends on the kind of work and the kind of metric used if working in pairs helped or not. If the test involved nearly the same kind of thing as the assignment then the poor student could just recall and BS their way to a better score from being exposed to the solution, without a greater understanding... it's understanding that is the goal. (or was trends seem to be in the other direction. thinking must be too dangerous.)
Your education SUCKS if you can't tell the difference between an online course and a classroom course. think about that. Also if your high school is offering college, it means your high school sucks and so does the freshman portion of a college education. I remember taking AP and testing out; I took the course anyway-- AP really is a scam... if you don't get it, then your unaware or your education sucked.