Note: I edited this post extensively after my first draft -- in trying to keep it readable and jargon-free, I ended up misrepresenting a research concept and the importance of certain studies' results.
There is some great discussion and commentary on this thread. I'll try not to be redundant or simply answer the question with, "It depends".
However... it depends
There is a lot of nuance in a comparison of student outcomes and student persistence in online vs face-to-face courses and programs. A lot of that nuance is lost in aggregate analyses that can not control certain variables. For those reasons, some people believe a straight, comparison across modes in aggregate is not appropriate. Some people believe any such comparison tends to focus us on the wrong things. I don't disagree.
Still, if you decide you want a straight comparison of online vs f2f for learning outcomes or for persistence rates I think it's worth pointing out two big pitfalls re. variables:
1. Different instructional methods -- i.e. the teaching or learning method, materials, or assessments in one course are not exactly the same as the other. If this were true one would expect to see no significant difference in learning outcomes. This plays out in the classic book by T.L. Russell, (surprise) The No Significant Difference Phenomenon, and is generally summarized by the position Richard Clark took in the "media debate" of the '80s and '90s -- i.e. "[...] media are mere vehicles that deliver instruction but do not influence student achievement any more than the truck that delivers our groceries causes changes in our nutrition".
If we are trying to compare online vs f2f, then presumably mode of delivery is our independent variable, and we control the other variables -- like instructional method -- as much as possible. If we don't, things tend to fall apart. One recent example of this that sticks in my mind is from a small 2014 study by Caldwell-Harris, Goodwin, Chu, & Dahlen which compared f2f and video courses and found that students in the f2f did *better* on the test than those students who watched recorded videos of the exact same lecture. The exact same lecture. What's somewhat hidden is that in the live, f2f setting, the teacher interacted with students. Even just eye contact and physically approaching students may have been enough of a difference in treatment, imo, to explain the difference in scores.
2. Different students / samples -- sometimes students in one course are not comparable to students in another course due to different levels of background knowledge, ability, engagement, motivation, financial stability, etc. This is hard to control, as we can't deprive a student of their choice in whether they take a course online or f2f. Indeed, student choice in mode is part of what makes online and blended learning so powerful.
Sample differences may explain findings like this study from 2018, which found, "Students that persist in an online introductory Physics class are more likely to achieve an A than in other modes. However, the withdrawal rate is higher from online Physics courses."
Weird, right?
I haven't dug into it, but I'd expect to see those factors in play in the study you mention re. California Community Colleges that found overall students who took online courses performed less well. These aggregate studies rarely compare one course to another with the same That doesn't mean those results are wrong, but they do no more likely they point to deficits in the design of online courses, e.g. some say it's easier to deliver a bad online course than a bad face-to-face course because at least f2f requires that both teacher and students to show up.
Re. persistence, you note the 2013 study, Does online learning impede degree completion? found, "that controlling for relevant background characteristics; students who take some of their early courses online or at a distance have a significantly better chance of attaining a community college credential than do their classroom only counterparts".
Those aside, I am aware of a couple of resources that may help you:
Good luck!
This disucssion post is outdated and has been archived. Please use the Community question forums and official documentation for the most current and accurate information.