Thanks for sharing Glenys!
I'm familiar with this particular DoE meta-analysis. My take from it is that the researchers concluded that the learning and teaching approaches used had a more significant effect than the modality of communication, i.e. face to face vs. online.
Advocates from one camp or the other can happily "cherry pick" research evidence to support their claims. Overall, I think the evidence is quite reasuring that online learning is at least as effective as face to face.
Something you might find interesting is a research paper on a 3 year blended learning project that uses principles of Socio-Cultural Theory quite ambitiously:
Designs for Collective Cognitive Responsibility in Knowledge-Building Communities: http://www.tandfonline.com/doi/abs/10.1080/10508400802581676