The authors carried out a detailed analysis comparing the students who enrolled in the on-line and classroom versions of the MS CS, finding the on-line students did as well or a little better. They also found higher completion rates than for MOOCs.
The authors are enthusiastic about the potential of on-line education for mid-career training. However, the author's knowledge of the history of distance education (DE) appears to be limited. They describe Georgia Tech’s Online M.S. in Computer Science, as "... the earliest model to combine the inexpensive nature of online education with a highly-ranked
degree program ...". While this appears to be an excellent program (as is the subsequent MS in Analytics), there have been inexpensive quality DE programs for several decades. An early example well reported in the literature are those from Open University UK (OUUK).
It has been known for decades that those who already have higher education qualifications and are working in a field, make better DE students. Also courses which charge money are known to have lower non-completion rates than those which are free, or change a token amount. Also it was known that on-line courses produce slightly better learning outcomes than campus based courses.
In addition, it appears that the authors of the study have missed the obvious point that those enrolling in a graduate computer science program are likely to have computer skills above those of the average student and so make better on-line students. The results for these students should not be taken as typical of the general population.
ps. Recently I completed a on-line Masters of Education at Athabasca University, (MEd). This program has been offered for more than twenty years. Most of my fellow students had a background in education, while mine is in computing (and teaching computing). When it came to writing essays, my English-teacher colleagues had the edge, but in navigating the on-line environment, my computing skills came to the fore.
Goodman, J., Melkers, J., & Pallais, A. (2016). Can Online Delivery Increase Access to Education? (No. w22754). National Bureau of Economic Research. Retrieved from http://scholar.harvard.edu/files/joshuagoodman/files/rwp16-035_goodman.pdf