I would add that students thinking of pursuing this path should be aware that the fraction of students who say they want a BA or BS, start at a two year college, and actually obtain a BA or BS is really low. This is true even if you condition on lots and lots of observed characteristics, as in this paper by my student Lock Reynolds. The NYT story provides a handful of success stories. They could have found many more stories of failure.
In thinking about why this might be, I would highlight three factors: peer quality, instructor quality and living at home / hanging out with friends from high school. Peer quality may be more important indirectly, through its effect on the level and pace at which courses are taught than any direct effect via personal interactions. I teach a harder and faster econometrics course at Michigan than I did at Maryland because the students are, on average, stronger here. I am sure that the course at Washtenaw Community College is easier and slower than the one I taught at Maryland.
While that college in Massachusetts noted in the article may have people who used to teach at Harvard (as teaching assistants?), I suspect most community college faculty have very different observed characteristics than faculty at four year schools. This is likely a mixed blessing for students. In some cases, community college faculty may have more time and enthusiasm for teaching than the research-and-tenure distracted faculty at four-year schools. In other cases, they will be teaching part-time to supplement whatever else they are doing.
My sense is that part of what makes four-year schools different from two-year schools is that students usually, though not always, do not live at home when they attend them. This adds to the cost, of course, but it also helps marginal students break away from their high school peer group and mindset and to form a new network more heavily weighted with the academically talented and ambitious. I suspect that this has positive effects on outcomes, though of course exogenous variation in living at home to use in teasing out causal effects is hard to come by. I would be interested in pointers to the literature - surely there must be some - on this point.
Finally, some of the advice, such as paying very close attention to the sequences of courses required to attain one's educational goals within four years, carry over to students who start at four-year schools. Bound, Lovenheim and Turner convinced me that institutional barriers due to limited capacity have real effects on time-to-degree.
Hat tip: Nora Dillon