Column #18, posted 3/17/03

Admission and Standards in an Online Program

When I first presented the idea of offering Master's programs online, explaining the potential numbers and market involved, the provost of Carnegie Mellon quipped that maybe we should put up golden arches over the campus that said over a million served. I didn't and do not see this as all that far-fetched, and it does bring to light some significant issues.

The first issue is that of standards. The second is that of maintenance of quality. The third is that of brand name. I will discuss these in order.

Every academic institution is concerned with standards. We don't take just anyone into our esteemed institution — you have to work hard to get in. Parents of college bound students obsess about the grades and test scores of their progeny for this reason. Everyone admires those who get into Harvard and Yale for this reason. How competitive a university is remains a big part of how it is perceived. U.S. News and World Report regularly ranks universities using average SAT scores of entrants as an important factor. When I suggested eliminating SAT's as an admissions critierion at Northwestern, I was told that this would severely affect the rankings and therefore could not be considered.

For these reasons, universities consistently concern themselves with how hard it is to get into their schools. A certain number of places need to be filled in every entering class, and it is a luxury to be able to reject applicants. It is a lot easier to get into some graduate programs at Harvard than others. It depends on how many are applying and how good the competition is. There usually is no absolute standard. It is simply a competition amongst those who meet the minimum standards.

When is comes to online education, one of the key variables has changed. The number of freshmen admitted into Harvard is determined by issues such as available dorm space, available classroom space, and available faculty. Harvard has convinced itself that only so many people should get into Harvard because it likes the kind of institution that creates. They could easily triple in size without causing quality to decline, but they don't want to do that. That is their decision, but that decision has its consequences, and these need to be discussed.

I was asked to join the Board of Editors at Encyclopedia Britannica ten years ago. Everyone else was about eighty years old. I was the computer guy in a room of people who had never used a computer. The Board believed that the Encyclopedia was just the right size. They worked on determining what was in and what was out. That was their job. I asked the Board if they would consider quadrupling the size of Encyclopedia if it made it no harder to produce or deliver and there were no negative financial consequences. They looked at me horrified, and said that it was just the right size now, and that they would never do such a thing. I told them the Encyclopedia would be all but dead in a few years then. The Web barely existed at that time, but it was clearly on its way.

A similar thing is happening with online education. Why not let everyone who applies in? There are no dorm rooms to worry about, and no class size or classroom issues. Faculty need not be tenured. In fact, they can and should be drawn from the best practitioners in the world. When location has no meaning in education, a lot of things change.

A university can imagine that its faculty are the best and that increasing its size would decrease its quality, but, especially in a practical field, in Master's programs, this is unlikely to be the case. Would-be practitioners are often best taught by current practitioners. Notions like clinical faculty and teaching hospitals exist in medical schools and other professional schools for a reason. The academics are not necessarily the best in practice. The size of the student body should be a function of the number of available mentors drawn from people who have practiced or are practicing the skills being taught.

So, who should be admitted? Anyone who can do the work, is the obvious answer. But, it should come as no surprise that not everyone who applies is equally prepared. In a teamwork based program, this can have serious consequences. On the other hand, in the real world, not every team member is equally prepared or equally skilled either. Dealing with such issues is part of how we learn to function in a field.

We have decided to admit anyone who has demonstrated sufficient experience and skills. For those for whom we cannot make this judgment, we have devised the pre-MSIT program. This program has three flavors. Working backwards, we have a short course one can sign up for. If a student succeeds at that course, if he or she can deliver the product that we teach him or her to build, then that student is admitted into the MSIT program of their choice. If the student is not ready for that course, we have a course that prepares him or her to handle it. And, if a student really know nothing about computer science, we have devised a full year program, the successful completion of which will qualify him or her for admission into the one year MSIT program. To put this more simply, there is a two year MSIT program that anyone who has a college degree can get into.

Why a college degree, you might ask? Simply because Carnegie Mellon is not ready to relax that requirement. Of course, I do not see why there should be any such requirement at all.

Maintenance of quality that is the issue, not admissions. We should concern ourselves with the quality of the output, not the quality of the input. A school should not brag about how hard it is to get in, but how hard it is to get out. The job of the faculty is to teach, after all. There probably is a point where one cannot find sufficient numbers of people who can mentor students and thus produce quality output. We have not yet met that point, however.

This leads to the issue of branding. What happens if thousands of people have Carnegie Mellon software engineering degrees? Wouldn't they suddenly be worth a lot less on the open market? This is an odd question for a university to ask, as it is simply an issue of supply and demand, and ought not concern to a major university as long as people want to sign up for the degree program. But, it does get asked.

To put this another way, Harvard likes it, in a deep way, that very few people can get into Harvard. Having a Harvard degree is seen as something special. This elitist notion seriously harms education in general, however. If Carnegie Mellon can teach many more people to be good software engineers, then an obvious consequence of this is that the world's software systems are more likely to work. Banks will function better, space shuttles will launch, airplanes will be less likely to collide, and so on. In a world so dependent on computers, isn't it a good idea to make as many highly qualified software practitioners as possible? It makes me crazy that this question is asked. It is clear why it is asked. Places like Harvard don't see it as their mission to better educate the world. If everyone had a high quality education, might business be more ethical, might there be less terrorism, might politicians be more effective? It sure couldn't hurt. Universities need to change their view on this. Online education has the power to make the world a better place if universities stop thinking about brand name and start thinking about their role in the larger world.