 I believe that higher ed is at a very critical moment here. New models are going to be required to have higher ed be sustainable in the future. We need to be better about what we do, and I believe analytics is one of the keys to that. Her enthusiasm for analytics is well-founded. It's currently a hot topic in higher education, and rightly so. It has the potential to transform the way we teach students, it can help us make informed decisions about resources, and it can even help us predict the future. But some have concerns around using analytics properly. Let me say one other thing about analytics that's really kind of orthogonal to everything we've talked about. A lot of the justification I've heard is really justification in terms of being able to identify at-risk students, help students that need help finding their way through a specific class or through the broader higher ed experience. And that's fine, I mean, again, how can you argue? But at the same time I think that there are some very important components of the university experience. Certainly there were for me when I went through it about learning to take responsibility for what you do, learning to sort of pace and take responsibility for your own learning and mastery of a body of material, and being given opportunities to fail. I'm not sure it's helpful to have things popping up all the time saying, you know, 60 out of 100 people who were in this particular configuration ended up badly, so you shouldn't do it because there, you know, there always are people who pull it off and sometimes learn something doing it. Some of the things that you learn the most at are things that maybe you fail at the first time. And you know, I just argue that there needs to be room in the system for that too. Another voice of concern comes from the director of professional development and innovative initiatives at Virginia Tech, Gardner Campbell. He says that in the realm of teaching students, using analytics might miss the underlying proof of learning. The danger I feel is that we're standardizing on a set of assumptions that are simply not correct. We're standardizing on the idea of memorization, repetitive behavior within a carefully proctored and managed online environment, like a learning management system. I say, what if we flip that? If we say it's an understanding fostering system, well, that starts to look different. It looks more like biological evolution. It looks more like the way Google has figured out how not only to support the web but to learn from it. Can that be quantified? Google is a great example of a place that has been doing analytics for a long time. But part of that is creative for them. They're looking for evidence of engagement, not simply trying to look at repeated behaviors. I think it was an early insight that a link represented not just an information transaction, but at least until the spammers got hold of it, a link represented a human judgment. Well, you get human judgments at scale. In something like the World Wide Web, you can begin to do some interesting things with the data you get from that. But let me speak even farther to the point you're asking about quantitative versus qualitative. I think it is true, I agree with Autul Gawande, that it's important to be able to count something. I'm absolutely not against quantitative analysis. I think it has its place and can be crucial. The question is what you're counting, of course. But leaving that aside, let's talk about the qualitative. This gets back to what I think is our highest ability as a species, which is what John Medina in Brain Rules calls this capacity for symbolization, capacity for symbolic representation. Douglas Hofstetter talks about analogy as the core of cognition. And he's talking about a certain ability to recognize patterns, regroup things that come in a series or events that follow each other in ways that begin to make meaning and construct knowledge together. I really don't know how far numbers alone will take us there. I think what we need to look at is, oh, if you frame it this way, if you group these elements this way, if you recognize these patterns this way, what does that liberate us into? Fuller human potential, greater human happiness? I hope so. I think that those are the goals. We're really talking about the human record. It sounds grand, but we're talking about civilization. And I don't think that you're going to be able to get there just by looking at behaviors in a narrow way. I think you really do have to look at this larger capacity for symbolization. Gardner Campbell and Cliff Lynch sharing their thoughts and concerns about analytics.