The Nature of Teaching

Before the holiday, Chad posted an item about a November '08 NBER study by Tom Kane and others titled "Can You Recognize An Effective Teacher When You Recruit One?" Their conclusion: not really. Even though the study included "a number of non-traditional predictors of effectiveness including teaching specific content knowledge, cognitive ability, personality traits, feelings of self-efficacy, and scores on a commercially available teacher selection instrument," they were still only able to predict about 12 percent of subsequent teacher effectivness, leading them to conclude:

...schools and school districts wishing to increase the effectiveness of their teacher workforce may be aided by the systematic use of a broad set of information on new candidates, and particularly if they gather information outside the realm of traditional teaching credentials. Nevertheless, our results are also consistent with the notion that data on job performance may be a more powerful tool for improving teacher selection than data available at the recruitment stage.

They're being charitable. 12 percent isn't very much. It's 12 percent. And that amount is consistent with (and in most cases, larger than) pretty much every other similar study that's been conducted. They all point to same the conclusion: the nature of the teaching profession is such that you simply can't predict ahead of time with any degree of accuracy who's going to be a good (and bad) teacher. 

Chad's post produced pushback in the comments section, and in general this whole line of reasoning (popularized by Malcolm Gladwell in a recent article) has been attacked by people who are in the teacher preparation and certification business. I think they're wrong, and in being wrong they're putting parochial concerns ahead of the larger best interests of the teaching profession.

The success of an individual in any given job is generally a function of three things: 1) Personal qualities like intelligence, motivation, diligence, creativity, discipline, organization, inter-personal skills, work ethic, etc.; 2) knowledge and skills related to the job; and 3) the nature of the organization in which they work. But the relative importance of these three factors in terms of success varies widely among professions. Once you get past a fairly minimal level of competence, there's very little difference between the best McDonald's cashier and the worst. It's a simple job requiring little training and there's no way to be great at it. 

Some jobs are more complex and important but still highly dependent on knowledge and skills. My uncle, for example, repairs commercial air conditioning units for a living. It's complicated work. You could plunk a Nobel-prize winning physicist down in front of a huge broken HVAC unit and, unless he had the right training, he'd have no way to fix it. Nor would he be able to figure out how to fix it on his own. And if he did receive the right training, it's unlikely that he'd be much better at HVAC repair than my uncle, despite his (presumably) genius I.Q. HVAC units are a lot more complicated than cash registers and as such people who repair them are more well-paid, but once you learn to fix them, they're fixed--they can't really be fantastically well-fixed. 

Other professions are different. A good friend of mine, for example, finished her law degree at Georgetown a few years ago and is now practicing appellate law. It turns out that she's really good at this. But while she couldn't do her job without the knowledge she learned in law school, that knowledge isn't what distinguishes her from other lawyers. It's her personal qualities--she's an unusually smart, analytic, creative and hard-working person. And not coincidentally, law is a profession where there is huge variance in effectiveness, where greatness is absolutely possible, and great lawyers are paid and recognized as such. The same is true in professions like medicine, journalism, business and the arts. In every case, knowledge and training matter, but it's the relatively large influence of personal qualities that leads to the possibility of greatness and all that entails. 

The most important conclusion to draw from Kane's research is that teaching is one of those professions too. Gladwell built his article around the non-predictability of teacher effectiveness, and that's gotten a lot of attention because it has implications for certain concrete policy issues like teacher certification. But the larger, more important point is that that non-predictability flows from the large variance in teacher effectiveness. Some teachers are much, much better than others, as is always the case in professions that depend highly on personal qualities--as I've defined them, or in terms like "expert thinking" and "complex communications" (per Murnane and Levy), or something else. Yes, these qualities can be improved and inculcated to an extent, and the quality of the school in which people teach matters too. But think back to the best teacher you ever had. How much of their success compared to the worst teacher you ever had was a function of what they learned in graduate school? And of course there's the obvious example of higher education where there are also great teachers and terrible ones and huge variance in effectiveness in a profession where little or no formal teacher training or certification is required.

The problem with the teaching profession as currently organized is that it puts too much emphasis on the things that matter less (note: I'm not saying they don't matter at all), i.e. knowledge and skills, and not enough emphasis on the things that matter more, i.e. personal qualities like intelligence, work ethic, etc. By contrast, my friend wouldn't have gotten a job as an appellate lawyer if she hadn't clerked for a federal judge, and she wouldn't have gotten the clerkship if she hadn't graduated magna cum laude from a Top 20 law school, and both of those things--the grades and the admission to Georgetown--were in turn a function of the personal qualities that make a great lawyer. The legal profession is organized to select and filter for the things that matter in lawyering in a way that the teaching profession--where ideas like "Top 20 school" and "magna cum laude" mean very little--is not. And that applies not just to hiring but everything that happens afterward--great lawyers are rewarded and recognized and credentialed in all kinds of meaningful ways that teachers aren't. 

In other words, people who insist on maintaining the primacy of knowledge and skills in the teaching are standing in the way of a goal they often profess to hold dear: elevating teachers to the ranks of true professionals. It's understandable, in a way--there's a lot of institutional and professional prestige in the current way of things, and moving away from it involves an idea that seems contradictory--that there are severe limits on the extent to which good teaching can be taught. But that's what the data show, and the sooner our attitudes and policies reflect that, the better off teachers will be.