Are Expert Tutors A Thing?
In my last post, I pointed out a few strange facts: 1) Tech companies are spending billions to build AI tutors for every child on Earth. 2) They are certain that 1:1 tutoring is the “gold standard” for learning. 3) They have no interest in professional tutors.
If you wanted to study the best practices of medical experts, you’d study doctors. But there aren’t any academic studies of professional tutors. Why not?
One uncomfortable possibility is that there’s no such thing as an expert tutor. The student knows something or he doesn’t. If he doesn’t, you tell him. Now he knows. Therefore, anybody who can provide explanations can be a tutor. A “professional” is just someone who does this all the time. He isn’t any better than the college kid down the street.
Of course, I think there’s a lot more to being a tutor. Two students can have the same score on the test, and even miss the same questions on that test, and a good tutor’s sessions with them will be completely different. If you eavesdropped, these are the kinds of phrases you might hear…
“How were you thinking about that one?”
“If I drew this line, would that help?”
“What’s your life like this week? How much time do you have for homework?”
Every one of these questions will generate a unique answer, forcing a unique response from the tutor. The idea is not to impose one “best” approach on the student. It’s to create a model of how the student is thinking and then offer the minimum assistance possible, with the goal of triggering an “Aha!” in the student, so that he gets excited and solves it himself.
And there’s more to tutoring than leading a student to mastery of individual question types. The expert tutor also has an understanding of what motivates the student, and how that spark can be fed, and sustained, and fed more, until it’s a fire that consumes large, dense logs that would have snuffed the spark earlier.
Sadly, I can’t prove any of this – there aren’t any studies. I can only tell you that I’ve experienced the power of great 1:1 teaching as a student, and I’ve seen it in dozens of other tutors.
But even if I could prove this expertise exists, I don’t think the tech executives would care. And maybe they are right not to. In my next post, I’ll explain why.
Just last night I was reminded of the importance of the details of reaction. I can usually tell when a student has each one of the following reactions to the moment when a question of what-is-the-answer/how-to-understand-the-answer (the "move on" moment) is reached:
1) I'm tired of this and want to move on;
2) I feel a little bad not understanding this so I'll say I understand it;
3) I understand it (but I don't understand it);
4) Oh, I see, now I think I understand it, kinda;
5) I get this and I'm not going to forget it; and the best of all
6) OMG my new understanding of the law of cosines is going to change my life forever!!!
You know, because you are a tutor, it is mainly the facial expression. Some students are inscrutable (or stuck on #1) and you can't really tell, but most react. However, they all react in their own way, using face, body expression and words, which requires a couple/few sessions to totally get (by the tutor) partly because you won't get all of these reactions in a single session, and it usually takes the student a period of trust-building to feel comfortable enough to not be putting it on most of the time. I try very hard to get that going (the trust thing) right away, but it can take time.
Anyway, take that, AI! Sure, you can do that, but somebody might need to tell you that you need to do that, and it is going to use so many resources that your keeper (the tutor builders) are going to turn that option off because it is too expensive.