### Be Careful What You Ask For...

...because you just might get it. Now it looks like I really will be teaching the math class for Education majors this fall. But it comes at the expense of what might have otherwise been a really nice teaching schedule.

Two courses worth the same amount of academic credit might, from a student's perspective, require different amounts of work. The same is true when teaching: adding another section of a course you already teach (and have taught before) is going to require less time and effort than adding a totally new course. But they both count the same in terms of calculating the teaching load. Large enrollment and no grader? Small enrollment and a grader? Counts the same.

The easiest path would be to teach the same few courses over and over again. And, especially if you don't have a grader, don't assign very much homework. The course evaluations that our students fill out ask more questions about the professor's classroom presence (voice, handwriting, etc.) than they do about how much the students learn. They ask several questions about the amount of work assigned but none about whether the professor (or even the grader!) provided feedback on graded work. It would seem that the easy path is the way to go. (Here's a sample evaluation form. Ours are fairly similar in their blandness.)

Someone I know (at another school) is working on a plan to completely automate teaching. Every problem in the course is categorized into a certain well defined type, and there is an online database of worked-out examples of problems of each type. Tests are generated by the system by inserting random numbers into certain parameters in standard problems. Students can take practice tests at home and re-try the practice problems until they get them right. Once a week the entire class is required to come to the computer classroom during the regular class time and to do one of these online tests under exam conditions. The system grades all the tests and compiles the scores. During the rest of the time when the class would meet, the professor is available to answer questions. I can think of few circumstances in which this is an appropriate way to run a college class.

Five years from now my students will be teaching math to elementary and middle school children. This is the last math class that many of these future teachers will ever take. I can maximize the time I can spend on my research (and probably my evaluation scores) by restricting their homework assignments to a few machine scored questions about arithmetic and basic algebra (or better yet, tell them homework is optional and the answers are in the back of the book). Or I can find problems that can only be solved by an understanding of place value, collect instructive examples of typical errors made by children, pose engaging and open-ended challenges — and spend the time evaluating my students' responses to these. Do I serve my interests or theirs? Do you trust me to make the right choice?

Two courses worth the same amount of academic credit might, from a student's perspective, require different amounts of work. The same is true when teaching: adding another section of a course you already teach (and have taught before) is going to require less time and effort than adding a totally new course. But they both count the same in terms of calculating the teaching load. Large enrollment and no grader? Small enrollment and a grader? Counts the same.

The easiest path would be to teach the same few courses over and over again. And, especially if you don't have a grader, don't assign very much homework. The course evaluations that our students fill out ask more questions about the professor's classroom presence (voice, handwriting, etc.) than they do about how much the students learn. They ask several questions about the amount of work assigned but none about whether the professor (or even the grader!) provided feedback on graded work. It would seem that the easy path is the way to go. (Here's a sample evaluation form. Ours are fairly similar in their blandness.)

Someone I know (at another school) is working on a plan to completely automate teaching. Every problem in the course is categorized into a certain well defined type, and there is an online database of worked-out examples of problems of each type. Tests are generated by the system by inserting random numbers into certain parameters in standard problems. Students can take practice tests at home and re-try the practice problems until they get them right. Once a week the entire class is required to come to the computer classroom during the regular class time and to do one of these online tests under exam conditions. The system grades all the tests and compiles the scores. During the rest of the time when the class would meet, the professor is available to answer questions. I can think of few circumstances in which this is an appropriate way to run a college class.

Five years from now my students will be teaching math to elementary and middle school children. This is the last math class that many of these future teachers will ever take. I can maximize the time I can spend on my research (and probably my evaluation scores) by restricting their homework assignments to a few machine scored questions about arithmetic and basic algebra (or better yet, tell them homework is optional and the answers are in the back of the book). Or I can find problems that can only be solved by an understanding of place value, collect instructive examples of typical errors made by children, pose engaging and open-ended challenges — and spend the time evaluating my students' responses to these. Do I serve my interests or theirs? Do you trust me to make the right choice?