"My philosophy of education is that I will try anything once, and if it works, I'll keep on doing it until it doesn't work any more." Sheila L. Skemp, a professor of history at the University of Mississippi, used those words in a paper Thursday to reflect a reality of the way many professors talk about teaching when they gather at scholarly meetings: They trade stories about what has worked -- and what hasn't -- in their classrooms.
This week, members of the American Historical Association are gathering in Philadelphia and members of the American Economic Association are meeting in Boston. Following are some of the ideas being shared about teaching at the college level -- many of which would apply beyond those two disciplines.
The Syllabus as a Tool, Not a List
When he started teaching, said Todd Estes, he was a minimalist about his syllabus, seeing it as a place to list assignments, schedules, etc. But Estes, a professor of history at Oakland University, in Michigan, said that his thinking -- and his syllabuses -- have evolved. Currently, he has an 11-page syllabus for his introductory American history course.
In a paper for a panel at the AHA meeting, Estes said that his primary goal for his students is to have them act and think like historians, not like students in a history class. In a class of 55 -- many of them "skeptical or even hostile to the notion that history has value" -- how can you do this? Estes argued that one way is through the syllabus, which isn't just a list, but provides context about the course, so that students are confronted with ideas, not just information, every time they look at the document.
Many students are uncomfortable with the idea that there is not a "single definitive answer" for some historical question. So Estes drives home a point about history, using the syllabus. A section of the syllabus is called "Why Historians Argue All the Time -- And Why YOU Will Be, Too, This Semester." In that section, Estes said, students are told why this is the case, and also that they will be required in papers to identify conflicting arguments in historians' writings.
Each book that is read isn't just listed on the syllabus, but is annotated -- again with the idea of showing students that there is a broader context to history and to the work of historians, not just a list of correct answers and assignments.
The syllabus also contains the usual information about assignments, work expectations, and plagiarism. Together with the context, Estes said, it tells students not only what they will do, but the kind of thinking they will be asked to try.
Cost-Benefit Analysis on Grading Homework
A paper  presented at the economics meeting noted that scholars in the field devote a lot of time to analyzing how people make various decisions, but apply little of that work to pedagogical issues. Problem sets are a common tool in economics (and other fields). Some professors grade them and include them as a portion of a final grade. Others do not. A logical question is whether students make a different cost-benefit analysis on how much effort to expend on problem sets if they are graded. Another question is whether any improved performance is significant enough to alter the choices of professors who don't grade.
At Syracuse University, a large economics course ended up being divided into two groups: one with graded problem sets and one without. Wayne A. Grove, an associate professor of economics at LeMoyne College, and Tim Wasserman, assistant director of institutional research at Syracuse, thought they had an opportunity and they analyzed the two groups.
The results: Grading matters. When accounting for other variables, the average grade was B- for those with graded problem sets and C+ for those without. The grading incentive, Grove and Wasserman wrote, appeared to primarily produce increased productivity in studying, not more time studying. While urging that more research be done, Grove and Wasserman wrote that their findings might suggest value in encouraging those teaching freshmen -- many of whom typically have difficulty and drop out -- to grade problem sets.
Using 'Question Architecture' to Avoid Awkward Silences
The differences in students' level of participation from class to class "can be maddening," wrote Henry Kamerling, an associate professor of history at Queens University of Charlotte. But what to do? How can professors create momentum in class when the question that is part of a "carefully-crafted, well prepared lesson plan" prompts only "dead silence."
Kamerling outlined a series of strategies based on what pedagogy experts call "question architecture." The approach is designed to engage students at all academic performance levels, to break out of the dry lecture format, and to encourage students to gradually grapple with more difficult questions. A professor might start with a series of questions that are primarily factual, then involve the class, and then pose more difficult questions.
For example, a session on the American Revolution might start by asking students to name all of the causes they can think of, while having a student jot them all down on a board in front of the class. Because there are tons of "right answers," lower performing students can get involved. Asking one or more students to write the answers on a board in front of the class sends students the message that they are participating in the teaching, and creates a "controlled chaos" in which students may be more comfortable offering ideas.
Then, Kamerling said, the professor would shift to higher level kinds of questions -- grouping together the various answers into categories and finally posing larger questions of what the revolution was about.
Few issues frustrate legislators more than remedial education at colleges. Many lawmakers complain about having "to pay twice" to educate students, and many suggest that there isn't proof that remedial education even works. Comparisons of the skills of students who did and did not take remedial courses have limited validity as they are not actual comparisons of the impact of remedial education.
But a paper  published by the National Bureau of Economic Research and presented at the economics meeting found evidence that remedial programs do work on a number of levels. The study examined 28,000 students at Ohio public colleges over a five-year period, and accounted for different levels of student preparation. The pool was large enough that the researchers were able to examine comparable students who received remediation and who did not.
The analysis found that students who need remedial education and get it are more likely than other underprepared students to complete a degree and less likely to drop out. These students are also more likely to transfer to a more competitive institution.
The study was conducted by P. Eric Bettinger, associate professor of economics at Case Western Reserve University, and Bridget T. Long, associate professor of education and economics at Harvard University.
Simplicity and Snapping in Large Lecture Courses
While many presentations focused on the latest ideas, some dealt with formats that are quite old, such as the large, survey lecture course.
In her paper for the history conference, Skemp of the University of Mississippi said that there are no magic solutions that work for everyone, but she said that large lectures could work -- even without lots of "bells and whistles" provided by the latest educational gizmos. She said professors must start with the basics, insisting on rooms with decent acoustics, a proper podium, etc. If you lack these basics, she suggested, the brilliance of lesson plans may be moot.
In terms of the substance of lectures, Skemp said "the larger the class, the more general you must become." She suggested that in survey courses, professors are advised to come up with a few key points per lecture, to stress those in various ways, and not to try to cover much more ground. "Don't deviate. Don't go off on tangents," Skemp recommended.
Skemp said a similar approach also helps with the course itself -- establish some basic themes and stick to them. While Skemp is dubious of gimmicks, she advocated one approach that she initially wrote off: telling students that if they don't understand something she says, they should snap their fingers.
"I was convinced either that they wouldn't do it at all -- or that they would abuse their privileges and do it all the time," she said. "But in fact, it works amazingly well. Very seldom is there just a click or two. Most of the time, what one person doesn't understand, at least half the class doesn't get it, as well. You will suddenly hear a 'loud snapping sound' cascade through the room."
And that's a good thing, Skemp said. "You will stop, explain yourself to their satisfaction, and move on."