THE TEACHING ECONOMIST - William A. McEachern 
| Home | About The Teaching Economist | Contact the Editor | Support |
Issue 30, Spring 2006
William A. McEachern, Editor
Homework Rules
Another memorable teacher during my junior year of high school was Miss Saunders, who taught Algebra II. She was nearly seventy. Homework was due every class. She would mark wrong answers and return the papers to redo until all were correct. Some were submitted again and again. Erasures sometimes would wear the paper through. I would guess that I had an average "float" of a half dozen assignments going back and forth. Multiply that by 30 students in the class and that's 180 papers in various states of mathematical correctness. Miss Saunders must have been a busy lady.
Research suggests that homework in high school is more valuable than homework in lower grades, and homework in math is the most valuable of all. Repetition is the mother of mastery.
Do we know much about the value of homework assignments in college economics courses? Wayne Grove decided that he would require students in his two sections of economic principles at Syracuse University to complete five problem sets during the term. Just before the term began, the department head asked him to cover two additional sections for a colleague injured in a bicycle accident. Since he expected to fill in only briefly, Grove felt he could not very well impose the graded assignments on students in those other sections. But figuring it could do no harm, he encouraged his accidental students to complete the assignments and he would later provide answers. Otherwise, he treated all four sections identically.
After several weeks, he was informed that his colleague would not be back that term. At that point Grove realized he had a natural experiment underway. Would the students required to do the homework assignments do better on the exams than the other students? The short answer is yes. He found that freshmen benefited the most—other things constant—improving on average from C+ in the sections without the graded homework to B- or better in the graded sections. These results were reported in "Incentives and Student Learning: A Natural Experiment with Economics Problem Sets," by Wayne Grove and Tim Wasserman, presented at the American Economic Association meeting in January (Grove's coauthor is with the teaching and learning center at Syracuse).
In research along similar lines, Susan Pozo of Western Michigan University and Charles Stull of Kalamazoo College report the effect of math homework on principles of economics test scores. A group of 157 students was required to complete math assignments and a group of 116 students was encouraged but not required to complete the assignments. Care was taken to otherwise treat the students comparably. Among students merely encouraged to complete the assignments, few did. Students required to complete the assignments did significantly better on the mid-term exam than did the other group. The benefit was more pronounced among students in the bottom half of the distribution. Oddly, there was no difference on the final exam. These results were also presented at the January AEA meeting in a paper entitled "Requiring a Mathematics Unit in Principles of Economics: Results from a Controlled Experiment."