To the Editor:
Most students at Indiana University Northwest take the mathematics-placement exam years after their last high-school mathematics class, not being aware that their knowledge has faded in the interim. The mathematics faculty at the university, which has a diverse student body, long ago realized that we needed something better than “a single standardized test that students generally don’t prepare for or take seriously” (“National Groups Call for Big Changes in Remedial Education,” The Chronicle, December 13).
The placement-exam structure that we recently created is accurate, simple, and inexpensive. Since the beginning of the full implementation, only 21 percent of students placed into the lowest mathematics level, compared with usual rate of 35 percent. Furthermore, detailed statistics are continuously collected, helping us with academic planning and with controlling the quality of exam questions.
Our mathematics-placement exam is broken into five levels, the topics of each matching our prerequisite courses. On arriving, the student is given five minutes to look at all five exams on paper and select the one that he or she wants to try on the computer. Passing the selected level allows the student to try the next higher level; failing allows the student to try the next lower level. This way we build student confidence, trust, and understanding of the accuracy of the placement. The five exams are programmed in the Indiana University Classroom Management Computer System in such a way that the answer to each question for all 1,300 students annually is saved. The mathematics department has instant access to this valuable data and it is already playing an important role in academic planning.
The placement questions are easy and quick. Some students still need a brief refresher on topics, even though most are not aware of it. Statistical analysis of students who failed to place into the lowest college-level mathematics classes gave us characteristics of our “in risk” group. At admission, such “in risk” students are informed that they have to attend a free two-hour workshop before taking the mathematics-placement exam. The workshops cover 600 students annually and cost us under $10,000, and the “in risk” students participating have shown significant improvement in their passing rate (80 percent vs. 20 percent).
We are fortunate that our university administration trusted us that having a student-centered placement structure is better than getting a single standardized-placement score for each student just for sake of comparability to other institutions. The price was countless hours of work by a group of enthusiastic faculty and staff members from mathematics, admissions, the administration, and the information technology staff—it would have been far easier and faster to make a call to a software company providing standardized tests and to shift the cost of the placement exam to our students. But improvement of the placement-exam structure needs to be recognized as a characteristic of a good university.
Co-Director of Introductory Mathematics
Indiana University Northwest