Students in nearly 100 classes this semester will fill out an experimental version of the Student Evaluation of Teaching along with the current version in an effort to develop an evaluation that is more clear and accurate, according to an independent committee of professors charged with investigating the system.
"[These students] will be looking at ... a shorter form, with shorter questions that we hope students will take seriously," said Lyn Stallings, an associate professor in the Department of Mathematics and Statistics, who heads the independent ad-hoc committee.
"I don't think most students take it seriously, and as a result the ratings are probably flawed," said Patrick Aberg, a junior in the School of Public Affairs.
Committees within each school use the evaluations to determine faculty tenure. The new form states in bold at the top, "Your responses to the Student Evaluation of Teaching (SET) form are very important to your instructor and to the University. This information contributes to: 1. improvements in the quality of instruction at American University 2. the annual decisions made on faculty retention, promotion, and salary."
"Faculty perceive that students understand very clearly what the forms are for," Stallings said. "The research seems to indicate that students do not have a clear idea of how the forms are used."
The experimental form asks more direct questions, uses more consistent rating scales and makes the purpose of the evaluation more clear than the current evaluation forms, Stallings said. It also focuses on evaluating teaching behavior instead of professor personality.
There are professors who do well on performance ratings because they have personality and "sparkle plenty" in class, Stallings said.
"Not all teachers are like that, but they're still good teachers," she said.
The new form substitutes questions like "Was the instructor stimulating?" with questions that ask students to rate a professor's ability to use class time productively and give useful feedback.
Stallings added that questions like "Is the instructor knowledgeable?" are difficult to judge.
"How would a student know how to answer that?" Stallings said.
"I think the questions are fine but the rating system ... tends to make students answer down the middle," said Erica Lyn Benjamin, a senior in the College of Arts and Sciences.
Also, the rating scale on the current evaluations is inconsistent, the committee found, according to Stalling. On some of the questions, the middle rating is "good," and on other questions it is "neutral." Some questions ask for ratings on a five-level scale and some on a six-level scale, she said.
"[Students] complete the forms in faith that they are consistent, but in fact they are not," Stallings said.
More than 50 tenured professors volunteered to give their students the experimental evaluation form, along with a feedback form and the current Student Evaluation of Teaching. Professors participating in the test teach in a variety of subject areas on the undergraduate and graduate levels.
There are two versions of the experimental form. Half the students in each class will fill out an experimental form that includes the recommendation question, "On a scale of one to seven, what recommendation would you give to other students about this course?" The other half will fill out a version that does not have the question.
The experimental version will be tested in spring semester classes as well. The committee will analyze the results during the summer and recommend what form the Faculty Senate should adopt. The current evaluation has been used for 15 to 20 years, according to Stallings.
Stallings decided to coordinate the effort when she realized that professors in her subject area were not doing well on performance-based evaluations.
"I had a particular interest ... I had a lot of concern because I work with faculty that have problems with their teaching," Stallings said. "I'm voicing concerns that some of the faculty had."
The new form is part of a larger effort to enhance the way professors are evaluated. Stallings said that now most of a teacher's evaluation is based on these Student Evaluation of Teaching forms.
"We want to put in place a more well-defined teaching file," she said. In May, her committee plans to recommend to the Faculty Senate that teaching files include syllabi, evidence of student learning and other materials as appropriate for the subject area in which a professor teaches.
"If the students like this form and the faculty like this form, then what we have is a much shorter form that gets right to teaching behavior," Stallings said.
