Every term, students get a short reprieve from their lectures, and professors quietly slip out of the lecture rooms. It is during this brief moment that students experience some small sense of deterministic power over their departments. However, right from the computer[[Eth]]scanned forms, there are serious problems which prevent student evaluations of profesors from being effective.

The computer[[Eth]]scanned forms are a central problems. Although some departments have substituted their own customized forms, the majority still use standard, cookie[[Eth]]cutter ones. Students regularly receive forms with many questions dedicated to the intracacies of the lab segment of a class that has no lab, or they are asked to rate the effectiveness of a discussion section or a TA when the class has neither. When they do receive a questionairre in their lab sections, students are asked about the textbooks they do not use or the exams they are not given. Furthermore, while these forms typically have room for comments, they do not offer any questions to focus these comments on. Many students, although far from being stupid, are simply too hurried or disinterested to write at length when they do not know what to write about.

Other students do something dramatically worse. Acting on spite, they fill a row of "Strongly Disagree" responses on the form without even reading the questions. Ostensibly, they do this just to make a statement or lower the averaged responses to the individual questions. However, this does not help the professor Ñ most undoubtedly discard or devalue such vehemently negative responsesÊÑÊand it does not help the students, since they waste their opinions. Most of these students do not write in comments, except perhaps for brief vulgarities, because they do not take the process seriously. In fact, most students do not write in comments. They are too busy (they can leave earlier if they do not write comments), they are too unconcerned, and they are convinced that their opinion is not valued.

There are also problems with the way that the information on the forms is used and interpreted. Recently, there has been discussion over the possibility that the student evaluations somehow have a bias against female professors. Various arguments have been levied in favor of this hypothesis. First, statistics showing lower ratings for female professors have been paired with the premise that students themselves are biased against female professors. For this reason, some have called for de[[Eth]]emphasizing student evaluations in the process of considering professors for promotions. Second, claims have been made that female professors are unappealing to students Ñ either because they do not "act" like their male counterparts, or because they do, particularly by acting more sternly than they would otherwise, thereby seeming unfriendly to students. Proponents of these arguments call for statistical modification of the results of the student evaluations, based on the gender of the professor.

Several things must change. Although some level of central control must ensure that evaluation forms are given for all courses at the University, departments should be made to create forms which ask only for evaluation on topics relevant to the courses in question, and which ask questions which are more specific to those courses. Everything possible should be done to avoid excessively generic forms.

Statistical modification should not be applied to the results of the forms with the overt intention of improving the image of female professors or any other group. Such actions are inherently discriminatory, vile, and unacceptable. However, while any given question should be applied equally to all professors teaching a class, some clearly bear more importance than others. Most students would prefer stern professors who teach well and give good assignments and exams, and then grade them fairly, to nice professors who teach poorly and grade arbitrarily. Issues, such as pleasantness, which are peripheral to the quality of education, in comparison with knowledge of a subject, should be weighed less heavily. However, this should be done to all professors in a department, and if any group rises or falls as a result of this, then this must not be discarded as "bias."

Finally, the results of these evaluations should not be used to judge professors numerically against each other. Publications such as Advice which grade professors by the responses to the evaluations only worsen the situation, by creating an atmosphere in which professors compete for praise from statistics. The tradition of students verbally communicating with each other, relating pleasant and horrific anecdotes, and helping each other decide which professors teach classes best is a good one. It relies on anecodotal evidence which has the explicit goal of helping students pick the best professors, rather than on statistical evidence which is used for questionable purposes.

However, before any modification to this system is proposed, it must be clear how important these evaluations are. While most professors do seem to care for the opinions of their students, they are often biased by their opinions of how hard the particular students work. Furthermore, students are often afraid to voice their opinions for fear that criticisms of a professor's performance will be weighed in the grading process or will otherwise harm them. By giving students a confidential means of criticising (or, believe it or not, complementing) their professor's work, the student evaluations provide what is often the only means students will have to effect the quality of the lectures, homework, and exams that make up a large portion of their education. For these same reasons, the evaluations provide professors with their only chance of obtaining an unbiased and honest evaluation by those their teaching affects the most Ñ their students. Whatever changes are made, evaluations should continue and they should continue to be taken at least as seriously as they are now.