This study examines how students perceive official student evaluations of teaching (SETs) and unofficial mid-semester evaluations (MSEs). It also examines whether completing a MSE affects students' perceptions of the course and the instructor. A survey revealed that participants (N = 80) believed SETs are valid measures of teaching; however, they had doubts about whether students or instructors take these evaluations seriously. On the other hand, participants had very positive perceptions of MSEs and instructors who conduct them. Furthermore, completing a MSE positively affected perceptions of the instructor's responsibility, his commitment to teaching, and his desire for the class to do well. Implications and limitations of these results are discussed.
Student evaluations of teaching (SETs) are widely used in higher education to assess instructors' performance and to provide a means of accountability for the college or university (Williams & Ceci, 1997). Administrators routinely use SETs when making decisions about the hiring, tenure, and promotion of instructors (Newport, 1996). These evaluations are also sometimes made available to students for use in course selection (Marsh 1987). However, there is a debate over the utility of SETs for these purposes.
Critics argue that SETs are prone to bias and that students are incapable of effectively evaluating teaching (e.g. Wachtel, 1998). Because SETs have become such an important factor in employment decisions, instructors may be temped to grade leniently, and as a result, academic standards may suffer. Proponents contend that students are indeed capable of evaluating teaching (Miller, 1988) and that well-developed SETs serve as one of the best measures of teaching effectiveness (Hobson & Talbot, 2001).
Research on the reliability and validity of SETs is mixed. West (1988) found that students were only moderately consistent in their evaluations of instructors over time. Furthermore, students' ratings can be influenced by a variety of factors--including class-size (Min & Baozhi, 1998), time of day (Greenwald & Gilmore, 1997), race and attractiveness of the instructor (Rao, 1995), and expected grade (Marsh & Roche, 2000). On the other hand, a number of studies lend support to the claim that SETs are reliable and valid measures of teaching (e.g. Gamliel & Davidovitz, 2005). Drew, Burroughs, and Nokovish (1987) validated student evaluations by assessing student-instructor agreement about day-to-day variability within courses. Students' reports were strongly related to those of instructors.
Although the reliability and validity of SETs has received much attention, relatively few studies have examined how students perceive these evaluations. Students seem to be generally willing to complete SETs and try to be fair and accurate (Spencer & Schmelkin, 2002). Students also tend to have confidence in their ability to assess instructors. In one survey, the majority of students believed that SETs provide an effective measure of teaching (Dwinell & Higbee, 1993). However, the results of other surveys suggest that students are skeptical about the utility of SETs. Spencer and Schmelkin (2002) found that students do not believe that instructors or administrators take these evaluations seriously.
Even if SETs are valid measures of teaching, they pose a fundamental problem for instructors being evaluated. Because SETs are typically conducted at the end of the semester, instructors do not have the opportunity to make changes to the course for which they are being evaluated. If instructors conducted their own mid-semester evaluations (MSEs), they should be able to better assess student learning and make immediate changes to their curriculum or teaching style. Price and Goldman (1981) found that conducting MSEs led to improved ratings on end-of-semester evaluations. Conducting unofficial MSEs may also convey to students a genuine interest in how well they are learning. …