S10-AS4-2 - Exploring Correlation among Different Elements of Student Evaluation of Teaching

2. Research-to-Practice Full Paper
Wen Cheng1 , Rani Vyas1, Ranjithsudarshan Gopalakrishnan1, Edward Clay1, Mankirat Singh1
1 California State Polytechnic University, Pomona

The full paper intends to explore the possible correlations among the distinct elements of the student evaluations of teaching. Albeit often controversial due to many demonstrable and potential flaws associated with student evaluations, they still either are one part of a larger set of evidence or make up the bulk of that evidence used to assess teaching effectiveness. It is especially a common case at teaching colleges where student-centered instruction is of great importance. The usual argument is that students ought to have a voice  on their learning experience even though they may not be the experts to evaluate every single perspective of the course, say, the pros and cons of the chosen course textbook(s) as compared against the set of alternatives. In addition, some of previous research indicates student evaluations may be imperfect measures, but they are also our best measures and may still provide the most accurate information available on teaching effectiveness.

As a result, the authors are dedicated to the analysis of the student ratings about teaching effectiveness and examine the possible association of the various instructional aspects. The data used include a selected group of teaching evaluations of more than one hundred course sections offered at Civil Engineering Department in California State Polytechnic University, Pomona. The teaching performance evaluations cover fourteen different aspects which include course structure and organization, course objective-defining and course-meeting, knowledge demonstration, concept-explaining, communication effectiveness, class discussion-stimulating, student question-answering, consultation availability, student learning facilitation by assignments,  concept-reflecting by exams, timely and constructive feedback on student work, and overall rating of the instructor. The rating scale with five response categories are employed which range from very good (1) to very poor (5).

To capture the reliable correlation amongst different teaching evaluation elements, some of the more popular rank correlation statistics include Spearman's ρ, Kendall's τ, Goodman and Kruskal's γ, and Somers' D were calculated which indicate the performance association from different perspectives. Some of the teaching activities are found to have stronger connection than do others, which tend to have isolated impacts on student learning. It is anticipated that the research findings would assist the faculty in doing the reflective work to diagnose what went awry and course-correct for the future teaching.

Keywords: Student evaluations of teaching effectiveness, rank correlation statistics, teaching diagnosis.