MichaelD. Johnson, Arunachalam Narayanan and William J.Sawaya

Abstract：Purpose (Hypothesis) This study seeks to understand if and how course and instructor characteristics affect SETs and thereby to improve the usefulness of these metrics. This article aims to statistically examine the relationship between course and instructor characteristics and SETs. Design/Method SETs from a large engineering college at a major public university were evaluated over a seven-semester period that covered 3938 courses taught by 549 unique engineering instructors. Course and instructor demographic data were statistically evaluated for their effects on SETs. Results Course characteristics such as class size, course level, and whether a course was an elective or required had statistically significant effects on SETs. Instructor characteristics of gender and academic rank affected SETs and average course grades, respectively. Average course grades were positively correlated with SETs. Conclusions Data analysis showed that course characteristics, faculty demographics, and average course grades had statistically significant effects on SETs; however, in some cases the effect sizes of these variables were small. Administrators and senior faculty members should be cognizant of these relevant factors and their effects when assigning faculty to certain courses and evaluating their teaching effectiveness using SETs.

Key words：Student evaluations of teaching (SETs); academic administration; engineering

总  结：给出了集的推广，使用权的重要性，并保留的决定，这是很重要的将分数在可能不直接相关的因素，教学评估有效性。考虑到这些因素，更重要的是兴趣集以外的学术成长。这项研究表明，一些课程，教师，以及相关的因素在工程大型公立大学设置分数显著的影响。而所提出的数据的样本量是比较大的，这些结果代表只有一个校园。其他高等院校的项目应进行自己的分析之前综合上述结果。有一个显着的负相关学科之间的尺寸和骨料组分。还有一个显着的正相关关系，总课程的设置。学院分配到较低级别的课程，往往有很大一部分可能是不利的因为他们有可能获得较低的设置，其他的事情都是平等的。管理者应以课程的特点，考虑在这些类型的教师绩效评价课程。女教师的教低年级的课程也有可能收到的略低设置分数，而在上层，有没有统计上的显着差异。在这研究，女性教师过多的低级别的课程教学（基于在课程总数教）。虽然效果大小是小的，那些与行政课程作业的责任应该认识到这些结果和分配教师因此，或至少考虑到这一点时评价集的结果。另一个显着的差异设置分数之间产生的核心和选修课程。教师教学选修课程得到平均得分大于0.12集核心课程教学。再次，即使在小规模效应的光，这些差异注意的。教师一直教选修课程可以充气装置相比与那些连续分配的核心课程。