Proposal Title

An Analysis of Biases in the Teaching Assessment

Proposal Abstract

This presentation reports an analysis of the biases that influence the student teaching evaluation (STE) interpreted from a faculty developer's perspective. Does the amount of student workload influence teacher ratings as in Greenwald & Gilmore, (1997)? Are expectations on the impact of the evaluation influencing student participation as Caulfield (2007) argued? Are students more generous in the ratings as they become more specialized (Atkinson, 2001)? How can the method for delivering instruction bias teacher ratings (Lucal et al., 2003)? Once presenting the context of our institution (EPFL) as well as the teaching and evaluation practice; we shall address the audience in search for leads to achieve a critical analysis from the standpoint of the Scholarship of Teaching and Learning of the biases here presented.

Full Proposal

This presentation reports an analysis of the biases that influence the student teaching evaluation (STE). At the Swiss Federal Institute of Technology, Lausanne(EPFL) we have examined whether the STE instrument is sensitive to variables others than the quality of teaching; such as the difficulty of a course, the size of class, the level of the course, and the response rate. Data were grouped in a regression model. Results show a significant but weak correlation (0.14) with regards to the difficulty of a course (STE vs. exams marks). An in-depth look reveals that first year courses (selection year) tend to receive lower marks (0.13) than the Masters (0.17). In addition, a negative correlation of -0.20 suggests a low impact of student participation on the evaluation, notably in smaller than 50 student classes. A null correlation between the evaluation and the number of teachers per course was also found.

Content:

1. Description of the session: A 15-20 minute presentation consisting of power point slides or overheads, it will be followed by a frank and open discussion. The presentation will focus on the results and will introduce the context of our school and faculty development activities.

2. Objectives of the session: The main objective for the presentation is to disseminate a fresh and complete analysis linking teaching evaluation results with student performance. A second objective is to share lessons learnt with regards to similar experiences in other contexts.

3. Involving the audience: The audience and other panelists will be invited to share any experience on the subject during and after the presentation.

4. Attendees can expect: A novel and first hand material matching teaching evaluation results with student performance (exam marks) in a European and higher education context. Our results will add an international update to the work of Aleamoni (1999).

Aleamoni, L.M., (1999). Student Rating Myths Versus Research Facts from 1924 to 1998. Journal of Personnel Evaluation in Education, 13:2, 153-166. Kluwer Academic Pulishers, Boston.

Further references: http://craft.epfl.ch/

Location

Room 2904 B

Publication Type and Release Option

Presentation (Open Access)

Share

COinS
 
Nov 1st, 11:00 AM Nov 1st, 11:45 AM

An Analysis of Biases in the Teaching Assessment

Room 2904 B

This presentation reports an analysis of the biases that influence the student teaching evaluation (STE) interpreted from a faculty developer's perspective. Does the amount of student workload influence teacher ratings as in Greenwald & Gilmore, (1997)? Are expectations on the impact of the evaluation influencing student participation as Caulfield (2007) argued? Are students more generous in the ratings as they become more specialized (Atkinson, 2001)? How can the method for delivering instruction bias teacher ratings (Lucal et al., 2003)? Once presenting the context of our institution (EPFL) as well as the teaching and evaluation practice; we shall address the audience in search for leads to achieve a critical analysis from the standpoint of the Scholarship of Teaching and Learning of the biases here presented.