Information Sheet for USAT Evaluation Reports

• The course evaluation report is generated from the scanned responses on the completed evaluation forms.
• Evaluation of courses whose enrolment is less than 10 is discouraged for two reasons: (a) the results are not considered to be statistically valid and (b) students may be identified through their handwriting.
• When the number of returned evaluations for a course is less than 10 (even though the enrolment might have been greater than 10), the results from that course are not included in the departmental means.
• The Dean, department head, and student copies of the reports show only the results to the university-wide and department-wide evaluation items.
• The instructor copy shows, in addition, the results to the instructor-chosen evaluation items.
• For each individual course/instructor, the report shows, for the university- and department-wide items, the results for this specific course/instructor (the filled-in histogram bar) and, for comparison, the results for the department (the open histogram bar). Departmental comparisons are produced only when there are more than two courses being evaluated in the session being processed.
• The numerical departmental results are calculated by summing the individual evaluation means and dividing by the number of course/section/instructors evaluated in the department - rather than summing all the responses to each item across the department (which would allow courses with a high enrolment to weigh the departmental mean).
• The percentage departmental results are calculated by averaging the individual course percentages - again rather than creating a percentage of the individual responses across the whole department.
• The standard deviation in the results for the individual course is given as a measure of the degree of distribution of the responses around the mean; the higher the standard deviation, the more spread-out are the responses; i.e. students show greater variation of opinion. The lower the standard deviation, the more in agreement are the students (standard deviation has been calculated using a population rather than a sample standard deviation function).
• The "Number of returned evaluations" gives the total number of scanned forms.
• The "Number of respondents" to each evaluation item gives the number of respondents providing a response to each item. Non-responders are not included in this number, so that this value may vary among the items and against the "Number of returned evaluations".