Sample size needed for student ratings of instruction

Area of Science:

• Educational assessment
• Psychometrics
• Higher education pedagogy

Background:

• Increasing administrative burden from student evaluation forms.
• Need for efficient and reliable methods for student ratings of instruction.
• Current practices may involve excessive data collection.

Purpose of the Study:

• To determine the minimum sample size required for accurate student ratings of instruction.
• To reduce the number of evaluation forms students must complete.
• To provide data for calculating necessary sample sizes across different class sizes.

Main Methods:

• Analysis of questionnaire items using four- and seven-category rating scales.
• Sampling data from class sizes of 40, 60, 100, and 140 students.
• Computation of standard error of the mean and permissible error index for various sample sizes.

Main Results:

• Figures illustrate the relationship between sample size and rating accuracy.
• Identified minimum sample sizes needed for accurate student evaluations.
• Demonstrated that sampling one-third of students in classes of 100-140 is adequate.

Conclusions:

• Efficient student evaluation of instruction is achievable with smaller sample sizes.
• Reduced data collection can be implemented without compromising rating accuracy.
• Provides a practical framework for optimizing student feedback processes.