Multi-Matrix Sampling: An Approach to Evaluation of Health Education Programs

Abstract

Sampling is often used in evaluation as an economical and efficient means of estimating population parameters. Among the approaches to sampling traditionally used, few provide the apparent flexibility of multi-matrix sampling. The current paper proposes multi-matrix sampling as an alternative to traditional sampling approaches for evaluation of health education programs. A comparison is made between examinee sampling and multi-matrix sampling. No significant differences were found between the estimates of mean test performance provided by the two sampling plans following completion of the fifth grade unit of the School Health Curriculum Project. The advantages and disadvantages of multi-matrix sampling are examined, and potential applications of this technique are illustrated.