LATIN SQUARE DESIGN: A METHODOLOGICAL INNOVATION IN EDUCATIONAL RESEARCH FOR IMPROVED TEACHING AND LEARNING

Abstract

A Latin square design (LSD) is a specific arrangement of integers, letters, or symbols, where each element appears exactly once in each row and column. Latin squares have long fascinated mathematicians, with a rich history and various applications in mathematics and other fields. It is more efficient and hence more powerful than reasonable alternatives such as Completely Randomized Design (CRD) or Randomized Complete Block Design (RCBD). Educational researchers also realized that Latin squares could be used in their work, but Latin-square designs do not appear to have been adopted in education anywhere near as often as in medical research. The review of how educational researchers have made use of Latin-square designs in their experiments. This work focuses on the performance of students of six classes in six different subjects by six different teachers using six different methods of teaching. A 6 × 6 Latin square design was employed. The descriptive statistics tables shows that SS3 students perform better with minimum average of failure (21.33%), all classes perform better in Ahmadu’s subjects that is Biology and Basic Science with minimum average failure (22.67%) and Method C (i.e. Deductive method) yield better result with minimum average failure (22.17%). From the Analysis of Variance, the p-value obtained for the three variables; Classes, Teachers and Methods of teaching given as 0.556, 0.612 and 0.842 respectively indicates that the null hypotheses are accepted and conclude that the differences between classes, teachers and method of teachings are not significant.