ABSTRACT Many construction management (CM) courses have low enrollment numbers, making them difficult to study using independent measures research designs. However, repeated measures (RM) research designs, which have historically been underutilized in CM education research, can be used to overcome the deleterious effects of low participation by increasing the number of measurements to achieve sufficient statistical power (i.e. 1-β greater than .80). This paper is the final installment in a research series documenting an RM experiment testing the effectiveness of instructional videos on undergraduate CM students. The emphasis of this paper is on the research method – a within-subject, withdrawal-of-treatment, crossover, RM experimental design – chosen for its ability to increase the statistical power for small sample sizes. The final, actual statistical power of the study was .99. Incongruous with most research testing the pedagogical effectiveness of instructional videos, the study found no statistically significant difference between students who were given instructional videos and the control group. Researchers should consider using RM designs to amplify statistical power in studies with low anticipated participation or to avoid missing clear opportunities to improve the quality and precision of their research findings by increasing the quantity of measurements.
Read full abstract