Studies in the Design and Analysis of Dental Experiments

IV. A Modified Analysis of Variance in a Clinical Toothbrushing Experiment

¹ School of Public Health and Administrative Medicine of the Faculty of Medicine, Columbia University, New York

Analysis of the data from a clinical toothbrushing study comparing an electric-powered and a standard toothbrush used by normal dental students during an 8-week period has been presented. No statistically significant differences in effects of the brushes as measured by the P-M-A Index were found. However, the P-M-A Index was higher for the lower than for the upper jaw.

The basic method of analysis was the analysis of variance. The analysis was greatly simplified by working directly with differences between the brushes or between the jaws. The sums of squares in the analysis of variance measuring brush differences were appropriately divided to measure differences in the early part of the study period, as well as the constancy of the differences throughout the study period. Various equivalent methods of making this subdivision were presented.