2.4 Statistical Analysis
Accurate power analysis cannot be done since studies of this kind have not been performed before; however, sample size was inferred based on an estimated difference in the mean SNOT-22 scores of 8.9 (clinically significant). Calculations were based on an effect size of 0.45 and a power of 0.8. Stratified groups and postoperative changes within groups were compared using the Wilcoxon two-sample test. P-values were considered significant if less than or equal to 0.05. Analysis was done on selected questions within the SNOT-22 and ETDQ-7 questionnaires that related to ear pain or ETD. All three instruments were analyzed for correlation of change between the preoperative and 6-week postoperative visit. Pearson correlation coefficients were calculated for SNOT-22, NOSE, and ETDQ-7 instruments for all groups.
Results