Reply to 'Right statistical methods should be driven to get the right answers from research studies

摘要

We thank the commentators for their criticism and their additional information and interpretation, but some misinterpretation by the commentators have to be commented on. Their main criticism is regarding the statistical methods used in our article, especially concerning statistics in Material and Methods. Correlation of pneumatization and courses of the structures, we did of course perform the chi-square test with Cramer's V-test (because of frequencies <5%) and bivariate test. We regret this possible misunderstanding term, but results and significances of the paper are completely correct. To repeat the results, the correlations between pneumatization and courses of the internal carotid artery and optical nerve as well as the correlation between the artery and optical nerve itself were all highly significant (P < 0.001; Cramer's V between 0.199 and 0.370; Person chi-square between 16.253 and 156.839). In agreement with the reviewers we did not include this information in the paper because focus of this study was not to demonstrate statistical correlations. The main aim was to demonstrate that cone beam computed tomography (CBCT) can visualize the sphenoid sinus as good as conventional CT or anatomical studies and to demonstrate frequencies of anatomical variants.