This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
In Reply.—We appreciate Dr Rosenfeld's comments that a nonparametric analysis of variance (ANOVA) would be more appropriate to examine the data in our article.1 In our study, the pairwise t test was used to determine the statistical significance of bacterial adherence between children and adults, and between patients with otitis media with effusion (OME) and normal control subjects, although the data distribution is skewed, and in some groups, the sample sizes are small. Rosenfeld suggested that the data should be normally distributed within groups, and that 30 or more samples are needed per group to apply any parametric test. In this letter, we consider Rosenfeld's comments and reexamine our results of bacterial adherence.
Increasing the number of samples, we examined whether the adherence of Haemophilus influenzae and Streptococcus pneumoniae is normally distributed or not. Nasopharyngeal epithelial cells were collected at random from 26 children, consisting of 11 patients
SHIMAMURA K, SHIGEMI H, KURONO Y, MOGI G. To 't' or Not to 't'-Reply. Arch Otolaryngol Head Neck Surg. 1992;118(4):446–447. doi:10.1001/archotol.1992.01880040113021
Customize your JAMA Network experience by selecting one or more topics from the list below.