Statistical package for the social science version 20

Statistical Package for the Social Science version 20 (IBM, Corp., Armonk, NY, USA) was used for the analysis of data. For quantitative data mean, standard deviation, minimum and maximum values; while frequency values were tabulated for qualitative data. For the comparison of categorical data, Pearson chi-square, and Fisher’s Exact tests were used. Data were evaluated at a level of 95% confidence level, A p-value<0.05 was considered significant. Parameters found to be statistically significant in the univariate analyses that were evaluated in a multivariate logistic regression to predict the risk of mortality.

The IBM Statistical Package for the Social Science Version 20 (IBM Corp., Armonk, NY, USA) program was used for the statistical analyses. The compliance of the variables with the normal distribution was evaluated with Skewness, Kurtosis and Histogram Analysis, the Kolmogorov-Smirnov Test and the variation coefficient. The data were not consistent with a normal distribution. Descriptive analyses were presented as mean±standard deviation and minimum and maximum for numerical data and by using frequency tables for non-numerical data. Data with numerical variables were compared with the Kruskal-Wallis test. Post-hoc analyses for pairwise comparisons were conducted by using the Mann-Whitney U test with Bonferroni correction. The total type-1 error level was accepted as 5% and the p-value as smaller than 0.05 for statistical significance.

The statistical analysis was conducted using Statistical Package for the Social Science version 20 (IBM Corp., Armonk, NY, USA). Friedman test was used to compare the mean differences between the ABRs recorded from two stimuli (click and LS chirp) and three stimulus polarities (rarefaction, condensation, and alternating) for the following variables, namely 1) ABR amplitudes wave I, III, and V; 2) absolute latencies of wave I, III, and V; 3) IPLs I-III, III-V, and I-V; 4) number of averages reaching the residual noise of 0.04 µV; and 5) the SNRs. A post-hoc analysis using Wilcoxon signed-rank test was then performed with multiple comparisons adjustments. The initial significance value of 0.05 was further divided by the six level of probabilities (3 polarities × 2 stimuli) resulting in the adjustment of the significance level to p=0.008. A non-parametric test was used since the data breached the parametric assumptions including the non-normal data distribution and in-homogeneous variance.