Stata ic statistical software release 14

We measured the internal consistency of the Spanish version of SCHNOS-obstruction domain (SCHNOS-O) and SCHNOS-cosmetic domain (SCHNOS-C) with Cronbach’s alpha, which was calculated along with a 1-sided 95% CL. Results of Cronbach’s alpha ≥0.9 were considered excellent; 0.9 > alpha ≥ 0.8, good; 0.8 > alpha ≥ 0.7, acceptable; 0.7 > alpha ≥ 0.6, questionable; 0.6 > alpha ≥ 0.5, poor; and <0.5, unacceptable.
We also determined the correlations between the items included in the Spanish version of the SCHNOS-O and the SCHNOS-C scales, with a Spearman correlation coefficient, which was obtained along with a 2-tailed P value. P values ≤ 0.05 were considered significant. All the analyses were carried out using Stata/IC Statistical Software: Release 14 (StataCorp LP, College Station, Tex.).
To measure the reproducibility of the questionnaire, we carried a test–retest in 40 patients, and we calculated a Spearman’s rank correlation and a Wilcoxon signed-rank test for matched pairs (P values) to compare the mean scores obtained for the SCHNOS-O and the SCHNOS-C during the 2-week interval.

In order to assess how well the model matched the observed data, the root mean square error of approximation (RMSEA) was used. First, the model fit was tested assuming there were no covariances between unique factors. After that, the modification indices suggested by the software were used to add covariance between factors (double-headed arrows in Fig 1) one at a time, each time testing the RMSEA closeness to the value of <0.05, or at least <0.08—the threshold for accepting the model fit.^{32 ,33 (link)} Every insertion was considered plausible if it made logical sense and did not violate the assumption that the common and the unique factors are uncorrelated. After achieving the RMSEA value of <0.05, no further covariances were imputed and the goodness of fit was assessed by χ² test. As the sample was relatively small considering the requirements of CFA, in an attempt to reduce dependence on sample size, the choice was the relative (or “normed”) χ² test. Relative χ² is a χ² estimate divided by the degrees of freedom. A relative χ² value <5.0 was considered an indication of a good fit.³⁴ All analyses were conducted using IBM^® SPSS^® Statistics for Windows^®, ver 22.0 (IBM Corp. Released 2013, Armonk, NY, USA); IBM^® SPSS^® Amos™, ver 23.0 (IBM^® Corp. Released 2013, PA, USA); and Stata/IC Statistical Software, release 14 (StataCorp LP, TX, USA).