Spss statistics package 22

Statistical analyses were performed with SPSS Statistics Package 22.0 (IBM Corp., Armonk, NY, USA). Distributions of several variables were skewed, thus, we performed both parametric and non-parametric tests. Because the results did not differ between approaches, we present only parametric statistics. To examine the effects of a missed nap on subsequent night sleep, paired t-tests (Baseline Night versus Recovery Night) of visually scored sleep measures (i.e., sleep onset latency, sleep duration, sleep stages) and quantitative EEG data (i.e., SWA, SWE) were performed (alpha=0.05; 2-tailed). Paired t-tests (alpha=0.05; 2-tailed) were also used to assess differences (Baseline Nap+Baseline Night versus Recovery) in sleep measures, SWA, and SWE. Summary measures are presented as means (M) and standard deviations (SD). Effect size (Cohen's d) was computed for each comparison [d=(MBaseline Night–MRecovery Night)/SD pooled or d=(M Baseline Nap and Night–M Recovery Night)/SD pooled]. Effect sizes of d=0.20, d=0.50, and d≥0.75 are considered small, medium, and large, respectively (Cohen, 1988 ).

We used the SPSS 22 Statistics Package. Descriptive analyses were carried out (based on the level of measurement of the variables, measures of central tendency, dispersion of continuous variables, and frequency and percentage of categorical variables), along with internal consistency analyses (alpha de Cronbach). Means comparison analyses were performed (t tests and analysis of variances) as well as Pearson's correlations. Afterwards, taking into account the exploratory purposes of our study, a multiple regression analysis was performed with the total score of GAD as the dependent variable and the different items from the resilience scale as predictive variables.
Finally, a binary logistic regression was performed to analyse the predictive influence of resilience skills on a possible GAD diagnosis according to the most critical cut‐off point, according to the theory (scores ≥10) (yes/no). The level of significance was set at P ≤ .05. In every case, the inclusion criteria for the variables in the multivariate analysis were of P (Wald) <.20 in the univariate regressions.³¹ Potentially related factors at P < .20 were included in the multivariate analysis. Then, BSTEP (LR) (forward stepwise logistic regression with the Likelihood ratio test) was combined, and the Hosmer and Lemeshow'³¹ good fit test was applied.