2.5. Statistical Analyses

STATISTICA 13 software (StatSoft, Inc., Tulsa, OK, USA) was used for all analyses, except for the calculation of interclass correlation coefficients (ICCs), which was performed using Predictive Analytics Software (PASW) v.18.0 (IBM Corp., Armonk, NY, USA). The basic statistical characteristics (mean, standard deviation, range) were calculated for quantitative variables, and the percentage distribution was estimated to describe the qualitative variables. Gender differences for continuous (anthropometric) variables were calculated using a t-test and the structure indicator was used for nominal (demographic) variables. ICCs were calculated to assess the reliability, and Pearson’s correlation was used to estimate the validity of the PAQE-PL. The interpretation of ICCs was based on the criteria of Koo and Li [21]: below 0.50 is poor, between 0.50 and 0.75 is moderate, between 0.75 and 0.90 is good, and above 0.90 is excellent. After Terwee et al. [22], we assumed acceptable reliability of the physical activity questionnaire if the ICC value was above 0.70. In the interpretation of correlation coefficients values, Cohen’s [23] classification was used. A correlation coefficient of 0.10 indicated a small relationship, and values of 0.30 and 0.50 were considered medium and large correlations, respectively. After Terwee et al. [22], we assumed the physical activity questionnaire was valid if the correlation was above 0.50 for accelerometer and 0.30 for other questionnaires.