2.3. Data Analysis

The statistical analysis was done with IBM SPSS Statistics 25.0 (IBM Corp. in Armonk, NY, USA) and MATLAB R2018b (The MathWorks, Inc., Natick, MA, USA). Pearson correlation analysis was used to evaluate associations between exercise measures (i.e., intensity-specific frequency) and other variables (age, cognitive functions, and mental health). The Chi-square test was used to evaluate whether there was an association between PA level and gender and between PA levels and the two frequencies of MVPA. Linear regression was used to evaluate to what extent exercise measures can predict outcome variables, with or without adjusting for covariates such as gender and age. One-way Analysis of Variance (ANOVA) was used to determine whether there were any differences in outcome variables between subjects at the three PA levels (Low, Moderate, and High), with or without gender and age as covariates (conducted as a general linear model). Bonferroni adjustment was used for post hoc comparison. Student’s t-test was used to compare differences between two groups means. The level of statistical significance was set at p < 0.05.