BESA Statistics 2.0 greatly enhances the capabilities of the previous version with two new analysis workflows for
Further new features include
As in BESA Statistics 1.0, it provides optimized, user-guided workflows for cross-subject analysis of EEG / MEG data. The statistical method used is parameter-free permutation testing on the basis of Student’s t-tests (Maris, E. and Oostenveld, R., 2007), F-tests (for ANOVA/ANCOVA), and correlations. The program is maximally user-friendly. All analyses are computed automatically with user-interaction minimized to defining time and / or frequency ranges of interest. Statistical values computed in BESA Statistics 2.0 can be directly used for scientific reports. No further analysis in other programs is needed. All results are visualized and can be directly used for publications.
BESA Statistics 2.0 integrates optimally with data that were analyzed in BESA Research, but it can also process data from other software packages as long as they conform to the BESA Statistics file format. The BrainVision Analyzer 2 native data format is supported for time and time-frequency data.
BESA Statistics will automatically identify clusters in time, and if applicable frequency and space where data of the input groups / conditions are not interchangeable, i.e. where the null-hypothesis that all groups / conditions are equal must be rejected. Results are considered corrected for multiple comparisons as only those clusters will be identified that have higher cluster values than 95% of all clusters derived by random permutation of data. Thus, results obtained by BESA Statistics are objective and robust.
For ANOVA/ANCOVA analysis, an additional non-parametric post-hoc Scheffe’s test is computed to determine, which pairwise comparison(s) were responsible for the group / condition main effect. A Bonferroni-Holm correction for multiple comparisons of the different pairwise combinations is applied subsequently.