update

util/evaluation.py

@@ -100,7 +100,7 @@ def statistical_tests(data):
         pair_name_score = f'{var1}{score_suffix} vs {var2}{score_suffix}'
 
         # Bootstrapped T-test for independent samples
-        t_stat, t_p = bootstrap_t_test(data[f'{var1}{score_suffix}'], data[f'{var2}{score_suffix}'])
+        t_stat, t_p = bootstrap_t_test(data[f'{var1}{rank_suffix}'], data[f'{var2}{rank_suffix}'])
         pairwise_results['T-Test'][pair_name_score] = {"Statistic": t_stat, "p-value": t_p}
 
     # Friedman test