Bonferroni correction - correctes for overall type I error, very strict, increases type II chance, corrected p value, Family-wise error correction - uses gaussian random field theory in SPM, probability of making 1 or more false positives errors when performing multiple tests, False discovery rate - controls expected proportion of false positives among all supra-threshold voxels (significant results), Permutation/resampling method - irregularities in data are kept in permuted data sets and included in estimation of permutation probability with no assumption that p-value distribution is correct, ROI comparisons - change basis fo statistical tests - each ROI can be thought of as a very large volume element, anatomically based ROIs do not produce bias,

Statistical corrections

โดย
สั่งพิมพ์
ฝัง

ลีดเดอร์บอร์ด

สไตล์ภาพ

ตัวเลือก

สลับแม่แบบ

คืนค่าการบันทึกอัตโนมัติ: ใช่ไหม