Health Sciences 3801A/B Lecture Notes - Lecture 2: Repeated Measures Design, Jeff Cooper, Mean Absolute Difference

• School: Western University
• Department: Health Sciences
• Course: Health Sciences 3801A/B
• Professor: Christopher Lee

Document Summary

Involves collecting two sets of scores, each set obtained under a different condition. Between groups design: participants assessed under one condition or the other condition. Md (mean difference: replace participant 1, changes score in condition 1, changes m1, so md would change. Repeated measured design: participants assessed under both conditions. There is always some uncertainty about the size of the mean difference. Is a measure of how much the mean difference is likely to vary its calculation depends on the design: (cid:862)a(cid:448)erage(cid:863) a(cid:373)ou(cid:374)t (cid:271)y (cid:449)hi(cid:272)h (cid:373)ea(cid:374) differe(cid:374)(cid:272)e (cid:449)ill flu(cid:272)tuate fro(cid:373) (cid:1005) sa(cid:373)ple to a(cid:374)other. A t ratio is calculated by dividing the mean difference by its standard error: = (cid:3032)(cid:3033)(cid:3033)(cid:3032)(cid:3030) (cid:3042)(cid:3033) (cid:3030)(cid:3042)(cid:3041)(cid:3031)(cid:3042)(cid:3041) If condition has an effect on scores, the absolute value of md should (cid:271)e greater tha(cid:374) e (cid:374)ot just a little greater, but greater to a degree that is unlikely to happen by chance. Indicates the number of standard deviations by which the groups differ.