PSYC 3000 Lecture Notes - Lecture 13: Standard Deviation
2 Aug 2016
School
Department
Course
Professor
Document Summary
When you see same treatment effect but different end result. You can get rid of individual differences by looking at the difference between scores for each participant. Individual differences are stable across a design, it is not affected by treatment. The numerator between independent and matched formula are the same. The use of standard deviation instead of s pooled creates a larger number. Use of 1/n instead of 2/n also makes it bigger. But the effect is much smaller than the change mentioned above. Problem with the independent samples formula is that it does not take into account individual differences in which the treatment has no effect on. It also assumes that both groups are independent from each other. The value of r can be found by using spss. If there is no correlation between both samples you should not do a matched samples formula. The formula is only useful if there is a correlation between both groups.
