This preview shows pages 1–2. Sign up to view the full content.
1
Psychology 321 (SP08) – Lab #3
Content:
A. Overview of Dependent Samples t test
B. Computing Dependent Samples ttest using SPSS
C. EeanyMeenyMinyMo
A. Overview of Dependent Samples t test
This week in class you (hopefully) learned something about the independent
samples ttest.
One of the assumptions of this test is that the two groups are independent
of each other. Contemplating this assumption invariably led you to the
question, while discussing stats over dinner, “Yeah, but what do you do if
you know the samples are related, like before and after tests on the same
people or husband/wife pairs rating their marital happiness??”
The answer you are seeking has been in your book all along (Chp. 13, btw);
the dependent samples ttest!! This is a wily test that has many aliases,
including relatedsamples, matchedsamples and pairedsamples ttest.
Don’t be fooled…it’s all the same thing.
The basic idea behind this test is to get rid of the dependence between the
two scores. That is done by finding a difference scores between the two
measurements (i.e. posttest – pretest). By doing this, you are also able to
remove some individual differences (which is a very good thing). Once you
have the difference scores, you just do a onesample ttest with them.
Just to make you memorize something else, there is a specialized formula:
t = M
diff
– 0
with df = N – 1
where N = # of difference scores
(s
D
/
√
N)
but notice that it isn’t actually different from the onesample ttest you
already know and love
t = M
sample
– M
pop
.
with df = N – 1
where N = # of scores
(s
sample
/ √ N)
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 04/22/2008 for the course SOC 321 taught by Professor Reiter during the Fall '08 term at Ohio State.
 Fall '08
 Reiter

Click to edit the document details