Lab #7 Preview
Paired Samples t Test
Purpose
Compare the sample means for two sets of scores obtained from the same individuals, or matched pairs of individuals. Is there a significant difference between the means?
3rd Graders at private school measured twice:
Before 3rd Grade After 3rd Grade
Is there an improvement in IQ scores from before to after 3rd grade as predicted by the school principal?
Measures (DVs):
- iqbef = IQ before 3rd grade
- iqaft = IQ after 3rd grade
One-tailed vs. two-tailed test: When there is a prediction about the direction of results, you can use a one-tailed hypothesis test. In SPSS, this involves simply using the One-Sided p value when comparing to your alpha level (.05).
Null Hypothesis: Mean IQ before 3rd grade ≥ Mean IQ after 3rd grade
Alternative Hypothesis: Mean IQ before 3rd < Mean IQ after 3rd grade (matches prediction)
Paired samples t test is a test of the null hypothesis. Based on test results, we either:
- Reject the null hypothesis (and conclude the alternative is correct), or
- Fail to reject the null hypothesis (and conclude the means do not differ in the predicted direction)
APA Style
The APA style write-up for the paired samples t test follows the same format as for the independent samples t test covered in the previous lab… report the two means, their standard deviations, whether the means differ (and which one is larger/smaller than the other) or do not differ, and the t-statement (with the same format of t(df) = xx, p = xx, d = xx).