Dependent samples t-test 16

# Dependent samples t-test 16 - .And Yet More About...

This preview shows pages 1–11. Sign up to view the full content.

..And Yet More About Hypothesis Testing One-tailed and Two-tailed tests

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Put all of the 5% probability in ONE tail Known Dist of Means Fail to reject Reject Sample is higher than most (95% or more) of the other samples One-tailed test
One-tailed Reject Null Fail to Reject Ho 5% Example: df = 19, .05 level t cut-off = +1.729 t cut off = +1.729 t calculated > +1.729 t calculated <= +1.729 Null Sample is higher than most (95% or more) of the other samples

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
One-tailed Reject Null Fail to Reject Null 5% Example: df = 19, .05 level t cut-off = -1.729 t calculated >= - 1.729 t calculated < -1.729 t cut-off = -1.729 Sample is lower than most (95% or more) of the other samples
One-tailed hypotheses Our sample mean will be higher than the known population mean. cut-off score is positive Example: People who play Braingames will have higher memory scores than people who did not play Braingames. Our sample mean will be lower than the known population mean. cut-off score is negative Example: People who take the diet pill will weigh less than people who did not the diet pill.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Split the 5% probability between TWO tails Known Dist of Means Fail to reject Reject Sample is higher than most (97.5% or more) of the other samples Two-tailed test Sample is lower than most (97.5% or more) of the other samples Reject
Two-tailed hypothesis Our sample mean will be higher or lower than the known population mean. cut-off scores are positive and negative Example: People who view negative political ads will be more likely to vote for the candidate or will be less likely to vote for the candidate than people who did not view the ads. Whether MORE or LESS, our research hypothesis will be supported . But can still allow only a 5% probability that we had an extreme sample (extremely low OR extremely high). So we split the 5% between the two tails.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Two-tailed Reject Ho Fail to Reject Ho .025 Reject Null .025 based on df = 19 df, alpha= .05 t cut off = +/- 2.093 t cut-off = +2.093 T cut-off = -2.093 t calculated > +2.093 t calculated < -2.093 Null Null Sample is lower than most (97.5 % or more) of the other samples Sample is higher than most (97.5 % or more) of the other samples
Single sample t-test: one-tailed Compare the sample mean (from unknown population) to the distribution of sample means (from known population). The mean of distribution of sample means (M m ) is equal to the population mean (μ). One-tailed hypothesis: M > M m Null: M < M m or M = M m One-tailed hypothesis: M < M m Null: M > M m or M = M m

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Compare the sample mean (from unknown population) to the distribution of sample means (from known population). The mean of distribution of sample means (M m ) is equal to the population mean (μ). Two-tailed hypothesis: M > M
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 05/26/2011 for the course PSCH 343 taught by Professor Victoriaharmon during the Spring '11 term at Ill. Chicago.

### Page1 / 49

Dependent samples t-test 16 - .And Yet More About...

This preview shows document pages 1 - 11. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online