Comparing Means Independentsamples ttest Population A Population B

Comparing Means: Independent-samples t-test Population A Population B OR Sample 1 Sample 2 Lesson 14

Independent-samples t test Select 2 independent samples l are they from same population? n Experiment l Select 2 samples l Each gets 1 level of independent variable n H 0: there is no difference l samples from same population n H 1: there is a difference l Samples from different populations ~ n

t test: Independent-Samples n Sample statistic: n Same basic structure as single sample n Independent samples [df = N 1 + N 2 - 2]

Estimated Standard Error n Standard error of difference between 2 sample means l s 2 p = pooled variance ~

Pooled Variance (s 2 p) n Average of 2 sample variances l weighted average if N 1 N 2

Assumptions: Independent-samples t test 1. Samples are independent 2. Samples come from normal populations 3. Assume equal variance s 21 = s 22 u does not require s 21 = s 22 Homogeneity of variance n t test is robust l uviolation of assumptions u. Little effect on p(rejecting H 0) ~

Example : Independent-samples Does exercising longer have greater health benefits? n Participants (N=14) l Randomly assigned to 2 groups l N=7 per group n Manipulation (IV) l Group 1 exercised 2 hrs/week l Group 2 exercised 5 hrs/week n Outcome (DV): Health rating ~ n

PASW Independent -Sample T Test n n n Data entry l 2 columns l 1 for IV, 1 for DV Menu l Analyze l Compare Means l Independent-Sample T Test Dialog box l Test Variable(s) (DV) l Options (to change confidence intervals) ~