This preview shows page 1. Sign up to view the full content.
Unformatted text preview: set of all pairwise comparisons among population means. In our previous Drug example, the possible pairwise comparisons are: Drug 1 with Drug 2, Drug 1 with Drug 3, and Drug 2 with Drug 3. To compare the pair of means we could … Compute a confidence interval for the difference between the two population means and see if 0 falls in the interval or not. Perform a test of hypotheses to assess if the two population means differ significantly. When many statistical tests are done there is an increased risk of making at least one type I error (erroneously rejecting a null hypothesis). Consequently, several procedures have been developed to control the overall family type I error rate or the overall family confidence level when inferences for a set (family) of multiple comparisons are done. Tukey’s procedure is one such procedure for the family of pairwise comparisons. If the family error rate is not a concern, Fisher’s procedure is used. Try It! Comparing 3 Drugs In the comparison of the three drugs, we rejected the null hypothesis at the 5% significance level. We follow with a multiple comparison procedure to determine which group means are significantly different from each other. SPSS multiple comparisons using Tukey's method and a family confidence level of 95%: Multiple Comparisons
Dependent Variable: TIME
Tukey HSD (I) DRUG
1
2 3 (J) DRUG
2
3
1
3
1
2 Mean
Difference
(IJ)
1.0800
1.4200
1.0800
2.5000*
1.4200
2.5000* Std. Error
.9485
.9485
.9485
.8659
.9485
.8659 Sig.
.505
.318
.505
.027
.318
.027 95% Confidence Interval
Lower Bound Upper Bound
3.5276
1.3676
1.0276
3.8676
1.3676
3.5276
.2657
4.7343
3.8676
1.0276
4.7343
.2657 *. The mean difference is significant at the .05 level. a. Use the above output to report about the three pairwise comparisons: Does the confidence interval for 1 2 contain 0? ___Yes____ Does the confidence interval for 1 3 contain 0? ___Yes____ Does the confidence interval for 2 3 contain 0? ___No_...
View Full
Document
 Summer '10
 Gunderson
 Statistics, Variance

Click to edit the document details