AMS588. One-way Repeated Measures ANOVA: The Univariate and the Multivariate Analysis Approaches Suppose there are k regions of interest (ROI’s) and n subjects. Each subject was scanned on baseline (soda) as well as after drinking alcohol. Our main hypothesis is whether the change between baseline and alcohol is homogeneous among the ROI’s. That is H 0 : 1 2 k , where j is the effect of alcohol on the jth ROI, j 1, , k . The Univariate Analysis Approach For subject i, let Y ij denote the paired difference between baseline and alcohol for the jth ROI, then the (univariate) repeated measures ANOVA model is: Yij j Si ij , where j is the (fixed) effect of ROI j, Si is the (random) effect of subject i, ij is the random error independent of Si . With normality assumptions, we have iid Si N 0, s2 and ij i 1, iid N 0, 2 . Let Yi Yi1 , Yi 2 , , n , where 1 , 2 , , Yik , we have Yi ' iid Nk , , , k and ' s2 2 s2 s2 1 1 2 2 2 2 s s s s2 2 with 2 and s 2 2 1 s2 s2 2 s 2 s2 2 . This particular structure of the variance covariance matrix is called “compound symmetry”. For each subject, it assumes that the variances of the k ROI’s are equal 2 and the correlation between each ROI pair is constant , which may not be realistic. The univariate approach to one-way repeated measures ANOVA is equivalent to a twoway mixed effect ANOVA for a randomized block design with subject as the blocks and ROI’s as the “treatments”. The degrees of freedom for the ANOVA F-test of equal treatment effect is k 1 and n 1 k 1 respectively. That is, F0 H0 Fk 1, n 1 k 1 . We will reject the null hypothesis at the significance level if F0 Fk1, n 1 k 1 . The Multivariate Analysis Approach Alternatively, we can use the multivariate approach where no structure, other than the usual symmetry and non-negative definite properties, is imposed on the variance covariance matrix in Yi iid N k , , i 1, , n . Certainly we have more parameters In this model than the univariate repeated measures ANOVA model, and thus we must CQC CY have at least k 1 subjects. The test statistic is T02 n n 1 CY n n i 1 i 1 where Y Yi , Q 1 1 0 0 1 1 ' Yi Y Yi Y , and C 0 0 0 Under the null hypothesis, T02 ' 1 ' 0 0 . 1 1 0 0 Hotelling ' s Tk21,n1 . The Hotelling’s Tk21,n 1 statistic has the following relationship with the F statistics: F0 T02 n k 1 k 1 n 1 H0 Fk 1,n k 1 . We will reject the null hypothesis at the significance level if F0 Fk1,nk 1 . When to use what approach? There are more parameters to be estimated in the multivariate approach than in the univariate approach. Thus, if the assumption for univariate analysis is satisfied, one should use the univariate approach because it is more powerful. Huynh and Feldt (1970) give a weaker requirement for the validity of the univariate ANOVA F-test. It is referred to as the “Type H Condition”. A test for this condition is called the Machly’s sphericity test. In SAS, this test is requested by the “PrintE” option in the repeated statement. Example. One-way Repeated Measures ANOVA (n=4, k=4) Subject ROI 1 ROI 2 ROI 3 1 5 9 6 2 7 12 8 3 11 12 10 4 3 8 5 ROI 4 11 9 14 8 SAS Program: One-way Repeated Measures Analysis of Variance data repeatM; input ROI1-ROI4; datalines; 5 9 6 11 7 12 8 9 11 12 10 14 3 8 5 8 ; proc anova data=repeatM; title 'one-way repeated measures ANOVA'; model ROI1-ROI4 = /nouni; repeated ROI 4 (1 2 3 4)/printe; run; SAS Output: One-way Repeated Measures Analysis of Variance 1. Estimated Error Variance-Covariance Matrix ROI_1 ROI_2 ROI_3 ROI_1 ROI_2 ROI_3 10.00 8.00 7.00 8.00 16.75 11.75 7.00 11.75 8.75 2. Test for Type H Condition --- Mauchly's Sphericity Tests (Note: p-value for the test is big, so we can use the univariate approach) Variables Orthogonal Components DF Criterion Chi-Square Pr > ChiSq 5 0.0587599 4.8812865 0.4305 3. Multivariate Analysis Approach --- Manova Test Criteria and Exact F Statistics for the Hypothesis of no drug Effect Statistic Value F Value Num DF Den DF Pr > F Wilks' Lambda 0.00909295 36.33 3 1 0.1212 Pillai's Trace 0.99090705 36.33 3 1 0.1212 Hotelling-Lawley Trace 108.97530864 36.33 3 1 0.1212 Roy's Greatest Root 108.97530864 36.33 3 1 0.1212 4. Univariate Analysis Approach --- Univariate Tests of Hypotheses for Within Subject Effects Source ROI Error(ROI) DF Anova SS 3 50.25000000 9 13.25000000 Adj Pr > F Mean Square F Value Pr > F G - G H - F 16.75000000 11.38 0.0020 0.0123 0.0020 1.47222222 Greenhouse-Geisser Epsilon Huynh-Feldt Epsilon 0.5998 1.4433 Interpretation Note that the multivariate F-test has value of 36.33, degrees of freedom of 3 and 1, and the p-value is 0.1212. While the univariate F-test has value of 11.38, degrees of freedom of 3 and 9, and the p-value is 0.0020. In this case, since the assumption for the univariate approach is satisfied, we use the univariate approach which is more powerful (smaller pvalue).