a) Was there a statistically significant difference among the Weight groups within the Very Good (1) diet subjects? Explain.

The default error term in MANOVA has been changed from WITHIN CELLS to WITHIN+RESIDUAL. Note that these are the same for all full factorial designs. * * * * * * * * * * * * * * * * *Analysis of Variance* * * * * * * * * * * * * * * * * 357 cases accepted. 0 cases rejected because of out-of-range factor values. 0 cases rejected because of missing data. 9 non-empty cells. 2 designs will be processed. * * * * * * * * * * * * * * * * *Analysis of Variance -- Design 1 * * * * * ** * * * * Tests of Significance for LBXWBCSI using UNIQUE sums of squares Source of Variation SS DF MS F Sig of F WITHIN+RESIDUAL 1423. 64 348 4. 09 DIET 03 2 . 02 .00 996 WEIGHT WITHIN DIET(1 2.27 N 1. 14 28 . 758 ) WEIGHT WITHIN DIET(2 13. 15 2 6.57 1. 61 . 202 WEIGHT WITHIN DIET (3 105. 98 2 52.99 12.95 000 (Model) 121.51 8 15. 19 3.71 . 000 (Total) 1545 . 15 356 4.34 R-Squared = . 079 Adjusted R-Squared = .057