MTB > read 'a:\ed257\drugrep.dat' c1-c3 20 ROWS READ ROW C1 C2 C3 1 30 1 1 2 14 2 1 3 24 3 1 4 38 4 1 . . . MTB > Table c2 c3; SUBC> stats c1. ROWS: C2 COLUMNS: C3 1 2 3 4 ALL 1 1 1 1 1 4 30.000 28.000 16.000 34.000 27.000 -- -- -- -- 7.746 2 1 1 1 1 4 14.000 18.000 10.000 22.000 16.000 -- -- -- -- 5.164 3 1 1 1 1 4 24.000 20.000 18.000 30.000 23.000 -- -- -- -- 5.292 4 1 1 1 1 4 38.000 34.000 20.000 44.000 34.000 -- -- -- -- 10.198 5 1 1 1 1 4 26.000 28.000 14.000 30.000 24.500 -- -- -- -- 7.188 ALL 5 5 5 5 20 26.400 25.600 15.600 32.000 24.900 8.764 6.542 3.847 8.000 8.861 CELL CONTENTS -- C1:N MEAN STD DEV MTB > name c1 'reactime' c2 'subject' c3 'drug' MTB > anova reactime = subject drug; SUBC> random subject; SUBC> restrict; SUBC> ems. Factor Type Levels Values subject random 5 1 2 3 4 5 drug fixed 4 1 2 3 4 Analysis of Variance for reactime Source DF SS MS F P subject 4 680.80 170.20 18.11 0.000 drug 3 698.20 232.73 24.76 0.000 Error 12 112.80 9.40 Total 19 1491.80 Source Variance Error Expected Mean Square component term (using restricted model) 1 subject 40.200 3 (3) + 4(1) 2 drug 3 (3) + 5Q[2] 3 Error 9.400 (3) -------------------------------------------------- Winer Table 4.3-2 arranges this anova as Source SS df MS F Between persons 680.8 4 Within persons 811.0 15 Drugs 698.2 3 232.73 24.76 Residual 112.8 12 9.4 Total 1491.8 19 use F(3,12)