Table 5.4. (b) Test at the 1 level oI signiIicance the hypothesis that the means Ior Iactor A populations (Iertilizers) are identical. (c) Test at the 1 level oI signiIicance the hypothesis that the means Ior Iactor B populations (pesticides) are identical. (a) See Table 5.23. Table 5.23 Two-Factor ANOVA Table Ior EIIect oI Fertilizers and Pesticides on Output Variation Sum oI Squares Degrees oI Freedom Mean Square F Explained by Iertilizer (between columns) SSA 156 c 1 3 MSA 52 MSE MSA
15.48 Explained by pesticide (between rows) SSB 72 r 1 2 MSB 36 MSE MSB
10.71 Error or unexplained SSE 37 (r-1) (c-1) 6 MSE 3.36 Total SST 265 rc 1 11 -
(b) The hypotheses to be tested are H 0 : 3 1 3 2 3 3 3 4 versus H 1 : 3 1 3 2 3 3 3 4 are not equal where 3 reIers to the various means Ior Iactor A (Iertilizer) populations. For Iactor A. F: 9.78 (Irom App, 7) Ior degrees oI Ireedom 3 (numerator) and 6 (denominator) and - : 0.01. Since the calculated value oI F 15.48 (Irom Table 5.23) exceeds the tabular value oI F, we reject H 0 and accept H 1 , that the means Ior Iactor A (Iertilizer) populations are not equal. (c) The second set oI hypotheses to be tested consist oI H 0 : 3 1 3 2 3 3 versus H 1 : 3 1 3 2 3 3 are not equal but now 3 reIers to the various means Ior Iactor B (pesticide) populations. For Iactor B. F 10.92 (Irom App, 7) Ior degrees oI Ireedom 2 and 6 and - : 0.01. Since the calculated value oI F 10.71 is smaller than the tabular value oI F, we accept H 0 , that the means Ior Iactor B (pesticide) populations are equal. Note that in two-way or two- Iactor analysis oI variance (with an ANOVA table similar to Table 5.23) we can test 94 null hypotheses, one Ior Iactor A and one Ior Iactor B.
5.27 Table 5.24 gives the Iirst-year earnings (in thousands oI dollars) oI students with master`s degrees Irom 5 schools and Ior 3 class rankings at graduation. Test at the 5 level oI signiIicance that the means are identical (a) Ior school populations and (b) Ior class-ranking populations.
Table 5.24 First-Year Earnings oI MA Graduates oI 5 Schools and 3 Class Ranks (In Thousands oI Dollars) Class Rank School 1 School 2 School 3 School 4 School 5 Sample Mean Top 1/3 20 18 16 14 12 X 1 16 Middle 1/3 19 16 13 12 10 X 2 14 Bottom 1/3 18 14 10 10 8 X 3 12 Sample mean X 1 19 X 2 16 X 3 13 X 4 12 X 5 10 X 14