Check the normality and constant variance assumptions graphically for the data of Example 11.17.

Example 11.17

Example 11.17 A consumer product-testing organization wished to compare the annual power consumption for five different brands of dehumidifier. Because power consumption depends on the prevailing humidity level, it was decided to monitor each brand at four different levels ranging from moderate to heavy humidity (thus blocking on humidity level). Within each humidity level, brands were randomly assigned to the five selected locations. The resulting amount of power consumption (annual kWh) appears in Table 11.8, and the ANOVA calculations are summarized in Table 11.9. Table 11.8 Power consumption data for Example 11.17 Treatments (brands) 1 2 3 4 5 Table 11.9 ANOVA table for Example 11.17 Source of variation Treatments (brands) Blocks Error Total 1 685 722 733 811 828 df 4 3 12 19 Blocks (humidity level) 2 3 792 806 802 888 920 Sum of squares 53,231.00 116,217.75 1671.00 171,119.75 838 893 880 952 978 4 875 953 941 1005 1023 Mean square 13,307.75 38,739.25 139.25 Xi. 797.50 843.50 839.00 914.00 937.25 fA = 95.57 fB = 278.20 Since fA = 95.57 F.05,4,12= 3.26, Ho is rejected in favor of Ha. We conclude that power consumption does depend on the brand of humidifier. To identify significantly different brands, we use Tukey's procedure; Q.05,5,12 = 4.51 and d = 4.51 /139.25/4 = 26.6.