1. A consumer group wants to compare a new brand of wax (Brand X) to two leading brands (Sureglow and Microsheen) in terms of Effectiveness of wax. Following data is collected for this purpose:
Brand Effectiveness Brand Effectiveness Brand Effectiveness
Sure glow 93 Mirror sheen 90 Brand x 105
Sure glow 96 Mirror sheen 97 Brand x 91
Sure glow 87 Mirror sheen 91 Brand x 95
Sure glow 91 Mirror sheen 94 Brand x 107
Sure glow 88 Mirror sheen 100 Brand x 90
Sure glow 85 Mirror sheen 95 Brand x 96
Sure glow 88 Mirror sheen 88 Brand x 92
Sure glow 91 Mirror sheen 92 Brand x 94
Sure glow 82 Mirror sheen 94 Brand x 84
Sure glow 91 Mirror sheen 89 Brand x 86
Sure glow 86 Mirror sheen 96 Brand x 82
Sure glow 93 Mirror sheen 91 Brand x 91
Sure glow 91 Mirror sheen 97 Brand x 106
Sure glow 87 Mirror sheen 92 Brand x 90
Sure glow 88 Mirror sheen 92 Brand x 91
Brand x 92
Brand x 91
Brand x 106
Brand x 98
Brand x 97
Brand x 91
Brand x 99
Brand x 86
Using data analysis run the identified ANOVA test to analyse this data. Copy and paste your summary table in your word for submission.
1 answer
First, we calculate the total sum of squares (SST):
SST = SSbetween + SSwithin
Next, we calculate the sum of squares between groups (SSbetween):
SSbetween = Σ(ni * (mean_i - grand_mean)^2)
Where ni is the number of observations for each brand, mean_i is the mean effectiveness for each brand, and grand_mean is the overall mean effectiveness.
We also need to calculate the sum of squares within groups (SSwithin):
SSwithin = Σ[(xi - mean_i)^2]
Where xi is each individual observation for each brand.
Finally, we calculate the mean sum of squares between groups (MSbetween) and mean sum of squares within groups (MSwithin):
MSbetween = SSbetween / (k - 1)
MSwithin = SSwithin / (N - k)
Where k is the number of groups (brands) and N is the total number of observations.
Using this information, the ANOVA table can be constructed:
Source | SS | df | MS
-------------------------------------------------
Between Groups | ______ | ____ | ______
Within Groups | ______ | ____ | ______
Total | ______ | ____ | ______
To calculate the values for SSbetween, SSwithin, MSbetween, and MSwithin, we need to find the mean effectiveness for each brand and the overall mean effectiveness.
mean_sureglow = (93 + 96 + 87 + 91 + 88 + 85 + 88 + 91 + 82 + 91 + 86 + 93 + 91 + 87 + 88) / 15 = 89.67
mean_mirror_sheen = (90 + 97 + 91 + 94 + 100 + 95 + 88 + 92 + 94 + 89 + 96 + 91 + 97 + 92 + 92) / 15 = 93.27
mean_brand_x = (105 + 91 + 95 + 107 + 90 + 96 + 92 + 94 + 84 + 86 + 82 + 91 + 106 + 90 + 91 + 92 + 91 + 106 + 98 + 97 + 91 + 99 + 86) / 23 = 93
grand_mean = (mean_sureglow + mean_mirror_sheen + mean_brand_x) / 3 = (89.67 + 93.27 + 93) / 3 = 91.31
Now we can calculate the sums of squares for each factor:
SSbetween = (15 * (89.67 - 91.31)^2) + (15 * (93.27 - 91.31)^2) + (23 * (93 - 91.31)^2) = 411.9 + 54.45 + 126.33 = 592.68
SSwithin = [(93 - 89.67)^2 + (96 - 89.67)^2 + ... + (90 - 93)^2 + (91 - 93)^2 + ... + (86 - 93)^2] = 563.44
Now we can calculate the mean sum of squares for each factor:
MSbetween = SSbetween / (3 - 1) = 592.68 / 2 = 296.34
MSwithin = SSwithin / (15 + 15 + 23 - 3) = 563.44 / 50 = 11.27
Finally, calculating the F-value:
F = MSbetween / MSwithin = 296.34 / 11.27 = 26.29
We can compare the F-value to the critical value at a chosen significance level (alpha) and degrees of freedom for between groups and within groups. If the calculated F-value is greater than the critical F-value, we can reject the null hypothesis and conclude that there is a significant difference in the effectiveness of these wax brands.