Question 15.49

15.49 Consumer-generated ads.

More and more companies involve consumers in the process of developing advertisements. Is it beneficial to let consumers know this? In one study, 125 undergraduate students were randomly assigned to a design.18 Each student watched one of two consumer-generated Doritos ads. Half of the students were told that the ad was consumer-generated and the other half were not. After viewing the ad, the students provided their reactions to the ad and the advertised brand with higher scores reflecting a more favorable opinion. Here is part of the ANOVA table for their reactions to the ad:

Source Degrees of freedom Sum of squares Mean square
A (Ad) 3.054
B (Informed) 7.813
AB 1.876
Error 146.807
Total
  1. Fill in the missing values in the ANOVA table.
  2. What is the value of the statistic to test the null hypothesis that there is no interaction? What is its distribution when the null hypothesis is true? Using Table E, find an approximate -value for this test.
  3. Answer the questions in part (b) for the main effect of advertisement and the main effect of being informed that the ad was consumer-generated.
  4. What is , the within-group variance? What is ?
  5. The mean score when disclosing that the ads were consumer-generated was 4.52. The mean score when this was not disclosed was 5.23. Using these results and your answers to parts (a)-(d), summarize the results.

15.49

(a)

Source DF SS MS
A (Ad) 1 3.054 3.054 2.52
B (Informed) 1 7.813 7.813 6.44
AB 1 1.876 1.876 1.55
Error 121 146.807 1.2133
Total 124 159.55

(b) , the distribution is
(c) Ad: , the distribution is Informed: , the distribution is
(d) (e) There is no significant interaction, as well as no significant Ad effect. Informed was significant, those who were informed that the ad was consumer-generated had a higher average opinion, 5.23, than those that weren’t informed, 4.52.