Taste-testing scales. When comparing the palatability of food items, numerical taste test scales are employed. For example, the Hedonic 9-point scale uses the following ratings:

1 = dislike extremely, 2 = dislike very much, 3 =

dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, and 9 = like extremely. Another scale, called the general Labeled Magnitude Scale (gLMS), uses ratings that range from -100 (for strongest imaginable dislike)

to +100 (for strongest imaginable like). Both of these scales were applied to food items in a study published in the Journal of Food Science (Feb. 2014). Students and staff at the University of Florida taste-tested five food/beverage items: black coffee (BC), cheesecake (CC), grapefruit juice (GF), orange juice (OJ), and pepperoni (PP). Half 1n = 2002 of the subjects used the 9-point scale, and half 1n = 2002 used gLMS to rate each of the food items.

(These ratings, simulated from information provided in the article, are saved in the TASTE file.) One of the objectives of the research was to compare the average 9-point taste ratings of the five food/beverage items.

a. Explain why these data should be analyzed using an ANOVA for a randomized block design. Identify the treatments, blocks, and dependent variable for this design.

b. An SAS printout of the ANOVA is shown above, and a multiple comparisons of means summary is shown below. Interpret the results practically.

c. Refer to part

b. Perform a similar analysis for the data on the gLMS rating scale. Again, interpret the results practically.

Transcribed Image Text:

SAS Output for Exercise 10.80 Dependent Variable = Hedonic 9-point Scale The GLM Procedure Source DF Sum of Squares Mean Square F Value Pr> F Model 203 3823.677000 18.835847 9.66 F PRODUCT 4 3386.686000 TASTER 199 436.991000 846.671500 434.21 F PRODUCT 4 3386.686000 846.671500 434.21