As part of a larger study, Bratanova et al. (2015)1 investigated whether a person's taste for biscuits could be influenced by the ethicality of the manufacturing company. A fictional biscuit company was used for the study. 112 undergraduate students from a Belgian university volunteered to participate in the study.

The students were randomly assigned to one of two groups: 53 to a group that were given a description that portrayed the company as environmentally friendly and the remaining 59 to another group that were given a description that portrayed the company as environmentally harmful.

Students in both groups were then given the same type of biscuit to taste and told that it was made by the company in the description. After tasting the biscuit, both groups of students were asked to rate on a 7-point scale how likely it was that they would buy biscuits from this company in the future (Future buy). For the purpose of analysing this data we will treat Future buy as a numeric variable where 1 = not at all likely, and 7 = very likely.

Two additional variables reported for each student were:

Last ate: The number of hours since the student had eaten (Less than two hours, Two to five hours, More than five hours

Hunger: The hunger of the student, on a 7-point scale where 1 = not at all hungry and 7 = very hungry

The researchers were interested in whether the hunger rating depended on when the student last ate. Note that one student did not provide a rating for hunger and so has been excluded from this analysis. For the purpose of analysing this data we will treat a student's Hunger rating as a numeric variable.

NOTE: Use the Computer Output for parts e to g . (a) For this study describe the:

(i) units,

(ii) treatment or factor of interest,

(iii) response.

(b) (i) Draw the appropriate plots for this data set.

ANOVA Hunger Sum of Mean Squares dif Square F Sig. Between Groups 12.672 2 6.336 3.607 030 Within Groups 189.724 108 1.757 Total 202.396 110 Multiple Comparisons Dependent Variable: Hunger Tukey HSD Mean 95% Confidence Interval () Treatment [J) Treatment Difference (I-J) Std. Error Sig. Lower Bound Upper Bound Less than two Two to five hours - 548 317 .200 -1.30 21 hours More than five hours -.848* 317 023 -1.60 -.09 Two to five Less than two hours 548 317 .200 -21 1.30 hours More than five hours -.300 206 .571 -1.00 40 More than five Less than two hours 848' 317 .023 09 1.60 hours Two to five hours 300 296 .571 -40 1.00 ". The mean difference is significant at the 0.05 level.Hunger by Last_ate Less_than_two_hours CODO 8 More_than_five_hours Last ate 8 Two_to_five_hours 8 O 2 3 4 5 6 HungerPrimary variable of interest: Hunger (numeric) Secondary variable: Last_ate (categorical) Total number of observations: 111 Summary of Hunger by Last ate: Estimates Min 25% Median 75% Max Mean SD Sample Size Less than two hours 1 2 2 4 5 2.677 1.301 31 More than five_hours 1 2 4 5 6 3.525 1. 396 40 Two_to_five_hours 1 3 A 6 3. 225 1. 271 40