Study Design

(Include your response to questions from question 1 and 2 on the paper.)

139 people were given fliers in a randomised block design before entering a resaraunt. One flier had an offer of a free drink and a $4 buffet whereas the other flier offered a free drink and an $8 buffet. Of the 139 participants, 122 respondants gave relevant resposes to a survey taken after their meal. This is not a blind experiment and follows a randomised block design.

Conclusion

(Include response to question 3.)

From the table of demographics, we see that none of the characteristics (Age, Gender, Height, Weight, Number in Group, and whether they were hungry or not when they walked in) produced significant p-values when comparing the two groups using ANOVA.

The respondants were asked to rate the extent that they agree with statements about their meal out of 9. 'The pizza, in general, tasted really great' produced a mean of 6.89 for the $4 group and 7.44 for the $8 group. This gave an F value of 4.24 and a p-value of 0.04 which is statistically significant. Therefore, by the anova test we can conclude that there is a significant difference between the mean rating of the food for the two different price groups.

There are, however, certain limitations. The subjects were not aware if the flier was a special offer or not. Experienced customers will know that this is not the case and so would be likely to act differently to inexperienced customers recieving the flier. As the level of experience was not included in the demographics table, this could be a source of potential bias.

Also, there was no direct evidence of the conditions of ANOVA being met. F values were produced for all comparisons but the conditions were assumed (or this has not been included in the report). Therefore the analysis of variances (ANOVA) in this case may not be justified and could affect the conclusions based on p-values.