In exercises that call for two-sample procedures, you may use either of the two approximations for the degrees of freedom that we have discussed: the value given by your software or the smaller of and . Be sure to state clearly which approximation you have used.

Question 7.67

7.67 More on smart shopping carts.

Recall Example 7.10 (pages 381–382). The researchers also had participants, who were not told they were on a budget, go through the same online grocery shopping exercise.

smart1

  1. For this set of participants, construct a table that includes the sample size, mean, and standard deviation of the total cost for the subset of participants with feedback and those without.
  2. Generate histograms or Normal quantile plots for each subset. Comment on the distributions and whether it is appropriate to use the procedures.
  3. Test that the average cost of the cart is the same for these two groups using the 0.05 significance level. Write a short summary of your findings. Make sure to compare them with the results in Example 7.10.

7.67

(a) For those with feedback, . For those without feedback, . (b) Both Normal quantile plots show the two variables are both roughly Normally distributed. (c) . The data are significant at the 5% level, and there is evidence the two groups are different in total cost for those with and without feedback among those who were not told they were on a budget. The results are similar to those in Example 7.10; feedback helped reduce spending.