Question 5: (10 points) A software company is trying to know whether a new accounting software helps with reducing the input mistakes made by accountants. They randomly selected 12 retail companies to participate in a trial. Each company is given both the old software and the new software. They are allowed to use each software for 1 month. After the month, the number of input mistakes are recorded in a table below Company 1 2 3 4 5 6 7 8 9 10 11 12 Old software 3 10 4 6 2 1 0 4 2 5 15 3 New software 2 9 2 3 3 1 1 5 1 3 10 2 a. (2) Name a possible confounding variable in this experiment. Suggest a way to neutralize the effect of that confounding variable. b. (5) Is there any evidence to suggest that the new software reduce the mean number of input mistakes? Perform a hypothesis test at 5% level of significance. c. (3) Create a 90% confidence interval for the average difference in input mistakes. Interpret your interval. What does this suggest about the new software