Question 410:
1Answer:
No answer provided yet.A different methodology might be sampling over a longer period than one hour for one week. Instead extend it randomly for 1 hour across the day across several days and weeks.
The null hypothesis is that there is no difference between campaigns as measured by satisfaction scores. I'd conduct a 2-sample t-test. I have a beta-version on my sister website usablestats. Paste the values in each column. You should see an observed difference of -3.4 (meaning group 1 was on average 3.4 sat points better). The p-value of .013 is less than .05 suggesting this difference is not likely due to chance alone. Conclusion, there is a difference in campaigns-- go with the Pepperoni Mushroom.