What do we mean by reliability? Validity? Define both terms as it relates to measurement. Can a measure be valid if it is not reliable? What statistical analyses can we deploy to ensure our measures are reliable? How do we check the validity of our measures?

2. Suppose your supervisor issues a survey to measure customer satisfaction at three locations in town. She suspects that there will be a difference between the three locations and that customers enjoy Location C more than Locations A and B, and is considering a change in management A and B.

She returns the data to you and asks you to compare the means in overall customer satisfaction between the three groups. What statistical test do you run to determine whether there is a significant difference between the groups? What is the logic behind this test (that is, what is going on "under the hood")? Your statistics program determines that there was a difference between the three groups and the significance level is p < .01. What does this value mean? How confident are you that a significant difference exists between groups?

3. List and define a) the measures of central tendency and b) the different levels of measurement. What are the strengths and weaknesses of the different measures of central tendency? Is it appropriate to use all the measures of central tendency with each level of measurement? Why or why not?

4. Imagine you are trying to understand what factors contribute to high customer satisfaction for a regional airline. You have four variables that you think might be related to customer satisfaction: 1) price 2) cleanliness of aircraft 3) available destinations and 4) quality of customer service.

What test would you run to determine whether the four above variables are related to customer satisfaction? Suppose the strengths of those relationships are .67, .17, -.33, and .90, respectively. Which value indicates the strongest relationship? Which value indicates the weakest? What do these values mean? If you wanted to know how well each variable predicts customers satisfaction, what analysis would you run? Can you make a claim of causality from either of these two tests? Why or why not?