Correlation and Regression vs. Cause and Effect

1. In correlation and regression we are examining relationships between two or more variables (i.e., ad spending and sales). However in most studies we do we cannot say that one variable (ad spending) causes the other (sales). Why not? What has to be the case for us to discuss one variable "causing" another?

2. By the end of this week you should know which statistics test you will be using for your project. In this week's discussion I want the class to determine which test you think you should be using and why for your project. I will keep an eye on the discussion to make sure it is going in the right direction. Hint...there is a reason you are doing this discussion this week.(Below is the upcoming project criteria).

The purpose of the Class Pricing Project is for you to determine if the prices at one retail establishment are lower or higher than another retailer. You will need to compare the prices of at least 30 items (minimum sample size) at two different retailers to see if there is a statistically significant difference between the prices of the two retailers. Be sure the 30 items are exactly the same items (product, brand) sold by each retailer. For an outline of the project, follow the Steps in the Process of Hypotheses Testing that starts on page 177 of your Ravid and Haan text. Be sure to explain how you calculated your statistical test. You can do the calculations by hand or can use software (i.e., SPSS, Excel). If you did the calculations by hand show the calculations as part of your report. If you do the calculations by software, include your output in the report. The Class Pricing Project is due on the last day of class. Each student will make a brief (5-10 minute) presentation of her/his project to the professor using Eluminate or Skype during the last week of class. You need to email me to determine a time to do the presentation. Feel free to let me know if you want to discuss this project on the phone or by Skype.I will submit this portion for completion later

T-Tests and Chi Squares

3. What is the difference between a t-test and Chi-square test? What does each do? When you we use each?

4. Explain the concepts of reliability and validity. Why are they important to research and statistics? How do we measure each of these? What happens if the questionnaire we use is determined to not be reliable or valid?