STEP 1

The data set contains 5 quantitative variables, 1 categorical variable and two factors (Gender and Ranking) which also qualify as categorical variables. The possible variable names are shown below this paragraph. They have no meaning so do not try to base any analysis on what the variable means.

You will have to create histograms and descriptive statistics for the 5 quantitative variables. Do not do statistical analyses on the factors. In this step you must review these descriptive statistics and graphics to decide if you can analyze the data provided. This requires application of the empirical rule. Based on this analysis, you must decide whether or not to use each variable in your analysis. It is expected that you may drop a variable after this step because you do not think it can be analyzed. In order to drop a variable,you must explain why it is a problem in statistical terms.Refer to slides or the book to develop your argument. (You may not drop "Job" or "Looking" and you should not analyze Element, Gender or Rank). For the remaining variables, indicate why you believe they can reasonably be analyzed using descriptive statistics. Look in Chapter 7 regarding the Central Limit Theorem.

Question 2 - Which variable do you think should be dropped?

Question 3 - What is the justification for dropping this variable?

Step 2

You must set an alpha for your analyses. This must not be the common ones, .10. .05 or .01. Provide an explanation of your choice and describe the impact it will have on your results in the context of Type I error. (Needless to say, you must also demonstrate an understanding of Type I error.)

Question 4 - What alpha do you choose?

Question 5 - What are the implications of choosing this alpha?

Question 6 - What is meant by type II error?

Describe Type II error and provide an explanation of how this will impact your results. Then answer two specific questions:

1. What precaution do we take to acknowledge the fact that we do not have any way to calculate Type II error?
2. What is one thing you can think of that might lower the risk of making a type II error. This is not something in your text. You will have to have an understanding of basic statistical principles and be creative.

Question 7 - How is this issue dealt with in basic statistics?