Suppose a company recently hired a well-known speaker to give a seminar to its employees about Business Ethics in Data Mining. The senior managers thought the seminar was enlightening but wanted to gauge how the other employees felt about the seminar. A short 10 question survey was given to each employee after the seminar was over and they were to respond with 5 point Likert Scale response (1=Strongly Disagree; 2=Disagree; 3=Neutral; 4=Agree; 5=Strongly Agree). The survey had the following questions: • Q1. I learned a lot from the ‘Business Ethics in Data Mining’ seminar. • Q2. I will use the information I learned in the seminar in my job • Q3. I hope the same speaker can return and give seminars on other topics. • Q4. I think the ‘Business Ethics in Data Mining’ seminar was useless. • Q5. I hope my I get the chance to attend other seminars on this topic. • Q6. I would recommend the ‘Business Ethics in Data Mining’ seminar to others. • Q7. During the seminar, I caught myself day-dreaming and thinking about other things. • Q8. I felt the ‘Business Ethics in Data Mining’ seminar was worth my time. • Q9. I was impressed by the speaker’s seminar on ‘Business Ethics in Data Mining’ • Q10. The speaker was engaging, informative, and knowledgeable.

Use “Final Exam data_question 1.jmp” to answer the following questions: a) There is probably some multi-collinearity in this data, since many of the survey questions will generate similar responses from the employees. Check the correlation matrix and comment about which questions are strongly correlated. Then, use the appropriate Factor Analysis to write out the equations to combine the 10 survey questions into Factors. Use only the Questions with Loadings greater than 0.40. b) Comment on how the questions that are combined with each other. Intuitively, does it make sense to combine the questions as suggested by the Factor Analysis? c) You should notice that Q4 and Q7 have negative Loading values. Comment on why you think this is happening.