Data Analysis for a Music Company

You are working for a company in the music business. They make stereos, musical instruments, produce CD's, etc. Their business is in a slump. So they decided to do a survey of college students on a variety of issues, presumably to develop an effective ad campaign and perhaps restructure their operation. Basically, they wanted to know about college students' interests and practices vis-à-vis music.

Their survey was administered to a random sample of students taking a course in elementary statistics at Penn State. Their resources were a bit limited so they decided to save money by picking a "typical" state university to do the survey. They asked a stat instructor if he would have students in his Stat 200 class fill out the survey. The instructor said he had already planned on having students fill out a survey for use in class projects, so he would just add in the items the company was interested in.

The president of the company hired the members of your group to look at the data, prepare a report to her, and do whatever statistical analyses appropriate to support your conclusions and recommendations. They hired you after the survey was completed, so you had no opportunity to suggest changes in the survey and/or how and to whom it would be administered.

The items on the survey form were as follows:

C1. What is your gender? Male ___ Female ___

C2. What is your race? Caucasian ___ Non-Caucasian ___

C3. Which kind of growth is most important in college? Intellectual ___ Social ___

C4. What type of TV show do you prefer to watch? Comedy __ Drama ___ Mini-Series ___ Other ___

C5. What is your favorite kind of music? Alternative ___ Classic Rock ___ Rap ___ Top 40 ___ Other ___

C6. What is your least favorite kind of music?

Heavy metal ___ Rap ___ 4: Country ___ Other ___

C7. Does MTV show favoritism in type of music played? No ___ Yes ___ Don't know __

C8. Do you belong to a music club? No ___ Yes ___

C9. Do you watch music videos? No ___ Yes ___

C10. Which do you prefer? MTV ___ VH1 ___ No preference ____

C11. How many music CDs do you buy each year? Number of CDs ____

C12. On average, how much money do you spend per month on Cds? Dollars _____

C13. How much are you willing to spend on a stereo system? Dollars _____

C14. Do you play at least one musical instrument? No ___ Yes ___

C15. How many "albums" have you bought in the last year? Number of Albums ____

C16. About how many CDs or cassettes do you own? Number of CDs or Cassettes ____

C17. If you were stranded on a deserted island, which would you prefer to have?

Telephone ___ TV ___ CD player/stereo ___

C18. Do you prefer to study with music or without music? Without music ___ With music ___

C19. Do you feel that your interests in music have broadened since coming to college?

No ___ Yes ___

C20. Do you have a significant other? No ___ Yes ___

The responses of the students surveyed were stored in the following data set: wlh_01.txt.

Project Requirements.

    1. This is a group project. Each member of the group is to contribute his/her share to the successful completion of the project. When the project is submitted, you will be asked to distribute a fixed number of points to each member of your group (excluding yourself) as an evaluation of their contributions to the project. Details will be given later. There will be an individual score (based on the group member's evaluation of your effort) and a group score.

     

    1. Your report must include an analysis of at least one of each of the following six inference types:

    1. One proportion
    2. Two proportions
    3. A measure of the center of a distribution
    4. Comparison of the center of two groups
    5. Relationship between two categorical variables with at least one of the variables having 3 or more values.
    6. Comparison of the centers of three or more groups.

    1. Your report must follow the guidelines for submission of the report, in terms of format and contents, which will be given to you separately.

Comments.

    1. There is basically only one procedure for analyzing proportions. Use the ones we have gone over in class.
    2. The analysis of the centers of numerical data can be done in two ways: in terms of means or medians. We have only discussed making inferences for means--these procedures assume the data is reasonably symmetric. If the data is highly skewed, procedures called 'nonparametrics' should be used. You can find a selection of 7 nonparametric procedures on the Stat Menu under 'Nonparametrics'. A description of them, in terms of when they are used, can be obtained using the 'Help' menu.
    3. We will devote two full labs to the project, so as to minimize time outside of class. I encourage you to communicate with one another both in-class and out-of-class on the project work. I will also be available to answer questions in the lab and through email and other means.