DOC PREVIEW
UW-Madison STAT 301 - Lectures 3 and 5 (Wardrop)

This preview shows page 1 out of 2 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 2 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 2 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

(1) Project Assignment for Statistics 301, Lectures 3 and 5 (Wardrop), Fall Semester, 2007You must perform one or two of the three projects described below and submit a written report for eachproject you select. If you perform two projects, one of them must be project 2. The report must besubmitted on paper; I will not accept electronic reports.You may work alone or as a member of a team of three or fewer persons. If you work as a team, onlyone report per project should be submitted by a team (remember to include everyone’s name on the report!)and all members of the team will receive the same grade. Neither my expectations for nor grading of reportsis influenced by the size of the team.Please submit legible reports. For many of you, this will mean words typed, and “nonwords” (e.g.tables, figures, formulas, and computations) drawn by hand. Effective use of color and illustrations cangreatly enhance a report.The maximum score for each project is 18 points.Project 1A. Due before 12:01 pm on Friday, October 19. Perform a balanced completely randomizeddesign on 100 trials with a dichotomous response to investigate some issue. Explain why the issue interestsyou and describe how you randomized. Present and summarize the data using techniques discussed in thecourse. Choose an alternative for Fisher’s test and explain your choice. Calculate two P-values for the dataand alternative: use the standard normal curve with the continuity correction, and use the website to obtainthe exact P-value. Compare your two P-values and comment. Your report should include a summary ofwhat you have learned from your study.Project 1B. Due before 12:01 pm on Friday, October 19. Perform a balanced completely randomizeddesign on 50 persons with a dichotomous response to investigate some issue. Indicate how you obtainedyour subjects and why these particular persons are of interest to you. Explain why the issue interests you anddescribe how you randomized. Present and summarize the data using techniques discussed in the course.Choose an alternative for Fisher’s test and explain your choice. Calculate two P-values for the data andalternative: use the standard normal curve with the continuity correction, and use the website to obtain theexact P-value. Compare your two P-values and comment. Your report should include a summary of whatyou have learned from your study.Project 2. Due before 12:01 pm on Friday, November 30. Perform 100 dichotomous trials under “fixedconditions” related to an activity of interest to you. Exercises 6–14 on pages 177–178 of the text presentsummaries of studies of this type that were performed by other students. Your report should describe theactivity, why it interests you, and why you are interested in the fixed conditions you study.Note that if a success is extremely difficult or extremely easy to obtain, 100 trials are not likely to revealmuch of interest. You will obtain more interesting data if you select conditions that make the probability ofsuccess on a trial be a number between 0.25 and 0.75.(a) Before you collect any data make a point guess of the value of p. (Point means single number. Forexample, if I were shooting free throws, before I collect data I would say, “My best guess is thatp = 0.58.”)(b) Use the data from your 100 trials to investigate the validity of Assumptions 2 and 3 of Bernoullitrials, using the techniques presented in lecture for Chapter 5. (Note: I put this in bold-face typebecause, historically, a large proportion of students have neglected to do this when told with regular-face type!) Your report must contain the time-order results of your 100 trials so that I can check youranalysis.(c) On the assumption you have Bernoulli trials, find the 95 percent confidence interval for p. Is yourpoint guess from part (a) in your CI? What does the CI tell you about the accuracy of your guess?(d) On the assumption you have Bernoulli trials, find the point prediction and 95 percent prediction inter-val for the number of successes in 100 future trials.(e) Perform the 100 additional trials and compare your results to your predictions from part (d).Grading of Reports. A report will not receive full credit simply because it contains no errors. Reports willbe graded on effort, creativity, report presentation (primarily clear writing), accuracy and appropriateness ofanalysis, and interpretation of results.Model projects are available on my webpage. Also click ontoResearch Papers: > Papers on Statistical Education > Small Student . .., pp. 3–9for ideas for project


View Full Document

UW-Madison STAT 301 - Lectures 3 and 5 (Wardrop)

Download Lectures 3 and 5 (Wardrop)
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Lectures 3 and 5 (Wardrop) and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Lectures 3 and 5 (Wardrop) 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?