STA 660 - Practicum in Data Analysis

Fall 2008

Course (Section)

STA 660 (A)

Meeting Time:

 

800-915 T Th (to be rescheduled - 8 is a bad time to get clients to meet with us. Target days/times - M Th/ 2-3:15)

Meeting Location

 

219 Bachelor Hall

Prerequisite:

STA 666
Grading: 

Pass/Fail

Professor

Dr. John Bailer

E-mail:

baileraj@muohio.edu

URL:

http://www.users.muohio.edu/baileraj

Office (phone):

292 Bachelor Hall (529-3538)

FAX: 529-1493

Office Hours:

  

12:30 - 1:45 M F

11:00-12:00 W

(other hours by appointment – don’t be shy!) 

Purpose of Course:

The university catalog describes STA 660 as "Supervised practice in consulting and statistical data analysis including use of computer programs." As part of this "supervised practice," we explore strategies for tackling research questions based upon available data, and written/oral presentations of analyses.

Course Objectives:

Chatfield (1995, p. 1) suggests that a statistician needs to be able to

1. formulate a real problem in statistical terms

2. give advice on efficient data collection

3. analyze data and extract maximum amount of information

4. interpret and report the results

STA 660 gives you experience with these 4 objectives plus the tools used to accomplish these objectives. Included in these tools are statistical computing software (e.g. SAS, R/S-Plus), software for graphical displays (e.g. R/S-Plus), word processing and presentation software.

Text(s) [ Required / Required / Recommended ]:

G. van Belle (2002) Statistical Rules of Thumb. John Wiley:  New York. ISBN: 978-0-471-40227-5

N. B. Robbins (2004) Creating More Effective Graphs. John Wiley:  New Yortk. ISBN: 978-0-471-27402-5
  P.I. Good and J.W. Hardin (2006)  Common Errors in Statistics (and How to Avoid Them)  2nd Edition. John Wiley:  New York. ISBN: 978-0-471-79431-8

Another good book (used in past 660s) ...

Chatfield, C. (1998) Problem Solving: A Statistician's Guide. Chapman & Hall/CRC Press. (suggested)

Other resources:

ASA:

  

www.amstat.org (you can join for $10 as students!)

ASA Section on Statistical Consulting:

  

http://www.amstat.org/sections/cnsl/

Miller

Wiley

 Beyond ANOVA, Basics of Applied Statistics

Madansky

Springer-Verlag

 Prescriptions for Working Statistician

Fox and Long

SAGE

 Modern Methods of Data Analysis

Venables and Ripley

Springer

 Modern applied statistics with S-PLUS

Derr

Duxbury

 Statistical Consulting: A Guide to Effective Communication

General practicum activities:

Meet with clients and determine analytic needs of the client. Evaluate available data and consider data management challenges. Analyze datta to address client needs. Produce a written and oral lreport of the analysis that will be presented to the client. Interact with cleint to complete project. The project is completed when the client is satisfied.

Most of these projects will involve a groups of students working on particular tasks. Each student will be expected to be the group leader for at least one project during the term. The group leader will be responsible for coordinating team activities and for meeting deadlines.

NOTES:

  1. Working with clients is naturally open-ended. The practicum is fundamentally different from other classes.
  2. Often, more than one method is appropriate for addressing a particular client question.
  3. You may need to learn about new statistical methods in order to address a client's research hypothesis. In addition, there may not be an exact methodological fit for evaluating the data and its support of a research hypothesis.
  4. Teamwork can be frustrating; however, it is a reality of most work environments.

Parting thoughts from John Tukey (http://www.brainyquote.com/quotes/authors/j/john_tukey.html; accessed: 20 August 2008):