Course Description
Provides undergraduate statistics majors with an exposure to
advanced statistical methods, beyond regression and analysis of
variance, and introduces the student to a data-analysis
experience related to a real scientific problem. In addition to
learning and applying statistical techniques, effective oral and
written communication of methods and results are emphasized.
Athena Title
Statistical Capstone Course I
Equivalent Courses
Not open to students with credit in STAT 5010W
Prerequisite
STAT 4220 and STAT 4230/6230 and (STAT 4365/6365 or STAT 4365E/6365E)
Pre or Corequisite
STAT 4510/6510
Semester Course Offered
Offered fall
Grading System
A - F (Traditional)
Student learning Outcomes
- Students will be able to understand advanced statistical methods which will round out their education.
- Students will be able to understand advanced statistical methods which will help them see that there is more out there than t-tests, ANOVA, and regression.
- Students will be able to understand advanced statistical methods which will give them more confidence in their own data analysis skills.
- Students will be able to understand advanced statistical methods which will make them more attractive to prospective employers.
- Students will understand how to write about statistical methods and results effectively.
- Students will have completed several writing assignments during the Statistical Capstone courses sequence, culminating in a final report of their data analysis project.
Topical Outline
- A wide variety of topics can be covered in this capstone course. Since all students will have gone through courses in which they learned the basic methods of regression and ANOVA, instructors can concentrate on other techniques according to their preferences, the preferences of the class, and the particular projects the students will be working on. Possibilities include time series analysis, data mining, factor analysis, classification and regression trees, smoothing, bootstrap, and categorical data analysis (contingency tables, loglinear models). The idea is not to go into any one of these topics in-depth and exclusively; rather, we propose covering many different topics over the course of the year, spending at most two or three lectures on each. In addition, the course will include units on effective communication (written and oral) and how to make a poster presentation. The rest of the class periods will be used for student presentations of their data sets and analyses.
