A Course on ''Web-Based Statistics''
Jürgen Symanzik & Natascha Vukasinovic
Utah State University
Department of Mathematics and Statistics
E-mail:
symanzik@sunfs.math.usu.edu
URL:
http://www.math.usu.edu/~symanzik
Introduction
Motivation
Statistics and the Web closely related these days
Previously no course that provided overview on Statistics and the Web
Provide broad overview on Web-related topics
Motivate students to further explore the Web (and read related articles)
Course Format
4 week summer course
15 lectures, 110 minutes each
Intended audience: advanced undergraduates & early graduates
Background in Statistics, Mathematics, Education, Computer Science, Administrative Sciences, or Natural Resources
Attracted audience: 7 students + 2 audits (from Statistics and some of the fields above)
Electronic classroom with PC for each student
Course home page, containing all material related to the class
URL is
http://www.math.usu.edu/~symanzik/teaching/2000_stat5810/stat5810.html
Topics
References
Computing Science and Statistics proceedings
CompStat proceedings
ISI Bulletin
Little published in journals
Activities
A1: Exploration and presentation of Statistical Summary Pages
A2: Exploration and presentation of Data Sources on the Web
A3: Exploration and presentation of Teaching Tools and Applets
A4: Exploration and presentation of Electronic Textbooks
A5: Exploration and presentation of Electronic Journals on the Web
Assignments
H1: Written summary of Activity 1
H2: Written summary of Activity 2
H3: Find a data set on the Web and describe how you succeeded (based on Activity 2)
H4: Written summary of Activity 3
H5: Written summary of Activity 4
H6: Preparation and delivery of an ''electronic introductory statistics'' (18 minute) lecture and evaluation of the other short lectures
H7: Written summary of software Web sites
H8: Written summary of a paper on the Web and (specialized) statistical applications
H9: Preparation and presentation of a 45 to 50 minute introduction of a Web-based statistical software package and evaluation of other presentations
H10: Solve simple questions using the Web-based software packages introduced in class and Homework 9
H11: Written summary of Activity 5
H12: Written summary of a paper on electronic / Web-based data collection
Answers to Homeworks available on the Web
Projects
1 paper submitted to Computational Statistics
1 paper in preparation for JSE or American Statistician
Feedback
Intro Survey
Questions:
Info about the course
Previous experience with Web
Future expectations
Responses:
Flyers in Stat Department or advisor
Used Web for literature search and fun
To use Web more for statistical applications (software + data)
Midterm Survey
Questions:
Expectations met so far
Format of the course
Amount of homework and reading
What to change
Responses:
In most cases yes (if any)
OK (some complaints about group work)
Somewhat too much
Reduce homework and reading assignments
Final (official) Course Evaluation (on a scale 1 - 6)
Overall quality of the course: 5.1
Instructor's effectiveness: 5.0
Comments:
+ : Interaction, in-class presentations, learned more than expected.
- : Too short time, too much work outside classroom.
Conclusions
15 lectures (110 minutes each) is too little to provide an in-depth discussion, but enough for a general overview
If more time is available, topics such as Web-based software and Web-based teaching tools could be extended
Depending on students' background and interests, additional topics could be added:
Basics of creating Web pages (html ?)
Advanced techniques of creating Web pages (Javascript, Java, xml, Flash, ...)
Analysis and visualization of Web traffic
...
Students liked the class and were surprised to find that much related to Statistics on the Web
Final survey (about 1 year after the course) is planned for this summer