|
Activity Number:
|
338
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Education
|
| Abstract - #307286 |
|
Title:
|
A Geometrical Approach to Introductory Statistics
|
|
Author(s):
|
Daniel Kaplan*+
|
|
Companies:
|
Macalester College
|
|
Address:
|
1600 Grand Ave., Saint Paul, MN, 55105,
|
|
Keywords:
|
geometry ; education ; regression
|
|
Abstract:
|
At Macalester College we have developed a new approach to introductory statistics with an emphasis on understanding complicated relationships among multiple variables. The course is oriented around multiple regression and multi-way analysis of variance and makes extensive use of modern computation. In our course, students learn about least-squares regression in terms of simple projections in two and three dimensions; they calculate correlation coefficients using a protractor; they perform analysis of variance using a ruler. The geometrical approach, in addition to being accessible to just about every student, makes it possible for students to reason about the results of statistical modeling in sophisticated ways. For example, Simpson's paradox does not seem like a paradox at all when presented geometrically --- it's rather obvious.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2006 program |
