|
Activity Number:
|
191
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 10, 2004 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Teaching Statistics in the Health Sciences
|
| Abstract - #300895 |
|
Title:
|
Development of a Web-supported Biostatistics Module that Includes Automatic Marking of Randomly Generated Datasets
|
|
Author(s):
|
Paul Corey*+ and Malcolm Koo and Andrew White and Raymond Tam
|
|
Companies:
|
University of Toronto and University of Toronto and Source Level Computing and Source Level Computing
|
|
Address:
|
First Floor McMurrich Bldg. Fac Medicine, Torornto, ON, M5S 1A8, Canada
|
|
Keywords:
|
web ; education ; biostatistics
|
|
Abstract:
|
Language study includes vocabulary (words) and their rules (grammar) of association. Applied statistics is a foreign language. The ultimate education has student and teacher analyzing data, relating statistical practice to a scientific question and making decisions in face of uncertainty. Preparation for such enriched education involves learning the vocabulary and grammar of statistics. The vocabulary includes means and standard deviations. Grammar arranges these "words" into measures of statistical evidence, conventionally consisting of p values and confidence intervals. Driven by ASP and MySQL, a web module was developed to support a graduate medical statistics course. Students receive questions selected in real time from a database of datasets randomly generated using SAS and stored securely on the server. Correctness of the statistical analysis is examined using a multiple-choice format. Feedback from quizzes prepares the student for an online timed exam with a random dataset. Quizzes and exam marks are automatically made available to students. This module can be used to support a course in statistics or as a screen of statistical proficiency for a course or program.
|
