|
Title
|
Room
|
|
Complex Power and Sample-Size Analyses Using SAS-Based Tools
|
M-International Salon D
|
|
Date / Time
|
Sponsor
|
Type
|
|
08/07/2001
1:00 PM
-
5:00 PM
|
ASA, Section on Teaching of Statistics in the Health Sciences*
|
Other
|
|
Organizer:
|
n/a
|
|
Chair:
|
n/a
|
|
Discussant:
|
|
|
CE Presenter
|
Ralph O'Brien
Ralph O'Brien
|
Description
After cursorily reviewing the material contained in Dr. O'Brien's "Intro" workshop on this topic, attention will focus on how to use UnifyPow and capabilities already in the SAS System to improve the depth of your sample-size analysis, to handle more complex situations, and to look at sample-size issues in ways that are often better than those currently employed. Methods to be discussed include common tests under the multivariate general linear model; using the 'exemplary data' approach to handle virtually any problem that falls under the generalized linear model, including logistic regression, log-linear models for contingency tables, and Poisson regression; tests for equivalence; and using data on hand to get point estimates and confidence intervals for the power of a proposed study. We will also show that although the inference method may be frequentist, using elementary Bayesian constructs can sharpen and simplify the sample-size considerations. For more information on UnifyPow and to download it, visit http://www.bio.ri.ccf.org/UnifyPow.
Fees: M-$75 (after July 13 $120) NM-$100 (after July 13-$150) SM-$45(no discount after July 13)
Continuing Education Units: 0.3
|
|
