|
Activity Number:
|
169
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistical Education
|
| Abstract - #304470 |
|
Title:
|
Accuracy in Parameter Estimation for Group Effects in Longitudinal Models: Sample Size Planning for Narrow Confidence Intervals
|
|
Author(s):
|
Ken Kelley*+ and Joseph R. Rausch
|
|
Companies:
|
University of Notre Dame and Cincinnati Children's Hospital Medical Center
|
|
Address:
|
Mendoza College of Business, Notre Dame, IN, 46545,
|
|
Keywords:
|
sample size planning ; longitudinal data analysis ; change models ; research design ; accuracy in parameter estimation ; polynomial change
|
|
Abstract:
|
Longitudinal polynomial models are widely used and are often an effective way of modeling a process over time. The accuracy in parameter estimation (AIPE) approach to sample size planning has as its goal a confidence interval whose expected width is sufficiently narrow or one where there is some desired degree of assurance that the confidence interval will be sufficiently narrow. Sample size planning methods for group effects in polynomial change models are developed from the AIPE approach. Because the group effect in a treatment/control designs is often of primary interest, planning a study so that the confidence interval will be sufficiently narrow is of key importance. The methods developed will be implemented in the MBESS R package so that they can be immediately implemented.
|
