| 148 | Tuesday, March 30, 2004 |
|
| 11:45 AM-1:15 PM | Convention Center:208 |
| Research Consortium |
| Measurement Symposium—New Development on Compositional Data Analysis: It Becomes Easy With Excel Freeware and SAS Routines |
| Literature has shown that the general approach (e.g., ANOVA), which is designed to analyze un-compositional data, should not be used directly to analyze compositional or percentage data. John Aitchison (1986) developed a new methodological approach for the statistical analysis of compositional data. However, there has been no easy-to-use software to help practitioners conduct compositional data analysis. Empirical researchers in kinesiology have experienced difficulty in appropriately analyzing compositional data. Recent developments in Excel-based software have made it much easier for empirical researchers to conduct the compositional data analysis. The purpose of this symposium is to introduce a set of appropriate methods and easy-to-use software for appropriately analyzing compositional data. The three presentations include the introduction of the recent development in compositional data analysis, and the application of CoDaPack (an Excel-based freeware) and SAS routines in kinesiology research. |
Keyword(s): measurement/evaluation, research
|
| Presider: Yuanlong Liu, Western Michigan University, Kalamazoo, MI |
|
Speakers: |
| |
Appropriately Analyzing Compositional Data in Kinesiology: Problems and Solutions
Yuanlong Liu, Western Michigan University, Kalamazoo, MI |
| |
CoDaPack- An Excel and Visual Basic Based Software for Compositional Data Analysis
Santiago Thió-Henestrosa, Universitat de Girona, Girona, Spain and Yuanlong Liu, Western Michigan University, Kalamazoo, MI |
| |
Ability Conceptions and the Influence of Natural Ability, Effort, Luck, Mood, and Task Difficulty in Physical Activity: A Compositional Data Analysis
Weidong Li, Louisiana State University, Baton Rouge, LA and Yuanlong Liu, Western Michigan University, Kalamazoo, MI |
|