Approaches to the nonparametric analysis of limited longitudinal data sets.
The traditional goals of longitudinal studies are many: consideration of stability and change; description of patterns of development and behavior; and understanding of the processes involved in disease, including disease onset, recovery, response to treatment, natural history of the aging process, and identification of factors that predict age-related outcomes. Researchers in aging seek to unravel the impact and interaction of physical and psychological processes on human development, health, and disease. From the point of view of statistical analysis, the critical aspect of data obtained from longitudinal studies is the inherent correlational structure of multiple measurements made on a single subject or other experimental unit, which must be appropriately treated in the analysis of the data. We discuss a series of nonparametric approaches that are both analytically accessible and particularly well suited to the analysis of sparse or otherwise limited longitudinal data.
Duke Scholars
Published In
DOI
ISSN
Publication Date
Volume
Issue
Start / End Page
Location
Related Subject Headings
- Statistics as Topic
- Longitudinal Studies
- Linear Models
- Humans
- Experimental Psychology
- 5204 Cognitive and computational psychology
- 5203 Clinical and health psychology
- 5202 Biological psychology
- 1701 Psychology
- 1117 Public Health and Health Services
Citation
Published In
DOI
ISSN
Publication Date
Volume
Issue
Start / End Page
Location
Related Subject Headings
- Statistics as Topic
- Longitudinal Studies
- Linear Models
- Humans
- Experimental Psychology
- 5204 Cognitive and computational psychology
- 5203 Clinical and health psychology
- 5202 Biological psychology
- 1701 Psychology
- 1117 Public Health and Health Services