Related Experiment Videos
Mixed models incorporating intra-familial correlation through spatial autoregression.
George J Knafl1, Kathleen A Knafl, Ruth McCorkle
1Yale University, School of Nursing, 100 Church Street South, New Haven, CT 06536-0740, USA.
Research in Nursing & Health
|July 20, 2005
Summary
Statistical modeling for family data can be complex due to interdependencies. This study adapts spatial autoregressive methods to model intra-familial correlation, accounting for relationships between family members and time.
Area of Science:
- Family research
- Statistical modeling
- Quantitative psychology
Background:
- Family researchers face challenges in analyzing statistical dependence within family data.
- Existing methods may not fully capture the complex correlations among family members.
Purpose of the Study:
- To adapt mixed modeling methods for analyzing continuous outcomes in family data.
- To introduce a spatial autoregressive approach for modeling intra-familial correlation.
Main Methods:
- Utilized mixed modeling techniques.
- Applied a spatial autoregressive approach to account for directional and distance-based dependencies.
- Incorporated dimensions for family members and temporal correlation in longitudinal data.
- Included general linear models for fixed effects.
Main Results:
- The spatial autoregressive approach effectively models intra-familial correlation.
- Demonstrated the ability to account for varying correlations between different family members.
- Showcased the integration of temporal correlation for longitudinal family data.
Conclusions:
- The spatial autoregressive approach provides a robust framework for analyzing complex family data.
- This method enhances the understanding of statistical dependence in family research.
- Offers a valuable tool for researchers studying intra-familial relationships and outcomes.