In earlier chapters we have gone quite some way in determining what associations may exist in a cross-sectional dataset.   Multi-way analysis as used in this manner continues on this path by examining the association of one dependent variable with a set of independent, determining or classifying variables.  The benefit in using multi-way analysis in addition to the other techniques, is that multiple variable linear regression is a more powerful tool for seeking out true associations when many independent variables are considered. Multi-way analysis is used to get to the heart of the research questions, to explore both causality and to create prediction models. The goal in analyzing nutrition data in this context is to contribute to policy and program design, analysis, and evaluation.  


Contents:

Page 1      When and Why?

Page 2      Causal Models to Address Confounding

Page 3       Causal Models to Address Interactions

Page 4         Test Yourself