Analyzing by two factors is often a first step in investigating whether an association is confounded or likely to be causal. Understanding the interrelated causes of malnutrition is important, considering the complexity of determinants that lead to malnutrition of children. Unless the various components that contribute to malnutrition are identified and understood in relation to one another, then an appropriate multi-faceted and multi-sectoral approach to alleviate the problem cannot be made.

Two-way analysis begins to link the determinants of malnutrition and explore potential causal relationships. This leads to the information needed to gear an intervention towards the factors most likely to have an impact, or in real terms to prevent malnutrition of children. Regression provides a very powerful tool for investigating potentially causal relationships, and will be introduced as a part of the two-way analysis. Initially, viewing tables offers a more intuitive way to look at the relationships of interest, and the information displayed can provide a check on the numerical regression output. Presentation of results will also utilize these tables as clear and simple venues of communicating results that have been confirmed through regression analysis.


Page 1
General Approach for Program Content using two variables and creating 2-way tables for presentation
Page 2  
Interaction of the relationship between the outcome and determinants – how to detect different levels of effect in different stratum (age groups, for example)
Page 3  
Regression Analysis with two independent variables – how they associate with the outcome
Page 4
Confounding of the relationship between the outcome and selected determinants– how to detect confounders and control for their effects in a regression analysis
Page 5    Test Yourself