The main question addressed in the present analysis is how education level of the
mother and child-care affects child nutritional status. The two-way ANOVA
approach can determine if there are overall differences in weight for height
Z-scores (ZWH) between different educational levels of the mother, between
varying levels of child-care and whether there is an interaction effect of educational
level and child-care on improving child nutritional status (Karpinski,
2003). The interaction effect can be thought of as saying that the effect of one
factor (e.g. educational level) depends on the level of the second factor (e.g.
child-care). For example, it may be the case that higher educated women provide
better child-care than lower educated women. Thus, we undertake the following
hypothesis tests:
Educational level effect:
H0: mean weight for height Z-scores do not differ by educational levels of the mother
H1: mean weight for height Z-scores differ by educational levels of the mother
Care effect:
H0: mean weight for height Z-scores do not differ by care levels by the mother
H1: mean weight for height Z-scores differ by care levels by the mother
Interaction effect:
H0: there is no interaction between educational levels and care levels
H1: there is an interaction between educational levels and care levels.