3 is a random error term that reflects heterogeneity in individual taste, health
endowment and other unobserved factors.
The outcome variables for our analysis are weight for age (ZWA) and height
for age (ZHA). These are computed in terms of Z-scores using the NCHS/WHO
reference values. The Z-score is defined as the difference between the value for
an individual (weight or height) and the median value of the reference
population for the same age or height, and divided by the standard deviation of
the reference population. A low height for age (often referred to as stunting)
results from long-termpoor health and nutrition and is often regarded as a stable
indicator of child nutritional status, since it is not subject to short-term fluctuations.
Stunting in infancy and early childhood often has adverse consequences
for later life, such as reduced productivity and increased reproductive and
maternal health risks (WHO, 1995).
On the other hand, low weight for age is influenced both by the height of the
child (height for age) and weight (weight for height). Its composite nature makes
the interpretation complex. One of the drawbacks of this measure is it fails to distinguish between short children of adequate body weight and tall thin
children. This indicator can also reflect the long-term nutritional experience of
the population. The XIC vector consists of age of the child in months
(AGEMNTH), and the squared term of child age (AGESQ). We include this
quadratic term since there was a curvilinear relationship (with bivariate plots)
between age of the child and nutritional status as measured by both height for
age and weight for age Z-scores. The rationale for such a relationship stems from
the fact that younger children are more vulnerable to various diseases such as
diarrhea, malaria, etc., and thus are more prone to malnutrition than older
children.