Depriving children of the nutrients needed for growth sets them up to fail in life.
When children are well nourished and cared for, they are more likely to survive, thrive,
and to meaningfully contribute to society.
This study assesses the association of characteristics of individual children under
age five in Ghana, their mothers, and their households—as well as socioeconomic and
environmental characteristics of the places where they live—with differential nutritional
well-being. What distinguishes this study from most research on young children’s
nutritional status in the Global South is its analysis of data for individual children, made
possible by use of Demographic and Health Surveys (DHS), and assignation of district
variables that capture characteristics of their places of residence to individual children as
cases. This enables assessment of the relative explanatory role of variables that describe
the socioeconomic and biophysical environments.
This study implements a three-level multivariate logistic regression analysis with
separate models for each of the nutritional outcome variables—height-for-age, weightfor-
age and hemoglobin—at each level. Descriptive statistics summarize the prevalence
of stunting, underweight, and hemoglobin and delineate frequencies and proportions for
selected independent variables at each level. Further statistical analysis relies on chisquared
(χ2) tests to determine significant bivariate associations. All significantly
associated variables in the bivariate analysis are subjected to binary logistic regression analysis. The results of fixed effects are reported with odds ratios (ORs) along with
confidence intervals for p<.05.
The following variables were found to be significantly associated with at least one
of the three nutritional outcomes in multivariate analyses at the child and district levels:
child’s age, months of breastfeeding, fever, mother’s health status, prenatal care,
mother’s occupation, mother’s ethnicity, household water supply, household wealth
status, population density, percent literate (vs. illiterate) in district, percent in rural (vs.
urban) locations, wealth status of district residents, and ecological zone of residence. As
found in much previous research, mother’s education and occupation, father’s education and occupation, household size and structure, and sanitation were significantly associated
with children’s nutritional status in bivariate analysis but not in multivariate analysis.
After controlling for the characteristics of children, mothers and households,
significant associastions with children’s nutritional status were found for population
density, percentage of literate (vs. illiterate) residents in a district, wealth status of district
residents, and residence in the Guinea Forest-Savanna Mosaic and Central African
Mangrove ecological zones. Other significantly associated variables in the final models
were the age of the child, months of breastfeeding, whether the child’s mother has health
insurance and the wealth status of a child’s household. Notwithstanding the shortcomings
of this study, its findings can potentially assist stakeholders by providing a better
understanding of the diverse set of factors that influence children’s nutritional status and some explanation for differences in nutritional status among places within Ghana.
University of Minnesota Ph.D. dissertation. August 2011. Major: Geography. Advisor: Connie Weil. 1 computer file (PDF); viii, 187 pages.
Nikoi, Ebenezer Goodman Ashie.
Child nutritional well-being in Ghana: an analysis of associated individual, household, and contextual health indicators and socioeconomic and biophysical environmental variables..
Retrieved from the University of Minnesota Digital Conservancy,
Content distributed via the University of Minnesota's Digital Conservancy may be subject to additional license and use restrictions applied by the depositor.