EQUITABLE ACCESS TO WATER IN A RURAL COMMUNITY IN KENYA

Abstract

Water, a fundamental human right, impacts human health through its quantity (i.e., physical amount and ability to access it) and quality. Consumption of poor-quality water can lead to a variety of waterborne illnesses, often manifested as diarrhoea. Millions of individuals worldwide lack access to drinking water that is free from contaminants and is available and accessible when needed. In areas where water is not piped to homes, several physical, demographic, socio-economic and health factors affect access to potable water. These factors may also influence which water point an individual fetches water (i.e. their waterpoint choice) from in the presence of multiple alternative waterpoints. Through this study, effects of various physical, health, demographic and socio-economic factors on waterpoint choice were explored. This study, based on datasets from a rural Maasai community in Kenya, implements a multinomial logit model to explore effects of various physical (travel time and water quality), health (aggregate frequency of self-reported diarrhoea stratified by age groups), demographic (average household age, household population, number of children under 5, number of women between 8-45 years of age and ratio of household population to number of women between 8-45) and socio-economic factors (education and income) on waterpoint choice. Travel time to the most probable waterpoint as predicted by the model was compared with the travel time to a household’s chosen waterpoint. Both travel times were calculated using the least-resistance path function incorporating slope and landcover. Results from model optimization showed that combinations of travel time, average household age, diarrhoea among adult women, income, education and number of women between 8-45 years were significant contributors to the three waterpoint choice models. The expected travel time to the most probable waterpoint predicted by these models and actual travel time to chosen waterpoint fit well, showing that the models explain waterpoint choice well.