Modeling village water demand behavior: A discrete choice approach

Abstract

This study presents a discrete choice model of households' water source choice decisions in developing countries. This model is estimated with data collected by in‐depth personal interviews with 69 households in Ukunda, Kenya, a small town south of Mombasa. The results suggest that households' source choice decisions are influenced by the time it takes to collect water from different sources, the price of water, and the number of women in a household. Household income, however, did not have a statistically significant effect. Essentially the same data were used to estimate a traditional water demand model which attempts to explain the quantity of water demanded by a household as a function of collection time, income, and other socioeconomic variables. The results of the discrete choice and traditional water demand models are compared in this paper.