Abstract
Estimating residential water demand in developing countries is complicated by the unique nature of water supply, characterised by unequal access and multiple water sources. Using cross-sectional data obtained from a survey of 1,300 households, the determinants of residential water demand were predicted using a multiple linear regression model. The determinants include access to water, household size, trip number, monthly income, payment for water, educational qualification, trip time and house type. The determinants predicted daily water consumption with an F(9, 1,014) = 81.063, P < 0.05, R2 = 0.450. Daily water consumption was found to be 2.8 times more per household and 4.4 times more per capita for those with on-site access, compared to those with off-site access. Moreover, consumption was influenced by the various water source categories.