Income, or its surrogate, has been an important predictor of residential water use. In the case of Israel, other socioeconomic variables not collinear with income have been found to be valid predictors and increase the sensitivity of residential forecasting models. For a sample of 1892 residences in Israel, aside from income those variables found to explain a significant portion of the variation in residential water use were number of persons per family, cultural origin, education and age of the head of the household, and number of rooms per household. The field of residential,water demand forecasting is replete with studies estimating the elasticity of price. A cursory review of the literature can be found in an article by Wong [1972]. However, only a few of these studies have examined socioeconomic characteristics as predictors of water consumption. The present study describes the use of income and number of persons per family as well as other variables from disaggregated data obtained in a sample survey held in Israel. It is hoped that projected changes of the distribution of demographic parameters and their characteristics can be associated with expected changes in residential consumption, accurate forecasting thereby being permitted. For example, income elasticity figures of 0.31-0.37 were found in a sample of 35 study areas throughout the United States [Howe and Linaweaver, 1967], 0.56 in the Toronto metropolitan area [Grima, 1972], 0.20-1'.03 in Chicago [Wong, 1972], and 0.28-0.58 in Kansas [Gottlieb, 1963]. All of these studies used either highly aggregated data or property values as income surrogates. When the data were disaggregated, Morgan [1973] and Grima [1972] derived another socioeconomic variable which explained large variations in residential water use, number of persons per family. SAMPLING DESIGN A questionnaire survey of 1892 households in four urban areas of Israel was conducted from October to December 1971, and water use data were obtained for the 1970-1971fiscal year. The objective of the survey was to obtain water consumption data for various segments of the population. Data were collected concerning the extent of water-using iristallations in the household, the frequency of their use, the degree to which families try to limit their water use, the perception of water as a commodity, and the household's willingness to pay more for water. Water consumption data, including water for gardening as well as for within-house consumption, were obtained for each household from the various municipalities.Although the family was the basic unit of inquiry, the head of the household, or the spouse, was not always the respond-Copyright ¸ 1975 by the American Geophysical Union. ent. Only persons over 20 years of age were interviewed, one person per household qualifying as a respondent. A household is defined as a married person with a family or an unmarried person at least 29 years of age. The subjective components of the questionnaire were representative of that individual's response, but the objective ...