Center for Development Research (ZEF)1, University of Ghana, Institute of Statistical, Social and Economic Research (ISSER) 2
Abstract Introduction Over the past decades (more especially since 2000), access to improved water supply and sanitation services in developing countries (including Ghana) have been widely discussed in the media, and in several conferences, meetings, workshops and seminars. This situation has assumed global prominence due to the disproportionate achievement of water and sanitation-related Millennium Development Goals (MDGs) between developed and low resource countries, and also increasing morbidity and mortality as a result of poor water and sanitation environment. Availability of quality water is essential for the general well-being of every human society. This is based on one of the most popular adage on water such as "water is life". In Ghana two types of water exist; (1) drinking water and (2) water for general use (such as cooking, washing, bathing, among others). There is the need for additional studies in understanding household decisions and choices in improved water supply and sanitation services; both as a separate and joint-decision making options. This study aims at estimating the determinants of household choice and use of water supply and sanitation services. A significant addition to the body of WASH literature by this study is the use of several estimating methods (logistics regression, bivariate probit, ordered probit and multinomial logistics) to estimate the determinants of household choice and use of water supply and sanitation services in Ghana. The analysis for the study is expanded to capture the nexus between water supply and sanitation services as single decision making unit at the household level. Methods/Materials To obtain information on the current state of water supply, sanitation and hygiene in the Greater Accra region of Ghana, a household baseline collection exercise was carried-out in April-May, 2014. Two districts (Ga South Municipal and Shai-Osudoku district) were purposively selected due to the presence of multiple water use systems and deficiency in use of improved water supply and sanitation services. Here communities along the coast, Volta river and Densu catchment area were of interest. An institutional survey (for public basic schools and WATSAN committees) was conducted in December 2013. The institutional survey yielded interviews with 36 WATSAN (water and sanitation management) committees and 48 public basic schools (based on school management) and this forms the basis of the initial sample frame. Public basic schools located in communities with multipurpose water systems and use of unimproved water and sanitation were then selected to form another stratum in the sampling design. School children in grade five to eight were randomly selected to represent the households to be surveyed. Household tracking in March 2014 identified all randomly selected school children and their respective households. Selected siblings were replaced with students with similar sex and grade from the same public basic school in order to obtain "unique" households. Consent and assent were obtained from participating households and school children during the tracking/listing exercise. A total of 505 households in about 16 communities were interviewed during the household baseline data collection. The 505 households are made up of a total of 3,072 household members/individuals. Household heads in the randomly selected households were interviewed by the use of the structured questionnaires. Other sections/modules were responded by the most knowledgeable person (for instance, spouse) in the household. The structured interviews involved several modules on water supply, sanitation and hygiene; household health; household expenditures and productive assets; interviewer observations, among others. Detailed information on households and their members form the basis for this study. Logit, multinomial logit, bivariate probit, and ordered probit models are used to estimate the choice probabilities across improved water supply and/or improved sanitation options. The dependent variables are based on WHO and UNICEF's Joint Monitoring Program indicators/measurements (including "drinking-water ladder" and "sanitation ladder"). Results and Discussion Based on JMP classifications about 72.5 percent of the households use improved main drinking-water sources whilst the remaining 27.5 percent rely on unimproved water sources. The results is lower as compared to the overall average of Ghana's improved water source usage of 87 percent and that of the rural communities/areas of 81 percent in 2012 (WHO and UNICEF, 2014). Taking into consideration the secondary drinking-water sources, about 67.7 percent of the households use improved sources whilst the 32.3 percent of the households rely on unimproved sources. In the case of main water supply for general purposes, about 59.3 percent of the households rely on improved sources whilst the remainder of 40.7 percent use unimproved water sources. Secondary water sources for general purposes come largely from unimproved sources (71.8 percent) whilst only 28.2 percent of the households rely on improved water sources. The main conclusion is that percentage of use of improved water sources by the households decrease along/across the different water supply options with the highest percentage being main drinking-water supply and the least percentage being secondary water supply for general purposes. In case of sanitation, about 43.8 percent of the households use improved sanitation facilities. This comprises of 14.8 percent of "improved not shared" sanitation facilities and 29 percent of "improved shared" sanitation facilities. The majority of the households (56.2 percent) rely on unimproved sanitation facilities. The result is higher than the overall national average of improved sanitation facility coverage of 14 percent and that of rural areas of eight percent in 2012 (WHO and UNICEF, 2014). The econometric analysis showed that household water supply and sanitation choices are affected by several socioeconomic factors. In general, the results show that age of household head, literacy status of household head, household wealth index (proxy as household expenditure), among other factors affect the choice of water supply and sanitation services both as a separate and joint decision making options. Conclusion The results show significant differences in choices of improved water supply and sanitation services. Moreover, the choice and use of improved water supply decreased consistently from main-drinking water to secondary general purpose-water sources. The econometric estimations indicate several socio-economic factors influence the choice and use of improved water supply and sanitation services; both as a joint and separate decision making options in Greater Accra region of Ghana. 1.WHO and UNICEF (2014). Progress on Drinking Water and Sanitation: 2014 Update. WHO, Geneva, Switzerland. pp.1-75.