Microcredit and Health Services Experiment in Bangladesh

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The project is a 4-celled factorial experiment designed to test the relative effects of separately and jointly introducing micro-credit (MC) and essential (health) services package (ESP) interventions on women's empowerment, economic well-being of the family and use of preventive health services in groups of villages within 64 unions (the political unit above a village and composed of 10-30 villages) in central Bangladesh. Baseline and follow-up surveys of samples of 3900 households (over-sampling of poor households) with about the same number of ever-married women will be interviewed in the 4 areas and the interventions will be carried out in the 2.7 interim years. The interventions will be implemented by partner agencies-Grameen Bank, Gonoshastra Kendra and Association for Social Advancement-which will expand activities in central Bangladesh and accept random assignment of areas. In each selected union, a cluster of villages without such programs will be randomly allocated to 1 of the 3 intervention arms or a control arm (with usual government services only). The main outcome variables of interest are women's empowerment, economic well-being of the family and use of specific health and family planning services. The follow-up survey is planned to be mostly in the same clusters as the baseline so the same women can be interviewed, allowing longitudinal analyses. A fraction of new clusters in each study arm will be sampled as well to permit assessment of a potential "testing" effect of the baseline survey. Hypotheses to test include: a) Improvement in empowerment indicators will be greater in experimental areas with MC than in areas without MC; b) The improvement in health, family planning and economic indicators will be greater in all experimental areas than in the control area; c) The combined or integrated MC and ESP programs will have greater effects on contraceptive use, health care utilization, and child health care than those of the simple addition of the separate programs. Simple statistics comparing before-after indicators by area as well as multi-level statistical models will be used to test the hypotheses. In addition, econometric methods will be utilized to estimate intervention effects at the individual level adjusting for self-selection. This research represents a joint effort of investigators at Johns Hopkins University Morgan State University (a historically black university), and Brown University, the partnership organizations in Bangladesh, and a consultant with expertise in women's empowerment.