Evaluation report on the contraceptive prevalence studies project

Special evaluation covers the period 9/79-7/83 and is based on document review, site visits, and interviews with Westinghouse Health Systems (WHS) contractor personnel and relevant experts. Funding uncertainties and political considerations in country selection (which led to the project"s concentration in small Caribbean nations) have put the project behind schedule, despite WHS"s impressive effort in recruiting and helping sub-contractors and in enlisting the participation of interested countries. By the project"s 9/84 expiration, 32 of 60 planned surveys (24 first-round and 6 second-round) will likely be completed in 30 LDC"s. Emphasis has been placed on first round rather than repetitive surveys; first and second round surveys have been conducted in only three countries. To date, eight country reports have been completed; an extension may be needed to complete the rest. Regarding survey documentation targets, four adaptable core documents - the core questionnaire, interviewer"s and supervisor"s manuals, and the coding/editing manual - have been widely published, and two supplemental modules - for male prevalence (in Barbados) and for abortion (in Nicaragua) - were added. Refinement and standardization could make the documentation more useful. Field operations have generally gone well and been sufficiently monitored to ensure a scientific sample, although careful monitoring of field operations is urged. A strong point has been overseas TA and training in data processing provided by WHS. Regional seminars and country workshops have proven effective means of disseminating project results, but there is a need to expand mailing operations and develop an ongoing feedback mechanism. The contract"s objectives of producing numerous baseline prevalence estimates notably hindered efforts to provide the data analysis sought by host country researchers. Key recommendations for a follow-on project are to broaden country representation, further analyze results, stress the improvement of LDC survey capabilities, and ensure sampling accuracy.