Is audio computer-assisted self-interviewing a feasible method of surveying in Zimbabwe?

Background. Research into reproductive health is dependent on participants accurately reporting sensitive behaviours. We examined whether audio comput er-assisted self-interviewing (ACASI), which increased sensitive behaviour reporting in the US, is a feasible method of surveying in developing countr ies. Methods. Zimbabwean women in three educational groups were surveyed about d emographics and family planning using interviewer and ACASI modes. An exit survey was administered to elicit information about the participants' opini ons and experiences using ACASI. Results. The majority of women (86%) preferred ACASI to interviewer mode. T he reasons mentioned were always related to increased confidentiality and p rivacy. Ability to use ACASI and user preferences varied with educational l evel. More women with primary school or less education (53%) reported probl ems with computer use than women in the higher educational groups (10-12%). The percentage of women having perfect response concordance between ACASI and interviewer modes increased significantly with education (64%, 81%, and 84% respectively; P-trend <0.001). Conclusions. Use of ACASI may be more feasible in Zimbabwe and other develo ping countries than was originally thought, but ACASI programs should conti nue to be improved and tested in various countries and population groups.