This paper describes the data collection methods used in a longitudina
l study of the coping strategies of 27 households in three villages in
the study area of the MRC/ODA Research Programme on AIDS in Uganda. A
fter pre-testing and piloting, 9 local interviewers made regular visit
s to the 27 study households over a period of just over one year. The
households were purposively selected to represent different household
types and socioeconomic status categories. Data were obtained through
participant observation using a checklist to ensure systematic collect
ion of data on household activities. Debriefing sessions with the inte
rviewers after the visits provided opportunities for the discussion of
the findings and exploration of themes for further study. On the basi
s of the study findings, and data from the Programme's general study p
opulation survey rounds, broad indicators of household 'vulnerability'
were identified. A participatory appraisal technique, 'well-being ran
king', was used at the end of the study in order to test the viability
of the chosen indicators. It is proposed that the example of the rese
arch method, which relied on local people not only as interviewers but
also as co-investigators in the research, be used to guide future res
earch approaches. The participation of the study community at every st
age of research and design, as well as monitoring and evaluation of su
pportive interventions, is strongly encouraged.