This paper summarizes the sampling procedure, survey instrument, implementation and
selected results of a comprehensive survey among 4,400 rural households conducted in 2007
in six peripheral provinces of Thailand and Vietnam as an initial wave of a panel dataset. The
sampling procedure applied was a three-stage cluster sampling with explicit strata for agroecological
zones in Vietnam and implicit stratification in Thailand. We find an atypical age
distribution of the rural population as a result of out-migration of the economically active
population, which is more pronounced in Thailand than in Vietnam. The educational
attainment of the adult population is rather low in both countries, but generally higher in
Vietnam. With respect to income and poverty, the accuracy of the survey estimates is similar
or better than for the national surveys in these countries and means are roughly comparable.
Per capita income varies stronger in Thailand than in Vietnam and consumption varies much
less than income in both countries. The provincial sub-samples differ also with respect to the
main sources of income and the importance and intensity of agriculture. The experience of
past shocks and the perception of future risks reveal that agricultural and demographic risks
are ranked as the most important threats to household welfare. We conclude that the rich data
provide a sufficiently accurate basis and a number of domains for comparison through the
coverage of distinct economic and ecological conditions. Moreover, the household
composition suggests that for further panel waves an inclusion of household members who
left for the cities should be included to adequately consider their contribution to the welfare
and risk profile of the rural households.