The data used in this study were obtained from a survey of
325 farmers randomly selected from 12 communities located in
two regions (Oaxaca and Chiapas) of Mexico. The selection of
the 12 communities from the two regions was not random, but rather systematic, aimed at sampling communities with households of different levels of the adoption of improved germplasm, different levels of poverty status and ethnic
backgrounds, as well as different land holdings. The selection
was aided by CIMMYT’s geographic information system
(GIS) Lab. The GIS was developed to define potential locations
for field work. In each community, 27 households were
randomly drawn for the interviews to ensure that the 12 communities
were equally represented.
The survey was conducted
between October and December in 2001. Information from the selected households was gathered by a survey of farm households.
The questionnaire contained questions about the kind of maize varieties that are planted by farmers, a key variable in the analysis.