Survey Data

The survey panel used in this analysis consists of data collected in four waves including a stratified random sample divided between 196 households surveyed in 1996, 196 surveyed in 2000, 406 surveyed in 2009, and 646 households surveyed in 2009. The sampling methodology and survey design were consistent between each of the waves. The sample size was expanded in 2005 (and therefore increased the size of the survey sample) for two reasons: 1) to include new settlements that had been established in the study region (on previous forest reserves or large ranches that were expropriated) and 2) to maintain a representative sample population. Also initiated in 2005, panels were maintained by property (owners of same lots always interviewed) and by family (same families always interviewed). Thus, households and individual tracking expanded the survey region in this year to include the original six municipalities in addition to surrounding municipalities.

The survey data provide (i) full information on farm production outputs and purchased inputs, (ii) hectares reported in different land uses, including forest, pasture, annual crops, and agroforestry and perennial crops, (iii) measures of wealth, including consumer durables, equipment, livestock, vehicles and reported value of parcels, and (d) a standard set of socio-economic characteristics. Data indicate that the rural population of small farm families has remained fairly stable over this study period (1996-2009), exhibiting a relatively low 5% annual attrition rate.

Data on land use change and socioeconomic dynamics in developing countries are often collected via paper and pencil interviewing (PAPI). In 2009 a computer aided personal interviewing (CAPI) methodology adopted for the fourth wave of a panel survey administered in our survey region. This survey instrument included a core set of questions included in previous rounds in addition to questions on consumption, new durable goods found in the region, among other questions on health. Ruggedized touchscreen laptops were used to address challenges associated with survey administration in this setting along with limitations associated with the PAPI method. In the paper “Improving Household Surveys Through Computer Assisted Data Collection: Use of Touchscreen Laptops in Challenging Environments", we discuss hardware and software considerations, methodological innovations, and test for mode effects on missing item response rates and enumerator learning effects.