The plot size plays a significant role in the accuracy of plot area measurement using the various methods, the validation sample was stratified on four plot size strata to ensure we could test the various methods on larger plots, which are much more rare. Four states were purposefully selected based on safety and past performance in area measurement (Benue, Osun, Oyo, and Kogi). Using the second wave of the GHS panel as the sample frame and the GPS measurement of the plot taken in the post-planting visit, every plot was assigned to some plot-size strata (strata 1: <=1000 sq. meters; strata 2: 1000-2500 sq. meters; strata 3: 2500-5000sq. meters; strata 4: >5000 sq. meters). One hundred plots were then randomly selected from each stratum. This process yielded the selection of 400 plots (211 households). However, in order to maximize the sample at minimal added cost, we included all plots from the selected households, not only the plots that were selected in the first step (totaling 518 plots).
From the 518 selected plots, 23 plots were unable to be measured (5 due to land disputes, 4 due to respondent refusal, 14 for other reasons). Therefore, the total number of plots measured and included in the farm area measurement validation study is 495, coming from a total of 202 households.
Stratification by plot size in the validation sample results in the unequal probability of plot selection within households from the GHS-Wave 2 sample. Household-level sampling weights were calculated for the validation sample to make them representative of the same household population sampled in Wave 2. Refer to Annex I of the Basic Information Document for details on the construction of the sampling weights.
