● Collecting and collating all vari

● Collecting and collating all variables from the programme baseline datasets. If the variable appeared in the project’s baseline data, the variable was marked with a “1” in the project’s respective column.● Organizing variables into thematic areas and subthemes. This was a subjective exercise that involved organizing the variables based on themes that emerged organically, and further organizing the themes into subthemes. For example, the survey question “How far is your farm from the nearest output market?” was grouped under the theme “Access to Services” and the subtheme “Accessibility”.● Flagging variables that were similar, but with slight variations. If two variables are measuring the same thing but are not immediately ready for aggregation because of standardization issues, a “0.5” tag was used instead of “1”. For example, the variables “crop calendar for crop X” and “month crop X was planted” were assigned 0.5 tags.● Highlighting variables that were common or nearly common across multiple projects. The total number of similarities for each variable was calculated by tallying the number of 1s and 0.5s tagged across the six projects. This gave a range of “total