An XLSForm method for fast offline collection of full-community social network data

The collection of social network data from small-scale communities has been an increasingly important part of long-form research projects on resilience, adaptation, and well-being. Various tools are available to facilitate such data collection protocols. Roster-based methods yield high quality data, but typically scale poorly with sample size. Name-generator-based methods, in contrast, scale well with sample size; however, they entail a large amount of highly error-prone post-processing after data collection in order to link elicited names to unique identifiers. Here, we introduce a hybrid XLSForm method for collection of social network data using name-generator prompts that are dynamically linked to a full-community photograph roster. The name-generator prompts result in fast data collection in situ, and a confirmation step using photographs certifies that elicited names refer to the correct individuals. We provide both an example XLSForm template, and an R package, XLSFormulatoR, that facilitates the automatic creation of larger multi-question XLSForm social network surveys. We present a complete methodological workflow using KoboCollect to implement our survey design. To validate our methodology empirically, we conducted a network study in two rural Beninese communities (n=614): we compare the friendship networks elicited using our app with friendship networks elicited from the same individuals using a roster-based design.