Quantitative analyses for resilience measurement

Published in August 2016 by the Resilience Measurement Technical Working Group (RM-TWG) of the Food Security Information Network, the briefing Quantitative analyses for resilience measurement seeks to provide guidance on how to carry out quantitative analysis for resilience measurement. The analytical aims of resilience measurement have been presented as two frameworks: the first focusing on the procedures used to construct resilience variables, and the second addressing the procedures used to examine relationships in which resilience may be a predictor variable. The importance of defining resilience clearly was emphasized.

To develop these ideas, resilience was considered as a multidimensional variable described as a function of a number of dimensions that express different aspects of resilience. Unlike temperature or rainfall, resilience does not have direct physical indicators or a straightforward corresponding count. Like many things measured in development, it cannot be directly observed and measured. However, resilience can be represented as clusters of indicators and each one can be examined separately. If the dimensions of resilience are aggregated into a unique index, factor analysis and structural equation models are the recommended approaches. If resilience is employed as a regressor in a well-being model, then endogeneity and multicollinearity problems should be addressed. To address nonlinear relations between resilience and its determinants or context-specific relationships, more sophisticated econometrics exist, but these are more computationally complex and not easily implementable in the field.