Modeling and mapping of girls’ female genital mutilation/cutting (FGM/C) in the context of economic, social, and regional disparities: Kenya demographic and health surveys 1998-2014

Document Type



School of Nursing and Midwifery, East Africa


A significant body of academic research has focused on the development of theoretical models of behaviour change that may afford insights into the dynamics of changes in FGM/C. These models fall broadly under two main paradigms: 1) individual-centred decision-theoretic models such as modernisation theory and 2) relational models that account for the influence of social norms within a reference community. FGM/C as a practice potentially maintained by both individual factors and collective social norms rests upon several assumptions that can be examined empirically with the increasingly rich body of survey data now available. Recent analyses of nationally representative survey data have, in several instances, been designed to investigate the effects of community norms associated with FGM/C, not of individual and household characteristics, but have been hampered by methodological limitations. The sampling strategy of DHS and MICS surveys lends itself to multilevel modelling techniques for simultaneous estimation of effects specific to individual respondents as well as those common to communities. In the first stage of sampling, clusters are selected from enumeration areas. In the second stage, a random sample of households is selected, and all women ages 15 to 49 are asked to respond to the survey. Characteristics common to a cluster, or in some instances larger geographic areas such as regions, estimate effects of the community, while simultaneously estimating individual and household effects.


This work was published before the author joined Aga Khan University.


Reproductive Health