Index of integration: toward a summary measure of a multidimensional concept

Fernando Rajulton, University of Western Ontario

Integration is social cohesion viewed at the individual level. Like social cohesion, integration is a multi-dimensional concept with dimensions in economic, socio-cultural, and political domains. We propose a method to derive an index of integration from the dimensions’ several indicators that are analyzed to identify a relatively small number of unobserved “factors” or “latent variables” representing relationships among the variables. We first use an exploratory factor analysis to identify latent constructs underlying the observed indicators. We then use structural equation model (SEM) to confirm the existence of theoretically established latent constructs. The final step towards deriving an index is to transform the latent scores from the structural equations models into one and same metric such that they all have a common distribution, and assigning weights for each of the domains. The method is applied to data gathered through the nation-wide Ethnic Diversity Survey conducted by Statistics Canada in 2002.