Multiscale Sociospatial Inequality Computations (MUSIC)

There is a long-standing interest within the social and economic sciences about the effects of the sociospatial environment in which people live on their individual outcomes, such as educational achievement, employment, income, and health. One of the main challenges in research on sociospatial effects is the measurement of the sociospatial context of people. Traditionally, this spatial context is measured at a single point in time, using one variable (such as the percentage of people in poverty) and at a single spatial scale (often a neighbourhood represented by an administrative unit). It is increasingly acknowledged that spatial context measures should be longitudinal, multidimensional and multiscale. The Netherlands has excellent and unique register data, which allows the creation of such spatial context variables, and their application in models in order to estimate how they impact individual outcomes. We have developed an approach for the creation of multiscale measures of population and successfully piloted the application of such measures and the assessment of spatial context effects using individual level register data from Statistics Netherlands (CBS). However, the creation of two simple variables alone took four months of continuous computation time on the CBS heavy server. As a result of the hardware limitations at the CBS, we were invited as one of three social science teams to use Cartesius as part of the ODISSEI Secure Supercomputer (OSSC) pilot project. After our successful pilot with Cartesius, we would like to expand our research and take the next crucial steps. We want to create a series longitudinal, multidimensional and multiscale measures of population, based on people’s sociodemographic characteristics, including their socioeconomic status, age, ethnicity and education. These detailed multiscale measures will enable us to answer here interrelated scientific research questions: How does the spatial structure of the social environment impact individuals and their educational, employment and health outcomes? In doing so, we will also be able to answer how segregation has developed over a long period of time in the Netherlands. Our third question of interest is how different neighbourhood characteristics have developed over time. Ultimately, we will examine the consequences of these processes for cities and for individuals.