Don’t throw the baby out with the water! Echos of human
spatial and information behavior in exhaust data, Aalto University, 13.9.2024
As spatial research is increasingly moving to analytics of large datasets that have not been directly collected for the exact purpose for which they are used, possibly generated as by-products (exhaust data), strong attention has been paid to fitness for use and data quality. A common step is often to clean datasets to retain only data that meet stringent requirements (for instance, for completeness). In this talk, I will focus on demonstrating the usefulness of data that may not always meet all of our stringent requirements, as is often the case in exhaust data, if approached from the deliberate, careful perspective of explicitly stating hypotheses about the data generation process. I will demonstrate how data problems led us to deeper thinking about the patterns inherent in datasets, and through this, to insights about human spatial and information behavior, in a range of studies about human mobility, population dynamics, and VGI, and thus inform the future research program of my group into urban networks and spatial causal inference.