Main argument
Satellites are a game changer but require careful thinking about how we believe place matters and this has implications for both theory and data.*
Center for Geographic Analysis | Harvard University
2026-05-20
I’m not here on behalf of the anti-pixel lobby
Satellites are a game changer but require careful thinking about how we believe place matters and this has implications for both theory and data.*
Why is this place more developed, happier, healthier than that one?
Which places are most vulnerable to effects of economic crisis or climate variability?
Why do people move from this place to that one?
How is this place changing over time?
How do place conditions affect outcomes…for people, businesses, policies?
What hazards are individuals exposed to as they move through these places?
Different applications demand different conceptualizations
Has political, economic, and/or psychological valence. Often has a name.
Could be formal and administratively defined, like a statistical unit, or could be perceptual, like your neighborhood.
Boundaries matter.
The unit upon which satellite images (and photos) are based. Resolution and boundaries are determined by a machine in space. Boundaries are arbitrary and meaningless.
Spatial influence on some outcome of interest (like education or health). Could be a place!
(probably isn’t a pixel)
Research should be grounded in a conceptualization of place that suits our purposes.
Then we need data for these geographies.
Project website: https://imago.ac.uk
rachel_franklin@cga.harvard.edu
https://www.rachelfranklin.org
Thank you!!
Geospatial Data and Quantitative Analysis in Social Science | GSSI | 20 May 2026