Analyzing the effect of time in migration measurement using geo-referenced digital trace data

Abstract

Geo-referenced digital trace data offer unprecedented flexibility in migration estimation. Due to their high temporal granularity, many different migration estimates can be generated from the same dataset by changing the definition parameters. Yet despite the growing application of digital trace data to migration research, strategies for taking advantage of their temporal granularity remain largely underdeveloped. In this paper, we provide a general framework for converting digital trace data into estimates of migration transitions and for systematically analyzing their variation along a quasi-continuous time-scale, analogous to a survival function. From migration theory, we develop two simple hypotheses regarding how we expect our estimated migration transition functions to behave. We then test our hypotheses on simulated data and empirical data from three different platforms in two internal migration contexts: geotagged Tweets and Gowalla check-ins in the U.S., and cell-phone call detail records in Senegal. Our results demonstrate the need for evaluating the internal consistency of migration estimates derived from digital trace data before using them in substantive research. At the same time, however, common patterns across our three empirical datasets point to an emergent research agenda using digital trace data to study the specific functional relationship between estimates of migration and time and how this relationship varies by geography and population characteristics.

Demography: Fast and Slow

“…2018; Fiorio et al. 2021). For instance, the use of daily mortality surveillance data has already been central to studying air pollution (Liu et al.…”

“…First, demographic data collection needs to fully take the speed of population change into account, collecting more up-to-date and frequent information on stocks and flows at higher frequencies, as opposed to relying on more traditional snapshots such as population censuses, or surveys, spaced 10 years apart. We have new opportunities thanks to the digital revolution: more frequent information on population and enhanced methods for "nowcasting" the present using register data as well as digital footprints (Cesare et al 2018;Fiorio et al 2021). For instance, the use of daily mortality surveillance data has already been central to studying air pollution (Liu et al 2019) andCovid-19 (Michelozzi et al 2020), and Facebook data at fine spatio-temporal granularity have been used to show the impact of Hurricane Maria on emigration from Puerto Rico (Alexander, Polimis, and Zagheni 2019).…”

Guest Editorial: Spatial demography in regional science

“…This then leads to the idea of spatial hazard (such as mobility and migration) being the counterpart of temporal hazard (such as survivorship). The growth of georeferenced digital trace data provides an emerging opportunity to test this approach in mobility and migration studies (e.g., Fiorio et al 2021). As the authors show, longitudinal methods need not be constrained to individuals and groups: the longitudinal transition of areas like neighborhoods is also a possible, but underexploited, extension.…”

“…Furthermore, modelling may benefit from artificial intelligence methods and machine learning for complex space-time data analysis (e.g., Molina and Garip 2019). Integrated public data infrastructures, that link administrative data from many sources, high-dimensional geo-referenced data from mobile phones and internet-based data on mobility can be particularly helpful in studying geographic mobility (e.g., Fiorio et al 2021). Finally, a greater focus may be expected on networks and spatial interactions (e.g., Small and Adler 2019).…”