This project analysed anonymized IP geolocated data, originating from repeated logins to Yahoo! web services. This approach allowed the research team at Stanford, the Qatar Computing Research Institute and Queens College to gain insights on long- and short-term patterns of mobility of over 100 million individuals. The data set spans over the time period from July 2011 to July 2012.
At the global level, this study demonstrates the persistence of traditional migration patterns, developed due to influences of geographic proximity, shared language, as well as existing economic and former colonial ties. Additionally, the results showed that the short-term movements somewhat correlate with traditional migration patterns, and that next to the USA as the centre of global migration flows also a number of regional migration hubs developed, such as India and China. Furthermore, the data illustrated that circular migration – going back and forth in between the same countries – becomes more and more prominent, in particular within the European Union. Overall, this project shows how innovative data can usefully complement migration statistics, especially due to its capacity of producing valid and reliable estimates of global flows and stocks of migration. Potentially, this approach can be replicated with multiple other data IP geolocation data sets, ideally using various data sets to test and counter the selection bias created by the users’ choice of web services.
(Image: © Stanford University)