What can we learn from the migrants’ web?From association sites to personal blogs to the use of social networks, migrants generate a large mass of information which can be used for research. This raw data provides new insights on migration flows, but also on how diasporas are organized on the web.
A team from Wolfram Research, a research laboratory working on online data exploitation, recently published a study based on the data received from 1.26 million volunteer Facebook users. By matching the residence and birthplace of internet users, researchers were able to highlight migration within the United States, but also around the world. At the same time, the Facebook Data Science teams - who are behind the making of the Graph Search engine - published a study based on the place of origin and residence of their users. They could thus underline the phenomena of coordinated migrations within developing countries. This is characterized by the displacement of at least 20% of the population from one city to another, in order to experience great economic advancement. Facebook's data is invaluable insofar as it relates to more than one billion people in the world, providing much more accurate results than poll-based research. However several problems still remain to be resolved in the particular case of this social network. First of all, Facebook is banned in China, preventing researchers from accessing the data of one of the largest diasporas in the world. Moreover, when Facebook Labs publish a study, they refuse access to data for other researchers to carry out checks. It is therefore impossible to know if either the data or the analyses are correct. Finally, despite access to numerous and varied profiles, one wonders if the sample of internet users on which the Facebook analysis is based is really representative of a community of migrants, or if it is just limited to the most mobile and educated users.
Fortunately, other types of analyses, based on free access data, can be carried out. This is the case for the study conducted by the team of scientists led by Dana Diminescu on e-diasporas. Rather than focusing on migratory flows, the study focused on all of the association sites, alternative media, personal blogs and activities on social networks, which form a multitude of unique ecosystems for each diaspora. By analysing the links which connect one site to another, the researchers drew up graph charts of these E-diaspora. In the end, it is not the use of these sites that is analysed, rather the content published there and the networks they form. After viewing sites being used as a hub or epicentre, the relationships maintained between websites and, consequently, the image they provide of an always moving virtual community formed by migrants, can be determined.
Thanks to this study, a very accurate picture of a community of migrants, where it is located and how it is organized, can be obtained. Thus the team of researchers could find that the E-diaspora was mainly localised on American sites, even when this presence is not geographically proven. This could be the case for Chinese or Palestinians migrants, which also shows that flourishing web activity depends more on the host country than on the number of migrants present. Other virtual ecosystems make it possible to analyse the relationships maintained between certain diasporas and their country of origin. Several communities, such as French migrants, keep strong links with their country, as websites put online by the state allow them to access voting or social welfare services. Conversely, links between the Moroccan and Mexican communities and their countries of origin are often one-way or even seem to be non-existent, due to the lack of links between different sites. This also shows that these migrants prefer to turn to associations, media or community sites in their home countries, rather than their country of origin.