Search for a New Home: Refugee Stock and Google Search

Abstract

Following the assumption that trends of online queries may indicate intentions and help to predict human behavior, this study addresses the general issue of analyzing, nowcasting, and predicting migrant decisions through an analysis of Google search patterns in the case of Syrians in Turkey. Aiming to contribute to the literature on predicting migration patterns, we examine the relationship between Google search queries for province names in Turkey and the number of Syrians under temporary protection across provinces from January 2016 to December 2019 and demonstrate a positive and significant association. Then, we explore the predictive power of Google searches in predicting the stock of Syrians under temporary protection in Turkey across provinces. We exploit the alphabetical difference between Turkish and Arabic as the method of differentiation between host and migrant populations. Our findings indicate that Google searches can be good predictors for estimating refugee stocks, especially when traditional data are not available. They can also be helpful in forecasting the changing pattern of migrant stocks at frequent intervals, to which conventional socioeconomic indicators are less sensitive due to their less frequent reporting periods.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 2 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.