Curbing Xenophobia? Assessing the Impact of COVID-19 Lockdown Measures on Anti-Refugee Sentiment in Turkey

Abstract

Is COVID-19 increasing or decreasing sympathy for refugees? How do COVID-19 restrictions on movement affect these sentiments? How does this influence inform policies to improve cohabitation between host communities and forcibly displaced populations? We explore these questions in the context of Turkey by investigating the effect of physical isolation—as opposed to movement—on sentiment. We analyze Google Mobility Reports and Twitter data to measure the impact of COVID-19 related restrictions on movement in Turkey on sentiment about refugees. We use measures of movement in Turkey during the COVID-19 pandemic made available through the Mobility Report, which we model against sentiment extracted from 7,988 tweets collected between February 15 and August 31. Descriptive analysis shows that xenophobic sentiment decreased somewhat during the pandemic (from March 2020) compared to the previous month. To explain this reduction, we use statistical tests aggregated at the country and region levels and show that four types of decreased mobility correlate with increased sympathy toward refugees: the more people stay at home, the more positive sentiment toward refugees they exhibit on Twitter. We conclude by proposing possible explanations for these findings, outlining policy recommendations, and describing the type of data needed to test the potential mechanisms.

Keywords: Refugees; COVID-19; Lockdowns; Natural Language Processing; Twitter; Public Opinion

JEL Classification: Z13, J6, I1, Z18