The interplay of migration and cultural similarity between countries: Evidence from Facebook data on food and drink interests

Abstract

Migration has been proposed as one of the factors that shape cultural similarities across countries. However, studying the relationship between culture and migration has been challenging, in part because culture is difficult to quantify. The traditionally used survey questionnaires have a number of drawbacks, including that they are costly and difficult to scale to a large number of countries. To complement survey data, we propose the use of passively-collected digital traces from social media. We focus on food and drink as markers of a country’s culture. We then measure similarities between countries in terms of food and drink interests using aggregated data from the Facebook Advertising Platform. Methodologically, we offer approaches to measure the similarity between countries with both symmetric and asymmetric indices. Substantively, we assess the association between migration cultural similarity between countries by comparing our measure of cultural similarity with international migration data. In most countries, larger immigrant populations are associated with more similar food and drink preferences between their country of origin and the country of destination. Our results suggest that immigrants contribute to bringing the culture of their home countries to new countries. Moreover, our study identifies considerable variability in this pattern across countries.

The value of cultural similarity for predicting migration: evidence from digital trace data

“…However, the authors included different types of features, from politics to national parks, and used a 'black-box' model to represent countries' culture, which makes it hard to assess what exactly their index is measuring. More recently, Vieira et al (2022) presented a scalable, data-driven methodology to measure the cultural similarity between countries in terms of the most popular food and drink in each country from a list containing more than 200,000 interests on Facebook Ads (Speicher et al, 2018). Compared to Obradovich et al (2020), the methodology proposed by Vieira et al (2022) compared countries in terms of fewer, but explicitly known attributes selected from a very large data set.…”

“…More recently, Vieira et al (2022) presented a scalable, data-driven methodology to measure the cultural similarity between countries in terms of the most popular food and drink in each country from a list containing more than 200,000 interests on Facebook Ads (Speicher et al, 2018). Compared to Obradovich et al (2020), the methodology proposed by Vieira et al (2022) compared countries in terms of fewer, but explicitly known attributes selected from a very large data set. In other words, interests that are not relevant to any of the countries were disregarded to reduce feature sparsity.…”

“…In this section, we describe the main data sources used to collect international data used in our prediction models. We present the data we used to measure the cultural and the food and drink preference similarity between countries: the World Values Survey (Inglehart, 1997), Foursquare (Silva et al, 2014), and Facebook Ads (Vieira et al, 2022). We also include a description of data sources for gravity model variables such as population, area, and geographic distance, as well as migration flow as the outcome variable.…”

“…These countries cover many major regions of the world and also have a high Facebook 3/14 penetration rate: Argentina, Australia, Brazil, Chile, Great Britain, France, Indonesia, Japan, South Korea, Malaysia, Mexico, Russia, Singapore, Spain, Turkey, and the US. Vieira et al (2022) used Facebook Ads data to collect data regarding Facebook user's preferences in food and drink and proposed measures of cultural similarity between countries. Their methodology consisted of selecting a subset of popular food and drink for each country and then creating a vector representation according to their Facebook users' interests in those food and drinks.…”

“…We propose the use of a food and drink preference similarity measure, which is derived from Facebook Ads data (Vieira et al, 2022) and evaluate its potential to improve prediction of migration. This measure is timely, costeffective and scalable, and uses freely and publicly available aggregate data from the Facebook Ads regarding users' food and drink preferences.…”