When people move to a new place, they consider lots of things: Do they speak the language there? Do they have family or friends already living there? Do their values match those of the new country? How far away is it from where they’re from?
In 2022, researchers at the Max Planck Institute for Demographic Research came up with a new way to measure how similar one country’s culture is to another’s, using data from Facebook. They found that adding this cultural similarity measure to models predicting migration patterns can give us better predictions.
A nuanced picture
Before this, these models mainly looked at things like population, location, language, distance, and shared history. But with Facebook data, the researchers discovered that people’s food and drink preferences also play a role in where they choose to move.
So, understanding people’s cultural connections is crucial in figuring out why and where they move to. This research shows how technology can help us uncover new insights into human behavior.
“Cultural distance is difficult to measure and has not been widely included in gravity models to assess and predict migration. However, culture does play a very important role in migration processes and we wanted to examine the importance of cultural similarity in migration research. Specifically, we tested measures of cultural similarity based on food and drink interests on Facebook to analyze international migration flows,” the researchers explain.
Why we move
The researchers analyzed Facebook data from 16 countries, including Argentina, Australia, Brazil, and others. They found that this data can help us understand and even predict how migration patterns change over time.
For instance, if there’s a rise in the number of Facebook users in the United States showing interest in traditional Brazilian cuisine, it could mean a few things. One possibility is that there’s been an increase in Brazilian immigrants moving to the U.S. As more Brazilians settle in the country, they bring their culture with them, sparking curiosity among Americans.
If these Brazilian immigrants form tight-knit communities in the U.S., it’s likely that more Brazilians will follow suit, leading to a continuing rise in the Brazilian immigrant population. In essence, the data helps us see how cultural connections on platforms like Facebook can influence migration trends.
“In this case, the number of Facebook users interested in Brazilian food and drink serves as an indicator of the size of the Brazilian community in the U.S. One of our key findings shows the importance of cultural similarities between countries in predicting migration flows between them,” the researchers say.
“Variables such as language, history, and geographic distance are static and symmetric, meaning that the distance between the U.S. and Brazil would hardly change, and it’s the same regardless of the direction from which it is viewed. We found that cultural aspects of daily life are sensitive to changes in the environment and can be represented as an asymmetric and dynamic measure of similarity between countries.”
Efficient data gathering
Using Facebook Ads data offers an efficient way to collect information passively, allowing researchers to develop metrics of similarity that are both timely and cost-effective. For instance, these metrics can swiftly capture changes, particularly when migration patterns shift rapidly due to crises or conflicts like the Russian invasion of Ukraine.
By tapping into social media data, researchers gain valuable insights into how cultures are evolving and how migration patterns are changing. This information equips policymakers to make more responsive and informed decisions, especially when addressing complex global challenges such as hosting refugees.
However, it’s important to acknowledge limitations. This study has constraints due to limited data availability and a small number of countries included. The methodology for measuring cultural similarities relies solely on Facebook users’ interests in food and drink, potentially overlooking other relevant factors like entertainment, celebrities, or sports. Moreover, social media data may carry biases that need to be considered.
