Large private sector data sources are now increasingly being leveraged to predict economic and social indicators. Data from Yelp You can predict changes in the number of businesses and restaurants. Google search data It is used in a variety of situations, including estimating GDP, unemployment rates, and the number of people infected with COVID-19. X (formerly Twitter) data It is used for real-time earthquake damage estimation, influenza prediction, etc.
These sources can be used to supplement survey and administrative data. For example, they can provide real-time metrics that are only available periodically from surveys, or they can provide more detailed data than is available from official sources.
Facebook marketing data in social science research
A new paper uses data from 59 countries to show that anonymized Facebook marketing data (among other data sources) can support estimates of poverty. The Facebook Marketing API allows you to query the number of daily and monthly active Facebook users (DAU and MAU) according to various characteristics for any location (up to 1 kilometer).
In a 2020 paper, Fatehkia et al. Initially, we tested the use of Facebook marketing data for poverty estimation in the Philippines and India and found that characteristics such as the proportion of Facebook users connected to Facebook through high-end phones were correlated with asset-based wealth indices. I found out that
We extend Fatehkia et al.'s approach by querying Facebook marketing data for 63,854 survey cluster locations from the Demographic and Health Surveys (DHS). Compare asset-based wealth indexes captured with DHS data to over 30 metrics from the Facebook Marketing API. We compared countries by characteristics such as the percentage of Facebook users interested in restaurants, the percentage determined to travel abroad frequently, and the percentage connecting to Facebook via high-end devices (similar to Fatehkia et al.). I discovered something that they all have in common. Everything is positively connected to wealth. Figure 1 shows a scatterplot comparing the percentage of Facebook users interested in restaurants to the wealth index for a given country.
