Full Citation

Title: Combining Social Media Data and Traditional Surveys to Nowcast Migration Stocks

Citation Type: Miscellaneous

Publication Year: 2018

ISBN:

ISSN:

DOI:

NSFID:

PMCID:

PMID:

Abstract: Social media and Web data offer new opportunities to improve demographic knowledge and to complement more traditional data sources. Face-book, for example, can be thought of as a large digital census that is constantly updated. However, its users are not representative of the underlying population. Contrastingly, the American Community Survey (ACS) relies on smaller samples that may be noisy and that are published with a substantial delay from data collection. Additionally, ACS samples are representative of the underlying population and have historical depth. We generate now-casts, present and near-future predictions, of migration stocks that combine the best of the two complementary sources using a Bayesian hierarchical model. Facebook data, obtained via the Marketing API are timely, but lack demographic constraints on trends in age patterns that we extrapolate from ACS time series. Combining data sources and modeling strategies enables the researcher to weigh down inconsistencies and extract valuable insights without ignoring existing information. Although the focus of this article is on migration, our methods are general and contribute to the emerging literature on complementing social media and traditional data sources in various contexts. 2

Url: https://docs.google.com/document/d/1Tuh0dyWtB-m4LAG70cgySCcmvWuo6WqZ0MrLk_EfYSA/edit

User Submitted?: No

Authors: Zagheni, Emilio; Polimis, Kivan; Alexander, Monica; Weber, Ingmar; Billari, Francesco C

Publisher:

Data Collections: IPUMS USA

Topics: Migration and Immigration

Countries: United States