Reproduce “A systematic machine learning approach to measure and assess biases in mobile phone population data” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Carmen Cabrera, Francisco Rowe
The authors introduce a privacy-preserving method to measure coverage gaps in mobile-app counts against census populations, without individual demographic records. They apply it to four UK datasets benchmarked against the 2021 census and use area-level features to examine associations with bias. Meta’s Local Authority District coverage was about 2% to 14%; the measure does not establish demographic representativeness or causes.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.