Code and Data

As a service to the community, we make the code and datasets from this study available to the public. Interested researchers can download them here.

This archive contains the partisan bias scales used in our study. Partisan scores are calculated per effective second-level domain, based on the relative fractions of registered Republicans and Democrats that shared links to that domain on Twitter. Scores are rounded, and all domains shared less than 50 times have been filtered out. The archive also includes the other scales we compare against in the paper including those from Bakshy et al., Budak et al. (aggregated by domain), Pew, and AllSides.

Citation

If you use this dataset, we ask that you please cite our paper as follows:

@article{ robertson-pacmhci-2018,
    author = "Ronald E. Robertson and Shan Jiang and Kenneth Joseph and Lisa Friedland and David Lazer and Christo Wilson",
    title = {{Auditing Partisan Audience Bias within Google Search}},
    booktitle = {{Proceedings of the ACM: Human-Computer Interaction}},
    volume = "2",
    number = "CSCW",
    month = "November",
    year = "2018""
}