I would be interested in some experimental data to play around with causality concepts. Ideally, this would be data where an accompanying paper is available as well. The data set could be relatively small, and should ideally come from a randomized experiment. It would be cool if there were some covariates. It would also be ideal if the experiment the dataset is based on would be fairly intuitive to understand for a layman in this area.

Can you think of a dataset, or sources for these types of datasets, that at least satisfy some of these criteria?

Many thanks

  • 1
    Loads of weather data has some covariates between different things (temperature at two different places, or temperature and rainfall, etc.), also plenty of papers on google scholar. For data sources, see this question: opendata.stackexchange.com/questions/10154/…
    – Swier
    Dec 27, 2016 at 10:21
  • Can this ever be experimental data?
    – clog14
    Dec 27, 2016 at 16:13
  • Can you say more about what you mean by causality in this context? My background is in causal inference through epidemiology, but I want to make sure I'm on the same page. I think there are a few datasets covered in some of the books on causal inference around... what have you tried? Dec 28, 2016 at 2:12


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.