5 ms·

Link to Full paper https://cseveren.github.io/files/FormativeExperiences_Paper_and_Appendix.pdf https://cseveren.github.io/files/FormativeExperiences_Paper_...

• refurb 4y ago
  Maybe I’m going nuts but it looks like they used 2000 US Census data on driving, so yes, they only looked at that year. They are literally comparing people who are older and younger with the cohort in question, so confounding results with how driving changes at various ages.
  • JumpCrisscross 4y ago
    > it looks like they used 2000 US Census data on driving, so yes, they only looked at that year The first major sentence under data: “The decennial census asks questions about commuting mode and time. ‘Journey to Work’ questions appear in the 1980, 1990, and 2000 censuses, and in the American Community Survey (ACS) (Ruggles et al. 2020). We use data from these three censuses, as well as the 2006/10, 2011/15, 2016, and 2017 ACS” [1]. Confirmation bias much [2]? [1] https://cseveren.github.io/files/FormativeExperiences_Paper_and_Appendix.pdf https://cseveren.github.io/files/FormativeExperiences_Paper_... bottom of page 6 [2] https://en.m.wikipedia.org/wiki/Confirmation_bias https://en.m.wikipedia.org/wiki/Confirmation_bias
    • refurb 4y ago
      It must be my years of reading shoddy economics studies that created that bias.
      • superhuzza 4y ago
        Given the exchange that just happened, are you sure they were all so shoddy? ;)