• Virkkunen@fedia.io
    link
    fedilink
    arrow-up
    2
    ·
    7 hours ago

    I’m not really trying to disprove or disagree with anything, I just think that knowing the sample is important. For instance, earlier in Hungary, we’ve had a lot of billboards and other media claiming that 99% of Hungarians were against things like sending aid to Ukraine and gender affirming politics. In a purely statistical sense, this was correct and could dissuade the common folk into thinking that’s representative of the country. However when you investigate further, their research was done on just a couple thousand citizens that were all either affiliated someway to Fidesz (the rulling party) or historically voted for them, which overwhelmingly skews the results towards one end.

    • airglow@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      ·
      edit-2
      4 hours ago

      Hey, I think you’re totally right to challenge a statistic when it looks questionable. Censuswide didn’t release the full data publicly, and the survey was commissioned by the Ghostery ad blocker, so there’s reason to suspect that the data is biased.

      I trust the YouGov data more, since YouGov is also a credible pollster and the data is being provided as market research data for businesses. However, since I don’t subscribe to their data service, I don’t have details of the methodology here, either.