• GaveUp [she/her]
    ·
    edit-2
    2 years ago

    Everybody, statistics is weird

    To get a 2% margin of error and a 95% confidence interval on your study representing 300 million people, you need to only properly do random sampling of 2401 people

    https://www.checkmarket.com/sample-size-calculator/

    @Shoegazer @WoofWoof91

    • moujikman
      ·
      2 years ago

      Yeah but your sample has to be actually random. Their methodology is "interviewed online in English" but then didn't explain how they found people online. Big red flag. And then they stratified the sample to align the demographics with the US population. IMO the data is trash and shouldn't be used for policy making.

      • bbnh69420 [she/her, they/them]
        ·
        2 years ago

        Goddamn it’s so easy to make a poll that’s worthless and write breathless articles about its findings