• anarcho_blinkenist [none/use name]@hexbear.net
    link
    fedilink
    English
    arrow-up
    22
    ·
    12 days ago

    it literally isn’t a good sample size, especially for their selection process in the breakdown of a place as diverse and varied in peoples and living conditions and environments like the US and its 350 million people. It’s a religious-focused NGO for “christian nationalism in politics”, has vanishingly few young people, does terrible breakdowns in the full report and tells us nothing about class or income, self-selects for those who are on consistent addresses in USPS lines with internet access who would be arsed to do these surveys (as well as has hundreds of self-selected opt-ins), the report is trash by a non-profit for “finding the intersection of religion and politics for clergy and the public”

    • GlueBear [they/them, comrade/them]@hexbear.net
      link
      fedilink
      English
      arrow-up
      25
      ·
      edit-2
      12 days ago

      No, it’s actually an excellent sample size. This study is absolutely worrying because it was conducted well.

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

      This is a sample size calculator,

      Punching in 350M for population size, 2% margin of error, and 95% Confidence interval we get a necessary sample size of at least 2401, this study had a sample size of 5352

      As for their margin of error, using the above population and confidence interval values but adding in 5352 for the actual sample size we get a calculated margin of error of 1.34%

      This study is valid and its findings should worry everyone.

      • Wertheimer [any]@hexbear.net
        link
        fedilink
        English
        arrow-up
        29
        ·
        edit-2
        12 days ago

        It’s a perfectly good sample size, you’re absolutely right, but @anarcho_blinkenist@hexbear.net raises good questions about whether it’s an appropriately random sample.

        Edit: They use weighting to put things closer to what the general population actually looks like, but here’s what their unweighted numbers are:

        • GlueBear [they/them, comrade/them]@hexbear.net
          link
          fedilink
          English
          arrow-up
          19
          ·
          12 days ago

          Okay I see the criticism, but would the more honest title:

          approximately 47% of Americans aged 30 and older believe immigrants should be put in militarized camps

          really have changed the conclusions and implications?

          It’s fucking disgusting no matter how you look at it.

          I suppose that number might go down to 30% at best if 18-29 year olds were properly sampled.

          • Wertheimer [any]@hexbear.net
            link
            fedilink
            English
            arrow-up
            9
            ·
            12 days ago

            I agree. Like, we could get the number down by asking the question a different way, too, but ultimately the issue is that ~97% of votes will go to candidates who will increase the power, reach, and budget of ICE and the military.

          • blobjim [he/him]@hexbear.net
            link
            fedilink
            English
            arrow-up
            4
            ·
            edit-2
            11 days ago

            It’s also in how the question is phrased and so on. They’re basically goading people into agreeing with right-wing stuff in these polls.

            Here’s the actual question and results:

            Q30e. Rounding up and deporting immigrants who are in the country illegally, even if it takes setting up encampments guarded by the U.S. military

            Strongly favor 22%

            Favor 25%

            Oppose 28%

            Strongly oppose 22%

            Skipped/ Refused 3%

            and the next question “Which statement comes closest to your view about how the immigration system should deal with immigrants who are currently living in the U.S. illegally? [ROTATE]” shows that over 60% of people say citizenship (56%) or permanent residency (11%)

            which also seems to entirely contradict the previous one lol.