• Flying Squid@lemmy.world
    link
    fedilink
    arrow-up
    50
    arrow-down
    3
    ·
    1 year ago

    I’m not super trusting of polls anymore, especially because they’re usually done by telephone. However-

    The poll had a sample of 1,032 adults, age 18 or older, who were interviewed online; it has a margin of error of plus or minus 3.2 percentage points for all respondents.

    This makes me a little more trusting despite that whopping MoE. It sounds like bad news for Trump overall.

    • Hyperreality@kbin.social
      link
      fedilink
      arrow-up
      40
      ·
      1 year ago

      I have a related degree. The reason people distrust polls, is because the media frequently misreports or misrepresents them.

      Eg. aggregated polling from the 2016 suggested Trump had a 1/3 chance of winning. If you believed some media coverage every poll said Clinton was certain to win. That was how the media reported on the polling, not the polling itself. Invariably Trump winning in 2016 was within the margin of error.

      that whopping MoE

      Not a large margin of error. You’re extrapolating from 1000 people to 300 million. It’s astonishing it’s that low if you think about it.

      because they’re usually done by telephone

      Not that common anymore. Often they’ll do a a telephone poll then supplement it with online or other methods. Here’s IPSOS’s article about this poll:

      The study was conducted online in English. The data for the total sample were weighted to adjust for gender by age, race/ethnicity, education, Census region, metropolitan status, household income, and political party affiliation. The demographic benchmarks came from the 2022 March Supplement of the Current Population Survey (CPS). Party ID benchmarks are from recent ABC News/Washington Post telephone polls.

      https://www.ipsos.com/en-us/politico-indictment-august-2023

      • CoggyMcFee@lemmy.world
        link
        fedilink
        arrow-up
        15
        ·
        1 year ago

        I remember reading 538 leading up to the 2016 election, and hearing them say repeatedly that if Trump has a 1 in 4 chance (or whatever amount) of winning the election, not only is it possible for Trump to win, but in fact it means you actually expect it to happen in 1 out of 4 times.

        • Asifall@lemmy.world
          link
          fedilink
          arrow-up
          10
          ·
          1 year ago

          Yeah I remember this too. I think the problem is that people simply don’t understand statistics and don’t realize a 70% chance of winning is totally different from getting 70% of the vote. I like what 538 has been doing in recent years by presenting odds rather than percentages, but people like echo chambers that confirm their biases so idk if this “polls don’t work” narrative is going to go away any time soon.

        • Alien Nathan Edward@lemm.ee
          link
          fedilink
          arrow-up
          9
          ·
          1 year ago

          That’s the article that has caused me to trust 538 above any other election prediction source. When HRC was doing a preemptive victory lap in Texas and HuffPo was publishing articles that said she had a 99% chance of winning, Nate Silver and Co were the only ones willing to admit the possibility of what would later become reality.

      • Monkey With A Shell@lemmy.socdojo.com
        link
        fedilink
        arrow-up
        8
        arrow-down
        1
        ·
        1 year ago

        Indeed, the problem isn’t polls themselves, assuming they’re well constructed they’re generally sound data, it’s the interpretation and packaging of it as reported to the larger populous that gets in the way. Sometimes it gets to the point of funny when someone does an infographic where 30% and 60% somehow appear to have the same weight.

        Lies, damn lies, and statistics…

          • Monkey With A Shell@lemmy.socdojo.com
            link
            fedilink
            arrow-up
            3
            ·
            edit-2
            1 year ago

            Thus why I’m not a writer, but at least the intent was there 🙂

            I’m doing good if it doesn’t look like a drunken baboon wrote it sometimes due to fat finger typing.

        • candybrie@lemmy.world
          link
          fedilink
          arrow-up
          1
          ·
          1 year ago

          Considering people tend to view probability as 100% A, 100% B or 50-50, I’m not sure showing a 30-60 split as the same weight is really a bad choice…

      • cerevant@lemm.ee
        link
        fedilink
        English
        arrow-up
        5
        ·
        1 year ago

        My biggest issue with polls is that the media tout them as predictions, ignoring the fact that even if the data is 100% valid, circumstances can change dramatically in just a couple of days.

        I maintain that polls are not actionable data for voters. They can help campaigns see trends and gauge the effectiveness of messaging, but they are useless to voters.

        • Hyperreality@kbin.social
          link
          fedilink
          arrow-up
          4
          arrow-down
          1
          ·
          edit-2
          1 year ago

          They can and do change in just a couple of days, but the real issue is that the media invariably fails to mention the margin of error or confidence interval.

          It’s always Candidate A 51%, candidate B 49%. When in reality it’s inevitably something like “There’s 19/20 chance that candidate gets between 48.5-53.5% of the vote, and that candidate gets between 46.5-51.5% of the vote.”

          And then when candidate B wins, the media will go “Why did the polls get it wrong?” when the election was always to close to call definitively.

          Oh, and this is obviously ignoring the far more sinister use of misrepresented polling data, micro-demographically targetted thanks to big data harvested from social media. Think Cambridge Analytica algorithms which have determined that women in village X with one child and dog, being more likely to vote party Y, and then targetting them on social media with stories about the polls showing the result is a foreglone conclusion and that there’s no point voting.

      • NotSpez@lemm.ee
        link
        fedilink
        arrow-up
        1
        ·
        1 year ago

        Thanks! I really like how Lemmy users with expertise in their area can add nuance to a lot of reporting, it really matters.

    • atzanteol@sh.itjust.works
      link
      fedilink
      arrow-up
      3
      arrow-down
      12
      ·
      1 year ago

      I’m not super trusting of polls anymore

      Like, do you not believe the people responded the way they say they did?

      • blackbelt352@lemmy.world
        link
        fedilink
        arrow-up
        9
        arrow-down
        1
        ·
        1 year ago

        Not the responses themselves but the methodologies of collecting responses don’t result in accurate representation of the population.

        Using collection methods that skew demographics in one direction or another, like older people being more likely to pick up a phone call.

        Failing to account for other potentially major variables. Like the 2016 and 2020 elections, pollsters failed to account for negative voter turnout, people who were motivated to vote against a specific candidate, which had major impacts on the elections.

          • blackbelt352@lemmy.world
            link
            fedilink
            arrow-up
            2
            ·
            1 year ago

            I like that, it’s a pretty good breakdown of controlled variables. And it looks like they’re factoring in socioeconomic factors too, which is always a good thing.

            After the last however long of bad polling, especially in the last 8 years, it’s refreshing to see some better methodology but it’s still going to take a while to get that general trust back.

      • YoBuckStopsHere@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        arrow-down
        3
        ·
        1 year ago

        Most polling is done via landline phone. Thus polling does reflect well on the actual voting population.

        • Mouselemming@sh.itjust.works
          link
          fedilink
          arrow-up
          1
          ·
          1 year ago

          It would probably particularly represent older voters, who might lean towards Trump. Although I’m over 60 and white and answer the landline phone, and I abhor him and get more progressive every day. Come to think of it, I also hang up on pollsters, so…

        • mrnotoriousman@kbin.social
          link
          fedilink
          arrow-up
          1
          ·
          1 year ago

          I’m not so sure that’s true anymore. I don’t go looking at every single study but I usually see a mix of landline, cell/sms, and online samples

          • YoBuckStopsHere@lemmy.world
            link
            fedilink
            English
            arrow-up
            1
            arrow-down
            1
            ·
            1 year ago

            That is why I said most, not all.

            About the Study

            This Ipsos poll was conducted August 18-21, 2023 on behalf of Politico Magazine, using the probability based KnowledgePanel®. This poll is based on a representative sample of 1,032 U.S. residents, age 18 or older, including 272 Republican respondents, 321 Democratic respondents, and 319 independent respondents.

            The study was conducted online in English. The data for the total sample were weighted to adjust for gender by age, race/ethnicity, education, Census region, metropolitan status, household income, and political party affiliation. The demographic benchmarks came from the 2022 March Supplement of the Current Population Survey (CPS). Party ID benchmarks are from recent ABC News/Washington Post telephone polls. The weighting categories were as follows:

            Gender (Male, Female) by Age (18–29, 30–44, 45-59 and 60+) Race/Hispanic Ethnicity (White Non-Hispanic, Black Non-Hispanic, Other, Non-Hispanic, Hispanic, 2+ Races, Non-Hispanic) Education (Less than High School, High School, Some College, Bachelor or higher) Census Region (Northeast, Midwest, South, West) Metropolitan status (Metro, non-Metro) Household Income (Under $25,000, $25,000-$49,999, $50,000-$74,999, $75,000-$99,999, $100,000-$149,999, $150,000+) Party ID (Democrat, Republican, Independent, Something else)

            The margin of sampling error is plus or minus 3.2 percentage points at the 95% confidence level, for results based on the entire sample of adults. The margin of error takes into account the design effect, which was 1.08 for all adults. The margin of sampling error is plus or minus 6.2 percentage points at the 95% confidence level, for results based on the sample of Republicans. The margin of error takes into account the design effect, which was 1.08 for Republicans. The margin of sampling error is plus or minus 5.8 percentage points at the 95% confidence level, for results based on the sample of Democrats. The margin of error takes into account the design effect, which was 1.11 for Democrats. The margin of sampling error is plus or minus 5.7 percentage points at the 95% confidence level, for results based on the sample of independents. The margin of error takes into account the design effect, which was 1.09 for independents. In our reporting of the findings, percentage points are rounded off to the nearest whole number. As a result, percentages in a given table column may total slightly higher or lower than 100%. In questions that permit multiple responses, columns may total substantially more than 100%, depending on the number of different responses offered by each respondent.