Welcome to DU!
The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards.
Join the community:
Create a free account
Support DU (and get rid of ads!):
Become a Star Member
Latest Breaking News
General Discussion
The DU Lounge
All Forums
Issue Forums
Culture Forums
Alliance Forums
Region Forums
Support Forums
Help & Search
Dem pollsters admit they blew it in 2020
PoliticoAfter 2016, when just about every poll heading into Election Day had HILLARY CLINTON winning, Democratic pollsters vowed they could never allow themselves to be that wrong again.
Now, a group of five leading Democratic polling firms is trying to explain, once and for all, why they blew it again in 2020, predicting victories for the party up and down the ballot that never materialized. As our polling guru Steve Shepard reports today, the competing firms — ALG Research, GBAO Strategies, Garin-Hart-Yang Research Group, Global Strategy Group and Normington Petts — banded together in an unusual collaboration after the election to conduct a self-autopsy of sorts. It acknowledged “major errors” and a failure “to live up to our own expectations” — yet, frustratingly, no easy solution to the problem of consistently overestimating how major Democratic candidates, including JOE BIDEN, would perform.
Among the culprits:
* Deteriorating public trust in institutions, government, the news media and, yes, the polling industry — driven by DONALD TRUMP’S bashing of those very institutions. Essentially, Trump voters were less willing to participate in polls.
* Pollsters again underestimated turnout among rural and white non-college-educated voters, who overwhelmingly backed Trump.
* Failure to detect late movement toward Trump and Republican candidates in the run-up to the election.
* Not accurately accounting for the fact that Democrats stayed home and answered their phones in greater numbers last year than Republicans who did not follow Covid-19 restrictions as closely.
But IDing the problems was easier than agreeing on fixes for next time. Even before Trump, the polling industry had been struggling to adapt to the evaporation of home landlines and the relative reliability they provided to survey takers. Trump has only exacerbated the challenge of measuring public sentiment when few people want to answer the phone and many actively distrust pollsters.
Now, a group of five leading Democratic polling firms is trying to explain, once and for all, why they blew it again in 2020, predicting victories for the party up and down the ballot that never materialized. As our polling guru Steve Shepard reports today, the competing firms — ALG Research, GBAO Strategies, Garin-Hart-Yang Research Group, Global Strategy Group and Normington Petts — banded together in an unusual collaboration after the election to conduct a self-autopsy of sorts. It acknowledged “major errors” and a failure “to live up to our own expectations” — yet, frustratingly, no easy solution to the problem of consistently overestimating how major Democratic candidates, including JOE BIDEN, would perform.
Among the culprits:
* Deteriorating public trust in institutions, government, the news media and, yes, the polling industry — driven by DONALD TRUMP’S bashing of those very institutions. Essentially, Trump voters were less willing to participate in polls.
* Pollsters again underestimated turnout among rural and white non-college-educated voters, who overwhelmingly backed Trump.
* Failure to detect late movement toward Trump and Republican candidates in the run-up to the election.
* Not accurately accounting for the fact that Democrats stayed home and answered their phones in greater numbers last year than Republicans who did not follow Covid-19 restrictions as closely.
But IDing the problems was easier than agreeing on fixes for next time. Even before Trump, the polling industry had been struggling to adapt to the evaporation of home landlines and the relative reliability they provided to survey takers. Trump has only exacerbated the challenge of measuring public sentiment when few people want to answer the phone and many actively distrust pollsters.
1 replies
= new reply since forum marked as read
= new reply since forum marked as read
