Oct 21, 2026 · 30 min
Polling’s old playbook struggles to reach voters
We try to fix election polling (update)
Reliable polling shapes political understanding, but the methods that once produced representative samples now miss too many people.
- 1Polling’s history shows that representative sampling matters more than simply collecting a large number of responses.
- 2Telephone surveys once made random sampling practical, but declining response rates and changing technology have undermined that model.
- 3Wisdom-of-crowds questions may reveal turnout and partisan sentiment, yet they cannot replace rigorous polling methods.
Don't miss
Nick Fountain makes more than 200 unsuccessful calls before finally completing a survey interview, turning polling’s response-rate crisis into a lived experience.
The brief
Planet Money revisits the polling crisis at Marist, where reporters train as callers before confronting how difficult it has become to reach a representative sample.
The episode traces polling’s early failures, from the Literary Digest’s oversized but biased sample to the 1948 error that pushed pollsters toward random sampling.
Telephone polling later flourished because landlines made random selection practical, but caller screening, cell phones, low response rates, and hard-to-reach groups broke that model.
The problem becomes tangible when Jeff Guo gets an interview quickly while Nick Fountain makes more than 200 unsuccessful calls before reaching one respondent.
Marist’s wisdom-of-crowds experiment offers clues about turnout and partisan sentiment, but its Republican tilt shows why the approach is not a complete fix.
Weekly email
State of the Pod, every Monday
The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.