Election polls are easy to summarize and easy to misread. A headline might say that one party is on 39% and another on 37%, but those figures are estimates produced from a sample of the electorate. They are not a precise count of how every voter intends to vote.
Start with the population
The first question is who was surveyed. Polls may measure all adults, eligible voters, registered voters, likely voters or another defined group. Those populations can produce different results. In election tracking, the label should always travel with the number.
Check the fieldwork dates
Publication date and fieldwork date are not the same thing. A poll released today may have been conducted several days earlier. When political opinion is moving quickly, that difference matters. Votendi records fieldwork dates wherever the source provides them.
Understand the margin of error
Probability-based surveys usually report a margin of sampling error. It describes uncertainty created by interviewing a sample rather than the full population. It does not capture every possible source of error. Nonresponse, turnout assumptions, weighting choices, question wording and survey mode can also affect results.
A small lead should therefore not automatically be described as a meaningful advantage. The uncertainty around the difference between two candidates can be larger than the reported margin of error for either candidate individually.
Sample size is important – but not everything
Larger samples usually reduce sampling uncertainty, but sample quality matters as much as raw size. Pollsters typically weight responses so the final sample better reflects the target population. Transparent reporting of sampling, weighting, interview mode and question wording makes a poll easier to evaluate.
Look for a trend, not a single surprise
One poll can be an outlier. Several polls pointing in the same direction over time are more informative. This is why Votendi shows individual polls alongside averages and trend charts rather than treating every new release as a new state of the race.
The core rule is simple: read the methodology before the headline. The more competitive the race, the more important the details become.
Sources
- AAPOR – Polling Accuracy: https://aapor.org/polling-accuracy/
- AAPOR – Best Practices for Survey Research: https://aapor.org/standards-and-ethics/best-practices/
- Pew Research Center – Understanding the margin of error in election polls: https://www.pewresearch.org/short-reads/2016/09/08/understanding-the-margin-of-error-in-election-polls/
Editorial note: This explainer is evergreen and can be linked from poll pages and the methodology section.
