Definition
Polling is the practice of estimating characteristics of a large population — voting intentions, consumer preferences, public opinion — from a randomly drawn sample. With well-designed random sampling, a few hundred to a few thousand respondents can give surprisingly precise estimates, with quantifiable uncertainty.
The mathematical engine is the sampling distribution of a proportion: for a sample of size n from a population with true proportion p, the sample proportion is approximately normal with mean p and standard deviation square root of p(1 - p) / n.
Why it matters
How it works
A pollster defines a target population, designs a random-sampling scheme to reach it, collects responses, and reports both the point estimate (sample proportion) and the margin of error (the half-width of the 95% confidence interval). For a simple random sample of 1,000 with a 50-50 split, the margin of error is roughly 3 percentage points.
The challenge in practice is response bias: not everyone called will answer, and those who do may differ systematically from those who do not. Modern weighting and adjustment use auxiliary information to correct for these gaps, but the corrections inherit assumptions that themselves can fail.