Opinion poll

An opinion poll estimates a population’s views or preferences by asking questions of a sample and analyzing the responses.

Sampling frame: A list or method defining the population units that can be selected for a sample. It determines who can be reached and represented in the poll.
Survey sampling: The selection of a subset of a population to collect data about the larger group. Opinion polls are a specialized use of survey sampling.
Election polling: Opinion polling that estimates voter preferences or likely election choices. It applies polling methods to candidates, parties, and electoral contests.
Census: A systematic collection of data from every member of a defined population. Unlike a poll, a census attempts to cover the entire population.
Probability sampling: A sampling method in which each population member has a known, nonzero chance of selection. Known selection chances support estimates of sampling uncertainty.
Target population: The full group about which a study aims to draw conclusions. A poll’s estimate applies only to its defined target population.
Exit poll: A survey of voters conducted as they leave polling places after voting. It measures reported vote choice and can examine voter groups.
Convenience sampling: A sampling method that selects participants because they are readily available. It can produce polls whose respondents poorly represent the target population.
Margin of error: A range around a sample estimate that describes uncertainty from random sampling under stated assumptions. It summarizes one source of uncertainty in a poll’s estimate.
Sample size: The number of units or people included in a sample. It affects precision, though it cannot by itself eliminate bias.

Linked from 8 pages