What is turnout bias in polls?

• A. When the sample under- or over-represents certain groups, skewing results.
• B. The margin of error around a turnout estimate.
• C. The total number of people surveyed.
• D. The confidence interval of a poll.

Answer: (A)

Turnout bias happens when the people who respond to a poll or who turn out to vote aren’t representative of the overall electorate, and those groups have different voting preferences. If the sample under- or over-represents certain groups, the poll’s results will be skewed toward those groups’ likely choices, making it not match what the true electorate will do on election day. This is why the option describing under- or over-representation of groups best captures turnout bias.

It’s not about how big the poll is, the average margin of sampling error, or the confidence interval. Those terms describe precision and uncertainty, not systematic differences in who shows up or responds. For example, if younger voters—who may lean differently—are harder to reach or less likely to respond, the poll could overestimate support among older voters and underestimate it among younger ones.