How American Voters Decide: Poll Methodology.

December 7, 2007

How American Voters Decide: Poll Methodology

Story by Mark Halperin and Amy Sullivan

Most Americans are just starting to tune in to the 2008 presidential campaign — and with headlines focused on security expenses for Rudy Giuliani's liaisons with his then mistress (now wife) and on Hillary Clinton's attacks on Barack Obama's kindergarten assignments, they may very well tune back out until after the holidays.

The citizens of Iowa and New Hampshire don't have that luxury. Campaign fliers and phone solicitations have been inundating them for months, and anyone looking to fuel up the family sedan risks running into a glad-handing candidate at the gas station. But that just makes them the perfect subjects for the first installment of our TIME election-year survey of the American electorate.

SRBI conducted interviews with registered voters nationally and in Iowa and New Hampshire. Below is an explanation of how the polls were conducted. The full story, together with an interactive chart, will be found at: http://www.time.com/time/nation/article/0,8599,1691613,00.html.

Methodology (National Poll)

This TIME poll was conducted by telephone between Nov. 12-19, 2007 among a national random sample of 1,863 adults, resulting in 1,503 self-reported registered voters, age 18 and older throughout America. The data have been weighted to reflect the demographic composition of adult Americans.

The margin of error for the entire sample is approximately +/- 3 percentage points. The margin of error is higher for subgroups. Surveys are subject to other error sources as well, including sampling coverage error, recording error, and respondent error.

The National Poll's partisan breakdown among Registered Voters is as follows: 32% Democrat 30% Republican 28% Independent

Methodology (Iowa and New Hampshire Polls)

These TIME polls were each conducted by telephone between November 12-19, 2007 among a random sample of 1,000 self-reported registered voters in Iowa and New Hampshire respectively, age 18 and older, likely to vote in the 2008 presidential election.

Likely voters included in the sample included those who said they were - 100% certain to vote in the 2008 presidential election, OR - Probably going to vote in the 2008 presidential election.

The data have been weighted to reflect the demographic composition of Iowa and New Hampshire adults respectively.

The margin of error for the entire sample for each poll is approximately +/- 3 percentage points. The margin of error is higher for subgroups. Surveys are subject to other error sources as well, including sampling coverage error, recording error, and respondent error.

Related Links