American Research Group, Inc.
The New Hampshire Poll


2004 New Hampshire Democratic Tracking Methodology

The results for this tracking survey are based on a minimum of 600 completed telephone interviews among a statewide random sample of likely Democratic primary voters in New Hampshire. Telephone households are selected by adding random digits to telephone exchanges selected in a random probability-proportionate-to-size sample of the state. Respondents in the telephone households are randomly selected and then screened for voter registration (including intent to register on or before primary day) and likelihood of voting in the Democratic primary. New Hampshire is a party-registration state and registered Republicans are excluded from the sample.

During each day of the tracking, a minimum of 200 interviews are completed among respondents saying they will definitely vote in the Democratic primary on January 27. The results reported are based on combining the last three days of interviews for a total sample size of a minimum of 600 likely Democratic primary voters. The composition of the sample by party is reported for each three-day period. Interviews began on December 26, 2003 and the results reported on December 28, 2003 are based on interviews conducted December 26, 27, and 28, 2003.

After the first day of interviews, the daily sample sizes will vary slightly because of the impact of callbacks. By the fourth day of the tracking (December 29), the three-day sample should reflect the geographic composition of the vote in the state and include difficult-to-reach Democratic primary voters.

The theoretical margin of error for a sample size of 600 likely Democratic primary voters reported in the tracking is plus or minus 4 percentage points, 95% of the time, on questions where opinion is evenly split.

The theoretical margin of error for a sample size of 700 likely Democratic primary voters reported in the tracking is plus or minus 3.7 percentage points, 95% of the time, on questions where opinion is evenly split.

The theoretical margin of error for a sample size of 800 likely Democratic primary voters reported in the tracking is plus or minus 3.5 percentage points, 95% of the time, on questions where opinion is evenly split.

*Please note that Carol Moseley Braun's name was removed as of the tracking beginning January 18.

*Please note that Dick Gephardt's name was removed as of the tracking beginning January 20.

Question wording:

If the New Hampshire Democratic Presidential Preference Primary were being held today between *Carol Moseley Braun, Wesley Clark, Howard Dean, John Edwards, *Dick Gephardt, John Kerry, Dennis Kucinich, Joe Lieberman, and Al Sharpton, for whom would you vote? (NAMES ARE ROTATED)

Return to Tracking Results