Like all polling firms, Rasmussen Reports weights its data to reflect the population at large. Among other targets, Rasmussen Reports weights data by political party affiliation using a dynamic weighting process (see methodology).

During the final two months of the election season, we update these targets on a weekly basis to stay current with the public mood. While partisan affiliation is generally quite stable over time, there are a fair number of people who waver between allegiance to a particular party or independent status. The intensity of a campaign season may cause subtle changes in the partisan landscape as Election Day draws near.

The targets are not set arbitrarily. Rather, they are established based upon survey interviews with a separate sample of adults nationwide completed during the preceding six weeks. A total of 500 nightly interviews are conducted for a total of 21,000 interviews over the six week period.

This week, for the first time in any week since mid-September, the partisan identification trends drifted slightly in the Republican direction. However, the electorate is still a bit more Democratic than it was in mid-September and quite a bit more Democratic than it was during Election 2004. When President Bush was re-elected, the number of Republicans and Democrats in the nation were essentially even.

For polling data released during the final days of Election 2008, the partisan weighting targets will be 39.9% Democratic, 33.4% Republican, and 26.7% unaffiliated. For the preceding week, the partisan weighting targets were 40.0% Democratic, 32.8% Republican, and 27.2% unaffiliated.

In mid-September, the last weighting targets established before the financial crunch became visible on Wall Street, the targets were 38.7% Democratic, 33.6% Republican, and 27.7% unaffiliated. Since that time, the number of Democrats has increased by 1.2 percentage points and the number of Republicans has declined by 0.2 percentage points for a net swing of 1.4 percentage points.

The current targets suggest a significantly better environment for Democrats than they faced in Election 2004. During Election 2006, there was a notable shift in partisan identification favoring Democrats as Election Day approached. During Election 2004, there was a notable shift in the opposite direction (see month-by-month results).

Outside of the election season, our baseline targets are established based upon survey interviews with a sample of adults nationwide completed during the preceding three months (a total of 45,000 interviews) and targets have been updated monthly.

In addition to weighting our election polling sample by partisan identification and other factors, Rasmussen Reports also looks at a person’s interest in the campaign and other factors to estimate the likelihood of voting. For example, someone who is following the campaign on a daily basis is considered more likely to vote than someone who says they have no interest in the campaign. Some who says they are certain to vote is considered more likely to vote than someone who says something might come up to prevent them from voting.