Results from the four surveys are based on telephone interviews conducted under the direction of Princeton Survey Research Associates International. Random samples of telephone households were called nationally, in New Hampshire, and South Carolina. In Iowa, samples of registered voters were called.
National Survey
Interviews were conducted Nov. 18 – Dec. 1 with a total of 469 adults 18 or older who identified themselves as Democrats or independents who leaned Democratic. All respondents said they were registered to vote and were very likely or somewhat likely to vote in a Democratic primary or caucus in their state next year. Respondents who later said they would not vote when asked for their choice among the Democratic contenders were excluded from the tabulations. Based on the total sample, one can say with 95% confidence that the error attributable to sampling and other random effects is plus or minus 5 percentage points.
Iowa Survey
Interviews were conducted Nov. 25 – Dec. 4 with a total of 394 registered voters in Iowa (registered Republicans were excluded from the sample). All respondents were age 18 or older, and said they will definitely or probably attend the January 19th Democratic caucuses. Respondents who later said they would not vote when asked for their choice among the Democratic contenders were excluded from the tabulations. Based on the total sample, one can say with 95% confidence that the error attributable to sampling and other random effects is plus or minus 6 percentage points.
New Hampshire Survey
Interviews were conducted Nov. 18 – Dec. 1 with a total of 585 adults 18 or older who said they were not registered as Republicans. All respondents said they were registered to vote and planned to vote in the Democratic presidential primary on January 27th. Respondents who later said they would not vote when asked for their choice among the Democratic contenders were excluded from the tabulations. Based on the total sample, one can say with 95% confidence that the error attributable to sampling and other random effects is plus or minus 5 percentage points.
South Carolina Survey
Interviews were conducted Nov. 18 – Dec. 1 with a total of 566 adults 18 and older who said they are registered to vote and planned to vote in the Democratic presidential primary on February 3rd. Respondents who later said they would not vote when asked for their choice among the Democratic contenders were excluded from the tabulations. Based on the total sample, one can say with 95% confidence that the error attributable to sampling and other random effects is plus or minus 5 percentage points.
In addition to sampling error, one should bear in mind that question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of opinion polls.
Survey Methodology in Detail
The samples for the national, New Hampshire, and South Carolina surveys are random digit samples of telephone numbers selected from telephone exchanges in the continental United States and the states of New Hampshire and South Carolina, respectively. The random digit aspect of the sample is used to avoid “listing” bias and provides representation of both listed and unlisted numbers (including not-yet-listed). The design of the sample ensures this representation by random generation of the last two digits of telephone numbers selected on the basis of their area code, telephone exchange, and bank number.
The telephone exchanges were selected with probabilities proportional to their size. The first eight digits of the sampled telephone numbers (area code, telephone exchange, bank number) were selected to be proportionally stratified by county and by telephone exchange within county. That is, the number of telephone numbers randomly sampled from within a given county is proportional to that county’s share of telephone numbers in the U.S. or within a given state. Only working banks of telephone numbers are selected. A working bank is defined as 100 contiguous telephone numbers containing one or more residential listings.
The sample was released for interviewing in replicates. Using replicates to control the release of sample to the field ensures that the complete call procedures are followed for the entire sample. The use of replicates also insures that the regional distribution of numbers called is appropriate. Again, this works to increase the representativeness of the sample.
At least 10 attempts were made to complete an interview at every sampled telephone number. The calls were staggered over times of day and days of the week to maximize the chances of making a contact with a potential respondent. All interview breakoffs and refusals were re-contacted at least once in order to attempt to convert them to completed interviews. In each contacted household, interviewers asked to speak with the “youngest male 18 or older who is at home.” If there is no eligible man at home, interviewers asked to speak with “the oldest woman 18 or older who is at home.” This systematic respondent selection technique has been shown empirically to produce samples that closely mirror the population in terms of age and gender.
Non-response in telephone interview surveys produces some known biases in survey-derived estimates because participation tends to vary for different subgroups of the population, and these subgroups are likely to vary also on questions of substantive interest. In order to compensate for these known biases, the sample data are weighted in analysis.
The demographic weighting parameters used in the national, New Hampshire, and South Carolina samples are derived from a special analysis of the most recently available Census Bureau’s Current Population Survey (March 2003). This analysis produced population parameters for the demographic characteristics of households with adults 18 or older, which are then compared with the sample characteristics to construct sample weights. The analysis only included households in the continental United States or the individual states that contain a telephone. The weights are derived using an iterative technique that simultaneously balances the distributions of all weighting parameters.
The sample for the Iowa survey was drawn from the database of registered voters maintained by the office of the Iowa Secretary of State, and excluded registered Republicans. This database is updated continuously and represents the most complete and accurate record of voter registration available. Most voter records contain a telephone number for the voter, and these were used to contact the sampled individuals. Data from Iowa were not weighted. Ann Selzer and Dana Birnberg of Selzer & Company provided helpful advice on polling Iowans.
