Americans on the Federal Budget

(Sept. 2000)

Appendix C: How the Study Was Conducted

To prepare this study, the Center on Policy Attitudes conducted focus groups, a review of previous polls done by other organizations and, together with Knowledge Networks, conducted a nationwide survey.

The Focus Groups
COPA conducted several focus groups around the country, specifically in Richmond, Virginia; Cleveland, Ohio; and Frederick, Maryland. In the focus groups participants went through the same budget exercise, but were able to discuss their feelings and thoughts as they went along, providing added insight into how Americans feel about budgetary priorities.

Nationwide Survey
The survey was conducted with a nationwide sample of 721 adults conducted via the Internet, using Knowledge Networks’ web panel. The survey was conducted July 14-26, 2000. A follow-up survey of 577 of the original respondents was taken July 28-August 6, 2000. Questions have a +/- 3.7% margin of error due to sampling. Sampling error is only one form of potential error in public opinion surveys.

Data for this survey were collected from a representative sample of Americans 18 years of age or older living in the continental United States, using web-based e-mail questionnaires. The primary sampling frame was generated by Knowledge Networks using random digit dial (RRD) methods. This form of selection makes it possible to reach every American household with a telephone. The RDD sample frame was generated by Marketing Systems Group, using the GENESYS single stage RDD sampling procedure. The GENESYS RDD methodology utilizes an unrestricted random sampling process within implicit sampling intervals. The GENESYS system employs a 1+ methodology, in which working blocks of phone numbers are defined as those with at least one working residential telephone out of the 100 possible numbers in that block. Knowledge Networks employed a complex sample stratification design that varied the probability of selection based on geographical location, the number of phone lines and people in each household, and whether the phone number was listed or unlisted.

Once the RDD sample frame was generated, households were contacted by an express delivery mailing and then by telephone for enrollment in the panel. Households for whom a mailing address could not be obtained are called without a letter. Knowledge Networks made up to 25 callbacks to reach a household. All households who agreed to participate were provided with a custom-designed web device, Internet access and ongoing technical support. This allowed anyone with a telephone to be represented in the panel. No training or web experience is necessary to take part in the survey.

Through its web-enabled panel, Knowledge Networks was able to provide the Center on Policy Attitudes with the ideal interactive format for the budget exercise. Respondents were able to see how much each increase or decrease in funding for a line item affected the remaining budget or surplus. Respondents were also able to see how their budget choices compared to the actual federal budget.