NATIONAL SURVEY

The data sets used throughout most of this book are part of the Democratic Processes project funded by the National Science Foundation (grant SES-97-09934). The telephone-survey data were collected by the Gallup Organization from mid-April to mid-May 1998. Gallup generated a random-digit-dial sample that provided equal access to all operating telephones. The standard methodology of the Gallup poll – contacting respondents with a three-call design using the “youngest male-oldest female” respondent selection procedure – was used. If a respondent was not reached in a household on the first call, Gallup called back two other times. If someone answered the phone, the interviewer asked to speak with the youngest male, eighteen years of age or older, who was at home at the time. If no male was available, the interviewer asked to speak with the oldest female, eighteen years of age or older, who was at home at the time. The average length of the interviews was 28 minutes and had a 53 percent response rate. The data have been weighted to match the sample with the population (adults in the United States, eighteen or older) based on the most recent U.S. Census. A total of 1,266 people completed the survey.

The survey items and scales used in the regression analyses in Chapters 3 and 6 were standardized to range from 0 to 1, making comparisons across variables easier. See King (1986) and Luskin (1991) for a discussion of how to interpret scales using this method of standardizing variables.