Numbers of News Stories
The total numbers of news stories on the different subjects were found by searching for “terror,” “crime,” “poverty,” and “education” in the Vanderbilt television archive for stories that ran on all networks (ABC, CBS, NBC, CNN, PBS, FOX, MSNBC, CSPAN, and CNBC) from September 11, 2001, to December 31, 2006, excluding commercials and special programs.
The Polimetrix Panel
Polimetrix maintains a panel of respondents, which it recruits through a polling website. Once individuals agree to become part of the Polimetrix panel, they provide demographic information and are offered opportunities to participate in surveys. Since recruitment into the panel is voluntary, this means that the larger panel is an opt-in sample that may be unrepresentative of the larger population. Opt-in internet samples tend to be more interested in politics as well as whiter than the general public.
To draw a nationally representative sample from a larger, nonrepresentative sample, Polimetrix uses a method called sample matching. It draws a random sample from the 2004 American Community Study run by the U.S. Census Bureau and then matches a respondent in the opt-in panel who is the closest to the census respondent based on the joint distribution of age, race, gender, and education, as well as imputed values of partisanship and ideology. The purpose of the matching is to find an available respondent who is as similar as possible to the selected member of the target sample, which results in a sample of respondents who have the same characteristics as the target sample. By matching respondents in the Polimetrix panel to those in the larger target population, Polimetrix samples become representative or close to representative.
|
