Browse Prior Art Database

Method of Conducting a Political Opinion Poll

IP.com Disclosure Number: IPCOM000020236D
Original Publication Date: 2003-Nov-04
Included in the Prior Art Database: 2003-Nov-04
Document File: 1 page(s) / 5K

Publishing Venue

IBM

Abstract

Public opinion polls are commonly used before elections to see how a political candidate is doing in each geography, to fine-tune the remainder of the campaign. Current art for conducting a public opinion poll is labor intensive and expensive. What is needed is a more automated way to measure the pulse of public opinion on a candidate or an issue.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 88% of the total text.

Page 1 of 1

Method of Conducting a Political Opinion Poll

    An autodialler with a recorded message from a candidate (or other widely recognizable public figure associated with the candidate) is used to telephone selected households within a geography. While the ostensible purpose of the message is to persuade the voter to vote, explain the candidate's platform, etc., a secondary purpose is to measure the public's reaction to the candidate and the messages by measuring "listen-time".

Initially, a benchmark is established for how long the typical person listens to the message before hanging up. This will account for the variation where the sample necessarily includes people who never listen to annoying phone calls regardless of content, children who answer the phone, etc. versus those who are willing to listen to a message and may be somewhat interested in the content.

At intervals during the campaign, the autodialling process is repeated to randomly-selected phone numbers within a geography and the "listen time" compared with the benchmark to compute a score. The score indicates whether the public opinion is becoming more or less favorable, and to what degree.

The correlation between listen-time and voter sentiment can also be calibrated through the use of standard telephone interviews such as the Gallup Poll, Harris Poll, etc.

In a variation, the set of households phoned can be preselected according to demographics, such as female voters, ethnic voters, high-income voters, etc...