Struck by Lightning by Jeffrey S. Rosenthal

Author:Jeffrey S. Rosenthal [Rosenthal, Jeffrey S.]
Language: eng
Format: epub
ISBN: 978-1-443-40167-8
Publisher: HarperCollins Canada
Published: 2005-03-17T05:00:00+00:00

10

Fifty-one Percent to Forty-nine Percent

The True Meaning of Polls

A number of national elections were held in 2004, including in Spain on March 14, in Canada on June 28, in Australia on October 9, and in the United States on November 2. The run-up to each of these elections featured large numbers of polls that attempted to gauge the opinions of the citizenry and predict the upcoming results. Most of those polls asserted various “margins of error” for their conclusions. For example:

● One month before the Spanish election, the Center for Sociological Research completed a huge survey of 24,000 Spanish voters. They predicted that the governing Popular Party would receive 42.2% of the vote, while the opposition Socialist Party would receive 35.5%. They asserted that their results had a margin of error of 0.64%.

● Two days before the Canadian election, EKOS Research Associates sampled 5,254 Canadian voters and predicted 32.6% support for the Liberals, 31.8% for the Conservatives, and 19.0% for the New Democratic Party (NDP), results they claimed “are considered accurate to within 1.4 percentage points, 19 times in 20.”

● Two weeks before the Australian election, an ACNielsen poll surveyed 1,397 voters and showed the governing coalition of Prime Minister John Howard leading the opposition Labor Party by 52% to 48%, with “a margin of error of plus or minus 2.6 percentage points.”

● Eleven days before the United States election, Reuters/Zogby surveyed 1,212 likely voters, giving George W. Bush a lead of 47% to 45% over John Kerry; they said that their margin of error was plus or minus 2.9 percentage points.

Similar claims are made all the time. What do they mean? Can polls really predict who will win an election? How closely do their results reflect the true opinions of the electorate? On what basis do they assert their accuracy or error levels? Are they really sure?

And why does it matter what the polls say, anyway?

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.