Home Page Company Clients Reports CallWeb SampleCalc EPlanCalc French

For every complex problem there is an answer that is clear, simple, and wrong.
(H. L. Mencken)

APRIL 16, 2001

Surveys under investigation: Quebec 1998

The last Quebec provincial elections, in 1998, gave a majority of seats in the House to a party which obtained fewer votes than the official opposition party. This anomaly of the single ballot uninominal electoral system was not predicted by surveys conducted during the campaign. Claire Durand, André Blais and Sébastien Vachon have recently published an article entitled A late campaign swing or a failure of the polls? The case of the 1998 Quebec election in the Spring 2001 issue of Public Opinion Quarterly. They write: "The first hypothesis put forward was that a movement toward the Liberal Party had occurred during the last days of the campaign — the "late campaign swing" hypothesis. The second hypothesis was that of a weaker turnout from PQ supporters, who were convinced that their party was way ahead. The third hypothesis was that the attribution of voting intentions to the nondisclosers was inappropriate." They conclude that "It is likely that polling methods and, most probably, factors related to coverage issues, sampling frame, and survey nonresponse explain the discrepancy." Interesting reading for those interested in survey research.

 Return to the home page
 List all news

To reach us:

General address : service@circum.com
Benoît Gauthier : gauthier@circum.com, @BGauthierCEEQ
Tel. : +1 819 775-2620, Fax : (no more fax; sorry)
238 Fleming Road, Cantley, Québec J8V 3B4

Benoît Gauthier, Mastodon