Archived Seat Projections

• Little Change Following Debate
  2011-10-03
• New Survey Confirms Dead Heat
  2011-09-1426
• Liberals Slip but Still in Minority Government Range
  2011-09-20
• New Polls Confirm Liberal Lead
  2011-09-14
• Conservative Lead Erodes
  2011-09-09
• Ontario Conservatives Just Shy of Majority Numbers
  2011-09-02
• Ontario Liberal Campaign Picking Up Steam
  2011-08-18
• Conservatives Making Gains in Ontario Election
  2011-07-19
• Conservatives in the Lead for Ontario Election
  2011-06-20

New Survey Confirms Dead Heat
       
A massive new survey of 40,000 from Forum Research published in this past weekend's Toronto Star confirms other polls conducted since the beginning of September that Ontario is heading for a minority government. In fact the most recent projection which includes a blending of samples with polls taken the previous week by Leger and Angus Reid is almost identical with the Sept.20 LISPOP projection. The Forum Research poll followed an IVR robocall format which has introduced some controversy into the polling industry, because its biases are not fully understood. However as it produced a virtual tie in popular support between the Liberals and Conservatives, similar to other recent polls, these particular data don't seem to alter current estimates about the state of the Ontario electorate as we move toward the leadership debate on Sept.27. That event might prove to be the last significant opportunity to shake up the static support pattern of voters since the election was called.

Projected distribution of seats by party, released September 26, 2011

	2011 Projection	2007 Results	2007 Projection
	42	26	29
	46	71	68
	19	10	10

Note:

The "regional swing model" is more fully explained in a paper originally prepared and presented by Dr. Barry Kay to the 1990 annual meeting of the Canadian Political Science Association, entitled "Improving Upon the Cube Law: A Regional Swing Model for Converting Canadian Popular Vote into Parliamentary Seats". It should be noted that the application of the model above does not make use of the "incumbency effect" described in that paper. In tests for past elections, using late campaign polls to project electoral outcomes, the model has proved to be accurate within an average of four seats per party since 1963. Readers interested in post-dictions for past federal elections dating back to 1963, for projections using pre-election polls dating back to the 1980 federal election and for three Ontario provincial elections, may contact me at bkay@wlu.ca.