The Perils of Polling: B.C. election delivers rude surprise to pollsters

The recent B.C. provincial election which delivered a shocking Liberal party victory of 50 seats(44.5% of vote) and a fourth consecutive term to just 33 seats 39.5 % of the vote) for the New Democrats should send pollsters back to their books. The turn out was extremely low just 53 % and the Greens did better than expected taking 8% of the vote and a key seat on Vancouver Island but the fact is that virtually no pollster had predicted anything like this outcome. Instead relying for the most part on on line sampling from a sample that resembles the provincial demographic they concluded that the New Democrats were on their way to a majority government. The polls were wrong. Vote splitting and a low turnout undoubtedly were factors. The combined NDP and Green vote exceeded the Liberal vote in some 12 ridings that the Liberals won. Complacency about the need to turn out and vote may have been a factor. But it should be clear now that polls that rely on on line sampling have a much larger possibility of error than those that sample the whole population in a statistically sound way. All polls have a margin of error 19 times out  of 20. In the remaining case they are simply dead wrong. Journalists, pundits and politicians often misunderstand statistics and probability. B.C. was no exception .

Advertisements

About haroldchorneyeconomist

I am Professor of political economy at Concordia university in Montréal, Québec, Canada. I received my B.A.Hons (econ.&poli sci) from the University of Manitoba. I also completed my M.A. degree in economics there. Went on to spend two years at the London School of Economics as a Ph.D. student in economics and then completed my Ph.D. in political economy at the University of Toronto. Was named a John W.Dafoe fellow, a CMHC fellow and a Canada Council fellow. I also was named a Woodrow Wilson fellow in 1968 after completing my first class honours undergraduate degree. Worked as an economist in the area of education, labour economics and as the senior economist with the Manitoba Housing and Renewal Corporation for the Government of Manitoba from 1972 to 1978. I also have worked as an economic consultant for MDT socio-economic consultants and have been consulted on urban planning, health policy, linguistic duality and public sector finance questions by the governments of Manitoba, Saskatchewan,the cities of Regina and Saskatoon, Ontario and the Federal government of Canada. I have also been consulted by senior leaders of the British Labour party, MPs from the Progressive Conservative party, the Liberal party and the New Democrats on economic policy questions. Members of the Government of France under the Presidency of Francois Mitterand discussed my work on public sector deficits. I have also run for elected office at the municipal level. I first began to write about quantitative easing as a useful policy option during the early 1980s.
This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s