CALGARY, AB (PRWEB) July 26, 2013 Converge Consulting Group Inc. reports they were contacted by a spokesperson from the Communications Office of Statistics Canada earlier this week and informed that the Chief Statistician for Canada, Mr. Wayne Smith, was refusing an offer to debate the reliability of the National Household Survey (NHS) before independent statistical experts.
The refusal was in response to a challenge made by Robert Gerst of Converge to “debate the science” of the NHS before the world’s leading statistical practitioners gathering at the Joint Statistical Meetings being held in Montreal this August. According to Converge, the spokesperson indicated that a debate was unnecessary because Mr. Smith had already made his views known “through the media.”
The reliability of NHS data has become an issue in Canada because the survey is now voluntary and has experienced low response rates. Mr Smith’s predecessor, Mr. Sheikh, resigned rather than conduct a voluntary survey stating in a letter to the Canadian Prime Minister: “I want to take this opportunity to comment on a technical statistical issue which has become the subject of media discussion … the question of whether a voluntary survey can become a substitute for a mandatory census … It cannot.”
Statistics Canada began releasing NHS results earlier this year (http://www12.statcan.gc.ca/nhs-enm/index-eng.cfm). In reviewing the results, Converge concurred with Mr. Sheikh’s position, issuing warnings to clients concerning the the lack of reliability in NHS data. In May, an opinion piece written by Robert Gerst calling the NHS data ‘worthless’ appeared in newspapers across the country. Mr. Smith responded in the Globe and Mail calling Converge and other NHS critics ‘irresponsible’. This led to Gerst’s challenge to debate the issue before the world’s leading statisticians to see “who is being irresponsible and who is not.”
In a statement, Mr. Gerst said; “Refusing to debate the science tells Canadians what they need to know about the reliability of NHS data. There isn’t any. Because the NHS is based on a non-probability sample, there is no way of determining how accurate or how much bias there is in the NHS estimates. The estimates could be off by 50% or more. Statistics Canada has a mountain of very expensive data. They are pushing municipalities, health care organizations, and businesses, very hard to use this data. A scientific debate would bring attention to the fact that the data is junk.”
Comments are closed, but trackbacks and pingbacks are open.