Poll: 51% “Concerned With Accuracy” Of Voluntary Census

“Poll based on the responses of 1,177 Canadians. Results are considered accurate within 2.99 percentage points, 19 times out of 20.”
Related

73 Replies to “Poll: 51% “Concerned With Accuracy” Of Voluntary Census”

  1. Ham at 3:14 PM: “Let the 77% (CBC poll) who want the long census form to be mandatory volunteer to do it. For me, I’ll take the short form.”
    Uh. No. Can you imagine how biased that would be? I would gladly fill out the long form if they will ask what I want done with the CBC.

  2. The question is not whether the voluntary survey can be cured of “bias” or not. That is a red herring.
    The question is whether the government has the right to force its citizens into the inquisition of the long census form.
    In fact, I would suspect there is a real strong legal argument to say that StatsCan is acting illegally and outside its own Act in that it is using the force of the federal government to gather information for third parties who would not be able to force the information out the citizens by itself.

  3. dunbar – what the heck are you talking about?
    First, surveys aren’t ‘science’; they are a method for collecting data about X. As methods, they have inherent problems that can be, up to a recognized limit, controlled but not eliminated.
    Who is talking about correcting self-selection bias?
    Who is saying that voluntary answers are more honest? The point most of us have been making is that self-administered responses are not reliable; the answers may not have a strong validity. Whether they are mandatory or voluntary is irrelevant.
    Saying these things is indicative of ‘laziness, complacency and orthodoxy’? Kindly, using scientific analysis, explain this conclusion of yours.

  4. Alain @ 1:14 PM: “Using the law to force people to answer questions does not guarantee accurate information, as totalitarian regimes have discovered. Indeed, people quickly learn to provide the answer that is required, but that does not make it accurate.”
    And in the case of Mao’s China it led to massive famines. Farmers reported what the government told them they were supposed to grow, not what they actually harvested.

  5. Texas Canuck:
    “…voting should be a mandatory duty. Kind of a small price to pay for living in a dominion.”
    What are you willing to do to me when I refuse to vote?

  6. “What are you willing to do to me when I refuse to vote?” You are free to leave the country or pay the penalty for breaking a law. A speeding ticket so to speak.

  7. A duty to vote or fill in a census form every decade is a lot easier commitment than conscription that many countries require from their citizens.

  8. It’s an easier commitment than forced labor in salt mines as well. Are you willing to send me there if your french fries are a little bland?

  9. @ Daniel Ream
    Read Phantom’s first post, where he sets up the false dilemma between self-selection bias and false answers. He asserts (in a wimpy, non-declarative way) that its ‘either or’, and my point is that we could just as easily now have both problems. So no, that wasn’t a strawman, although it was quite an easy argument to knock down.
    As for dodging the moral/decriminalization argument, I simply don’t find it compelling and didn’t feel like discussing it. I think you can find others here who are more eager.
    I am not interested in your hair-splitting on the subject of science vs. math.
    @ET
    1) Not interested in quibbles about statistics. We all know what stats are, lets move on.
    2) The search function is your friend. Look at Phantoms first post. I think an honest reading of this thread shows that many people either think self-selection bias is not a problem or that it can be controlled for in some way. That this view is promoted by our current government is why I used the word orthodoxy.
    3) See Kevin’s first post, its phrased as a question but basically says what you claimed no-one is saying.
    4) I think its lazy to repeat the governments talking points, such as that our data will be still be useful with a voluntary form, when they are don’t stand up to even the weakest scrutiny. Others, with this thread as proof, feel differently.

  10. dunbar – with your interest in statistics and science, then, you ought to try for accuracy of reporting.
    I did not claim that ‘no-one’ says that forced data is more/less reliable. I said that ‘most of us’ (which is not logically or statistically equivalent to the syllogistic ‘all’ or ‘no-one’) are claiming that a self-administered survey is not a reliable data source. The key term is ‘self-administred’; not mandatory vs voluntary.
    Nor did Kevin say this. A question is not an assertion, by the way.
    You seem to ignore the real issues here, which is the validity of a self-administered questionnaire. The ‘mandatory’ vs ‘voluntary’ is not related to the validity of responses. It’s related to Big Government – that’s all.
    For you to claim that IF one argues a point and it happens to also be the same perspective as the govt..THEN..one is merely ‘lazy’ and ‘repeating the govt’. That doesn’t make sense. After all, I could very well happen to think of an issue and come to a conclusion, all by myself, ..and the govt’s conclusion could be similar. For you to claim that IF I agree, THEN, I am copying..is irrational.
    Do you then claim that if one disagrees with the govt, that one is ‘by your definition’ – not lazy?
    Self-selection bias is not a key issue here, for the focus has to be on the validity of the data. Even those who choose to respond can provide false data; and to declare that insisting on mandatory response (to deal with self-selection bias) also does NOT provide valid data!
    My second point is that to base a data pool on only ONE source is statistically unreliable.

  11. Dunbar said: “And as for the brilliant argument that it’s easier to correct for self-selection bias (How? Please be specific) then for false answers, this is begging the question.”
    Dunbar, PLEASE don’t tell me you can’t figure this out yourself. OMG. With the voluntary form you have a LIST of everybody who didn’t fill it out, because you have a list of everybody you sent it to. No list of fake Jedi Knights, dude.
    What’s your degree in, Fine Art maybe?

  12. Dunbar said: “I think an honest reading of this thread shows that many people either think self-selection bias is not a problem or that it can be controlled for in some way.”
    Oh ghod, this is worse. EVERY medical research project that relies on self report has to manage self-selection bias. There are many methods of doing this, the most reliable is to go back and directly observe a percentage of the sample.
    Dunbar also said: “…our data will be still be useful with a voluntary form…”
    Are you kidding me? You think the mandatory form generates better data than a voluntary one would?
    Dunbar, self report, whether voluntary or mandatory, is one of the weakest types of evidence going. Only telephone questionnaires are worse. Its the type of study one performs when trying to generate a model or hypothesis, or for marketing. A self selection based questionnaire can barely establish correlation, even when used honestly.
    Basing public policy on such flimsy evidence is -stupid-, which is clearly the case given the wealth of stupid public policy we labor under in this damn fool country. The most one can reasonably hope for from a census is to count the people in a geographic location with reasonable accuracy. All else is wishful thinking.
    The true question, which you continue to evade, is the morality of imposing these weak-@ss social engineering data collection efforts by threat of jail. You know my opinion, what’s yours?

  13. Which would have more accurate information: a mandatory response (with the threat of jail time for not complying) from 20% of the population or a voluntary response from 30% of the population?

  14. IF …
    you believe that it is right for the outcome of a vote to be applied on an individual basis based upon how one voted if the issue is MPs voting to cancel special pensions for MPs
    BUT …
    you believe that it is wrong for the outcome of a vote to be applied on an individual basis based upon how one voted if the issue is citizens voting to cancel a mandatory census
    THEN…
    you might be a liberal.

  15. Akin is trying to appear unbiased in his report. He can’t hide it. He is also attempting to copy Goldstein’s style of dissing ‘his own’…the MSM. I don’t think he is comfortable in that suit yet, but if Sun Media is giving him a chance to tread those waters, there may be hope for him.
    Once again the difference in the comments at the Sun site vs. CBC reveals the clear divide in readership.
    Socialist lefties abound on the CBC board.
    Libertarian conservatives make up most comments at the Sun.
    Dunbar…U of T? You’ve got the lingo down to a T. Bet the socialist profs loved that purple piffle.

  16. $500 Mil? That’s chicken feed for these guys.
    When I was in business I received a monthly survey from Stats Canada. LIke an obedient Canuck I filled it out for months. When I found out that Stats Canada was selling this info and that seemed to be the primary reason for doing it I rebelled. I simply started doing bad arithmetic. If the data was supposed to add up to 100% I made sure it came to 85% or 120%. AFter about 3 months the surveys stopped coming! 🙂

Navigation