A very bad idea, the 2011 census long form will be voluntary

Tonington

Hall of Fame Member
Oct 27, 2006
15,441
150
63
So what's so difficult about feeding data into bank where the names are automatically "blacked" out?

How many times are we going to go over this? To come close to replacing the census data we have now means you need complete records from all levels of government. That means you need to employ people in all offices to put the data into files electronically. And you still wouldn't have all the information contained in the census.

So your plan would actually cost more money, make government larger, and give us less information. It's not so easy to replace a census.
 

Tonington

Hall of Fame Member
Oct 27, 2006
15,441
150
63
Why a Mandatory Census is Necessary


One talking point I have heard a fair bit the last few days is: "There are a lot of voluntary surveys out there that seem to be accurate. Are you suggesting they're all flawed?"

The answer is that those voluntary surveys rely on a mandatory census and as such, the removal of the census will weaken the value of those voluntary surveys.

Below the fold I'll describe why making the census voluntary will reduce the value of voluntary surveys and lead to too many baseball diamonds being constructed.

I grew up in the east end of London, ON in the 1980s and back when I was a kid my neighbourhood was made up partly of 1st generation/immigrant kids (mostly from Poland, Italy or Portugal) and partly of nth generation Canadian WASPy kids like me.

Let's suppose it's the year 1984 and the city is planning on building a new community centre. The city has a limited piece of land and have a limited budget, so they're trying to figure out how many soccer fields to put in vs. how many baseball diamonds vs. "is it worth the space and expense to have an outdoor pool"? The city commissions a polling company to conduct a (voluntary) poll of the area to determine the optimal mix for the residents. The polling company believes that given the demographics of the area simply calling up a few hundred households and taking an average may end up being misleading because:

  1. Different country-of-origin groups may respond at different rates, either by not picking up the phone when called or refusing to participate in the survey.
  2. Different country-of-origin groups may have different needs for the new facility. For instance, soccer may be more popular with some groups than with others.
From my childhood experiences in the neighbourhood, I can guarantee that WASPy families will end up being over-represented in the unweighted survey and baseball will appear to be more popular than it actually is.

The polling company is smart, though, so in the survey they not only ask about patterns of use for sports and activities, they also ask each person their country of origin. Suppose it turns out that 8% of the survey respondents are from Poland, but we know 11.7% of the neighbourhood is Polish immigrants. We can correct for this by weighting the responses of each Polish immigrant higher, so the total responses for the 'Polish' part of the survey has a weighting of 11.7% - identical to the weight they make up in the community.

But this only works if we know the actual proportion of the Polish community is, in fact, 11.7%

How do we know that proportion? From the long-form census! Of course, the government isn't talking about getting rid of the long-form census, just making it voluntary. But remember point one from before:

  1. Different country-of-origin groups may respond at different rates, either by not picking up the phone when called or refusing to participate in the survey.
With our voluntary census, we'll have the same problem we had with the voluntary survey - that Polish immigrants are underrepresnted. But with the survey, we could re-weight the results against the mandatory census. But in the case of the voluntary census, we have nothing to re-weigh the results against!

Note that this problem does not go away if you increase the number of people you voluntarily survey. If Polish immigrants are less likely to respond to the survey relative to guys named Moffatt, then you'll get the same results if you send the survey to 20% of the population or 30% of the population.

Unless the government is planning on putting questions such as race, ethnicity, income and religion into the mandatory short-form census, we're going to have a really difficult time doing any kind of planning in heterogeneous neighbourhoods, as shown by this example.
 

taxslave

Hall of Fame Member
Nov 25, 2008
36,362
4,340
113
Vancouver Island
How many times are we going to go over this? To come close to replacing the census data we have now means you need complete records from all levels of government. That means you need to employ people in all offices to put the data into files electronically. And you still wouldn't have all the information contained in the census.

So your plan would actually cost more money, make government larger, and give us less information. It's not so easy to replace a census.

No, I think what JLM means is that the census could and probably should be done without names or addresses of the participants. I basically believe that the less governments of all levels know about me the better off I am. I try really hard to do my taxes honestly but that is all the information I am willing to give to an office full of government employees that I have no idea of how honest they are. Its not that I have anything to hide that I am aware of, It is just on a need to know basis and no one in Ottawa needs to know. A lot of people where I grew up simply do not trust governments.
 

Tonington

Hall of Fame Member
Oct 27, 2006
15,441
150
63
No, I think what JLM means is that the census could and probably should be done without names or addresses of the participants.

Names sure. Places, nope. Did you read the example I posted above? Without addresses, we have no data at the community level.

You know there is a reason that the backlash is so swift against this right? You may not see it, but the census data is used by a lot of different organizations, all the time. It's important data to have. The Romans were collecting census data before Christ walked the Earth. A census makes planning far more comprehensive, and far more accurate.

Ask yourself this question, are you happy paying taxes for government waste? The census is the data departments use, the data which is used to allocate budget funds. I'm a budget hawk too, and I like to see the purse strings in Ottawa as tight as possible. Without reliable census data, it will be very difficult to fund programs efficiently. Inefficiency drives me nuts.

The government won't even know anything about you. CSIS doesn't even get access to personally identifiable census data held by Stats Canada. The privacy commissioner has had no damning reports on the practices Stats Canada uses. There is prudent caution, and then there is just plain paranoia.

What the government is doing is whipping up the paranoia.
 

Tonington

Hall of Fame Member
Oct 27, 2006
15,441
150
63
Tony Clement today, on accountability for the decision to make the long form a voluntary survey:

That was a government decision, no question about it. I'm not trying to suggest otherwise," he said. "We made that decision, not StatsCan. I've made it quite clear that StatsCan would have been quite happy to move along with the status quo.

No, Tony Clement actually was clearly stating that it was Stats Canada that suggested a voluntary survey, which is patently absurd. No statistics organization on the planet would suggest such a thing.

The opposition hopefully will fix this mess when Parliament is back in session. Obviously our elected government doesn't know what they're doing...
 

Nuggler

kind and gentle
Feb 27, 2006
11,596
141
63
Backwater, Ontario.
:-(I was watching the parliamentary committee "grill" Antonio Clemente, and the chair asked him directly why he lied about the head of Stats Can supporting the demise of the long form.

HE NEVER ANSWERED THE QUESTION: He rambled on about being concerned about privacy, blah blah blah blah, but never answered the question.

!!!!8O Here's the kicker. The committee NEVER CALLED HIM ON NOT ANSWERING!!!

Which leads me to believe what I have all along. The whole thing's a farce. Maybe a bit of cross party collusion here, as the members of the "committee" never know when they'll get caught out, and have to be in the hot seat. Guaranteed they won't get treated as gently by the cons as they treated Antonio. (with his shiny green suit)

:glasses11:And it is duly noted that the "fighter jet multi billion acquisition non fiasco" has slipped from the front page and from the radar.

How did that feel, Canada. You like it.?? More to come.

At this very moment, Stephano, and Antonio are probably slurpin some vino, readin the bible and greasin up for another go at us.

badda bing badda boom.
 

Tonington

Hall of Fame Member
Oct 27, 2006
15,441
150
63
The Opposition tried to get him to give straight answers, but the chair of the committee was actually encouraging Clement's obfuscation.
 

Omicron

Privy Council
Jul 28, 2010
1,694
3
38
Vancouver
They need the census for just a little longer.

Once cash is eliminated the govt can buy corporate databases which know what type of toothpaste you buy and what your gas mileage is.
Which means, city planners and provincial budget ministers will know exactly what kind of toothpaste you use, and what your gas mileage is, but won't have a clue how to zone your neighborhood.

The London Economist ranked Stats Canada as the best information gathering service of any of the G8. Specifically, they called it the "Cadillac" of national statistics services. Stats Canada does such a good job that some of their information is payed for by other nations, i.e., the US buys it's studies on the effects of various industrial contaminants on the environment (a study which no corporation will sponsor).

Harper wants to neuter Stats Canada because it makes the effects of his puppet-masters' policies too easy to see the damage of.

He's not the first right-wing PM to try that. Mulroney tried to neuter Stats Canada too.

Why is it that left-wingers and middle-of-the-roaders always want more information out there - the more extensive, accurate and comprehensive the better - whereas right-wingers always want it held in and controlled close to their chests?

What's odd is that governments, any government, whether it be left (NDP), centrist (Liberal), right (the now-defunct Progressive Conservatives), or fringe-right (Conservatives, which is just the Reform Party under a new name) need accurate "snap-shot" information to do their job, which is to Govern.

Since Harper wants to neuter Stats Canada, it means he doesn't think he needs accurate information to do his job, which means he has no intention of actually Governing... it means he wants to RULE!
 

Tonington

Hall of Fame Member
Oct 27, 2006
15,441
150
63
Clementonomics

Toronto Star: "Industry Minister Tony Clement says opponents of the Conservatives' decision on the census are just whining because they once had a “good deal” to get information they needed while letting Ottawa force citizens to supply the data."

Let's suppose you owned a bakery and you believed your bread prices were too low for whatever reason (margins are too low or having low priced goods doesn't mesh with your bakery's value proposition, etc.).

Those of us who work in the private sector could think of a variety of solutions. You could raise your prices. You could look for lower cost suppliers of raw ingredients. You could offer the bread only as a bundle with a higher margin item. You could even stop selling the product altogether.

If you're Tony Clement, you go out and buy a 30 million dollar machine that inserts broken glass into the bread. When your customers complain about the broken glass, call them a bunch of freeloaders. After all, they did benefit from years of inexpensive bread. After awhile, you won't have any more annoying customers to deal with.

Maybe that logic makes sense on Parliament Hill. But as one of those tax-paying private sector 'freeloaders', I have to admit I'm scratching my head.
 

taxslave

Hall of Fame Member
Nov 25, 2008
36,362
4,340
113
Vancouver Island
Tonington: Still trying to sort all this out. I know that you favor mandatory census because it provides important data. In post #142 if I understand it correctly it would appear that you manipulate the data from your hypothetical survey based on what you think you know about the population of the area to give the answer you want. How do you know that your Polish citizens actually want soccer as opposed to baseball. Perhaps they want baseball to try to fit in better?
In #144 You seem to agree that names could be voluntary but not places. I am willing to tell them that I live in Qualicum Beach and possibly even postal code since it covers a fairly large area but not to provide my name or residential address. That is something that I feel they do not need to know. Under these conditions I would honestly answer the long form. Otherwise I will either refuse to fill it out or lie on it. Either way it will skew their data. Would that not give the data needed without compromising my identity?
 

Tonington

Hall of Fame Member
Oct 27, 2006
15,441
150
63
You don't manipulate data, you weight the sampling protocol. Samples of the Canadian population are often taken using about 1000 respondents. If you don't weight the sampling, then the probability of a skewed sample is high. Using the census data, pollsters and scientists can weight their surveys so that the population is evenly sampled. Without proper weighting, you may end up sampling, for example, no french speakers. Or you might get too many french speakers. Without proper weighting, you need huge sample sizes to avoid sampling bias.

So, now that our census is moving towards voluntary reporting, some groups will be under-represented, and others over-represented. The USA tried voluntary census, as an experiment. To fix the data would have been hugely expensive, and the data quality still would have suffered.

If we don't have a real census, that is a real snap shot of Canadian society, then those surveys can't be weighted properly.

The polish example is just that, it's a teaching example. But there are very real parallels in Canadian society.
 

Omicron

Privy Council
Jul 28, 2010
1,694
3
38
Vancouver
I am willing to tell them that I live in Qualicum Beach and possibly even postal code since it covers a fairly large area but not to provide my name or residential address. That is something that I feel they do not need to know.

Why?

What are you afraid of, given that if Ottawa really wanted to track you down in order to tyrannize and suppress you, all the information they need for that is in the CRA database?

The only thing the long-form data does is provide accurate information to a government seriously intent upon actually trying to Govern.

A government would *not* care about accurate data if it has no intention of Governing, rather, if its intention is to Rule... forcing whole populations to do things one way or another regardless of who the individuals are and what their wants and needs as a people might be.

Furthermore, if the government's intention is to Rule rather than Govern, then not only is accurate information about the society irrelevant, it get's in the way, because it reveals too much about how damaging are the effects of their Rule.

Don't get me wrong, I think they should have information on the public, but some of the questions got far to personal for my tastes.
I've never filled out a long-form, so I'm curious... what questions did they ask that were far too personal for your tastes?
 
Last edited:

taxslave

Hall of Fame Member
Nov 25, 2008
36,362
4,340
113
Vancouver Island
Why?

What are you afraid of, given that if Ottawa really wanted to track you down in order to tyrannize and suppress you, all the information they need for that is in the CRA database?

The only thing that long-form data does is enable a government seriously intent on trying to actually Govern the information they need in order to make decisions based upon the situation they are Governing.

They only way a government would *not* care about accurate data is if they have no intention of Governing, rather, if their intention is to Rule, forcing whole populations to do things one way or another regardless of who the people are and what their wants and needs might be.

Furthermore, if the government's intention is to Rule, and not just Govern, then not only is accurate information about the society irrelevant, it get's in the way, because it reveals how damaging the effects of their Rule are to the happiness and well-being of individuals and to the population as a whole.

As I stated before where I grew up most people did not trust the government so part of it is inherited. Other than that there is a lot that bureaucrats simply do not need to know. Like my ethnic background, what, if any religion, how many pets. That is irrelevant to running the country. Nor do they need to know how many bathrooms anyone has. The less information the government has on us the better. If they want to know if I work they can just call CCRA. All the information they require is on the short form and it does not need a name and address attached to be accurate since according to Tonington's explanation they manipulate that data anyway.I don't trust any government employee not to sell all that info to some marketing firm either.
 

Tonington

Hall of Fame Member
Oct 27, 2006
15,441
150
63
All the information they require is on the short form and it does not need a name and address attached to be accurate since according to Tonington's explanation they manipulate that data anyway.I don't trust any government employee not to sell all that info to some marketing firm either.

I never said they do manipulate data, I said that they would have to manipulate the data to remove the bias if they move to voluntary census forms. Weighting a sample is not manipulating data...

And all the information they require is not on the short form. The workplace survey, used to assess trends in employment, is weighted using the data in the long form. Do you think it's wise for a government to have no knowledge of emerging trends in the working population of Canada?
 

taxslave

Hall of Fame Member
Nov 25, 2008
36,362
4,340
113
Vancouver Island
I never said they do manipulate data, I said that they would have to manipulate the data to remove the bias if they move to voluntary census forms. Weighting a sample is not manipulating data...

And all the information they require is not on the short form. The workplace survey, used to assess trends in employment, is weighted using the data in the long form. Do you think it's wise for a government to have no knowledge of emerging trends in the working population of Canada?
I don't see any difference between weighting and manipulating data. Two words for the same thing. To be accurate everyone would have to fill out the long form. Partly because of the disparity in population between cities and rural areas and their idea of regions. Data collected in Victoria cannot be considered accurate when applied to the rest of Vancouver Island. Today I filed my EI even though I work steady like most loggers I never know from one week to the next if I will still be working. At the bottom is a link to the job bank with two jobs I am qualified for and what they think is in my community. One is in Victoria and the other in Tofino. I live in QUalicum Beach.(Both are 2hrs+drive) They think Vancouver Island is one region even though it is nearly the size of England. Also I know that even employment data is skewed since when your EI runs out you are no longer considered unemployed even if you are still not working.
There is still no need to have names and addresses on the long form even if it is mandatory to fill out. So they could have made life easy just by eliminating that portion of it.
 

Trex

Electoral Member
Apr 4, 2007
917
31
28
Hither and yon
So, apparently in 2011, the long-form will not be mandatory. Previously, the form was sent to approximately a fifth of Canadian households of which the response was mandatory. Now it will be replaced by more forms, only the response will be voluntary.

This is not good news.

Some might find this a trivial technical matter, but it's really not. Having access to good population information is critical for legislators, and having good access to bad information makes it far easier for governments to manipulate the information we do have.

It's well known in sampling literature that people with low income and lower education have lower response rates than those with higher income and higher education. So making it voluntary to fill out the census is perverse, in that the lower income/educated citizens will be systematically under-sampled, and when law makers craft new legislation, when R&D considers market trends, when NGO's analyze planned policies, the impact in that segment of society will be unknown. Or rather, the result won't be an accurate appraisal of what will happen.

Consider this, the census is the only data Canadians have access to on aboriginal academic achievement. How will we be able to make objective assessments of policy alternatives?

All sample surveys that are not mandatory have bias, but it can be corrected, if you know what the true population distribution is. When you have a complete census, you know what the distribution is. So it's really shooting Statistics Canada in the foot by hamstringing what data they have.

This really is an assault on democracy, but because of the poor statistical/mathematical understanding of Canadians, this won't get turned into the issue that it needs to be. At least not by Joe and Jane Public.

The government tries to frame this issue by relating it to privacy concerns. The truth is, that the researchers who use this data never see it. They send their estimation codes to StatsCan, just like the recent HST model simulations in BC. StatsCan just runs the inputs the researchers are interested in, and gives them the results. At no time would anyone see the actual data, except for the professionals at StatsCan.

Basically, the census form is no more an invasion of privacy than is filing a tax return. Anyone who has dealt with StatsCan knows that they are compulsively obsessive about protecting this information. It's almost annoying how protective they can be.

If you still think this is not a big issue, consider this. Canada has a well-known productivity gap. I've yet to read of a single economist that doesn't think that Canada needs to make improvements that address the knowledge sector of our economy. That is where countries with good productivity are making those gains. If we don't have good population data, we don't have good access to good data, and we make it more difficult for:
-Investors
-Marketing
-Policy makers
-Academics

So basically everyone involved in the knowledge economy.

This decision was obviously made without consultation from the relevant stakeholders in the Canadian public. It also represents further movement towards secrecy in Ottawa, despite pontification to the likewise.

This was actually tried before, by Brian Mulroney. The backlash from the business community was severe. That put the kibosh to that plan. Let's hope that good sense prevails again.

I disagree.

Keep in mind that all the Conservatives want to do is decriminalize the Federal census.
It is a purely Libertarian move.
The census itself remains the same.

Why on earth should we use the authorities to harass and threaten our citizens over a census.
At tax payer expense no less.
And a very, very important point to be made here is future governments can include or delete any questions they see fit.
That means any future Federal governments be they Conservative, Liberal, NDP, Green or whatever the future holds politically can rejig the questions.
And then what; fine or incarcerate our citizens for failing to answer possibly offensive or improper questions?
Again lets keep in mind the census and all mandatory completion enforcement costs are at tax payers expense.

For what, good data?
For whom?
Since my wife works at a reasonably high level within the research university system and also happens to be a registered Federal lobbyist I have seen some of the letters the research Universities are firing off to the Federal Government.
Which is no big deal, it's all public knowledge.
Its all pretty much along the same lines as Tonington's posts except more factually accurate.
The Uni's of course have a huge vested interest.
Research at Canadian Universities is Federally funded.
Except for the funding they receive when they get in bed with the private sector.
So all that census data (collected at taxpayers expense) flows to the research universities who either discover something and then patent it or sell the data or the research to the private sector.
So the Federal government is using taxpayer moneys to collect census data at gunpoint.
They then sell off the data to the private sector for a profit or give the data to research groups who then sell anything discovered of value to the private sector.
In effect your taxes are being used to enforce data collection which could then be used to create a more potent cigarette or a more marketable haemorrhoid cream.

Now lets talk about the indigenous peoples.
The old if we don't have an enforceable census then they become underrepresented.
I really don't buy that.
I think its a flawed and erroneous argument.
We make them criminals for not answering a census?
Have the authorities harass them for not completing it to their satisfaction?
Then what, incarcerate them?
Its the old "we have to kill them to save them" logic.
Poor people, uneducated people and indigenous peoples have the right to free will.
If they don't want to answer the Fed's intrusive questions then more power to them.

In WW2 the Nazis used the census data in Germany and Austria to locate and round up Poles, Slavs, Jews and Gypsies(Romanians).
Then they killed them.
Then they went through the census data to try and search out the mentally handicapped or the physically disabled with an inheritable trait.
Then they sterilized them.
Or killed them.
As soon as the Nazi's invaded a country the SS went straight for the local census data.
People then got rounded up and loaded on trains.
Lots and lots of people.

Back home in Canada the authorities went straight to the census data to round up the Japanese-Canadians and the German-Canadians.
Off to the internment camps they went.
Down south the Americans did the same thing.
Built the camps and then hit the census data.
So its pretty well established how handy accurate census data can be.

Next up to bat are the countries who have decided to get rid of mandatory (enforceable) census data.
Norway (highest GDP/PP in the world), Sweden( that hotbed of rightwing control freaks), Denmark( rated the happiest country in the world) and Iceland (the greatest density of fantastically hot chicks on the planet) have all punted the mandatory census.
They feel its both invasive and intrusive.

France, Germany and the UK are all at present debating getting rid of the mandatory census.
Pretty much everyone agrees the harassment and enforcement issues do not justify the results.

And just how good is the data anyway?
In this day and age of RIFID's and data tracking every time you make a purchase without using cash.
All your insurance data, your tax data, your health information and your every purchase is sold to the highest bidder and collated and massaged daily.
Census data is however generally considered "stale" data.
Quite often the census data is just to dated to be very useful in this age of mobility,daily surveys and internet marketing.

I really don't care if the Fed's are going to pimp out my census data to heave into the great data collection pot.
But to make my tax dollars pay for it?
And then to enforce mandatory compliance, again at tax payers expense?

Trex
 

Omicron

Privy Council
Jul 28, 2010
1,694
3
38
Vancouver
As I stated before where I grew up most people did not trust the government so part of it is inherited.
Me-thinks you're confusing Government with Ruler-ship. Your ancestors probably came from a part of the world controlled by Rulers (aka Tyrants), and life under the Rulers was so bad that they taught you that "government" is a bad thing.

But for a whole bunch of neuro-sociological scientific facts about human nature that I can get into if you want, we can't operate on any level of organization without some form of "government". Crumb... even organizations as small as two people - i.e. a married couple - have to work out rules over how final decisions for their two-member group's actions are going to be made.

The issue isn't to eliminate government... you can't... not unless you're going to be a perfect monastic hermit (and by "monastic" I mean in its original sense, derived from the Monastic Age, wherein a solitary individual would live away from other humans, subjecting himself to the invisible government of his Deity, which means, even then he hadn't eliminated government; he just replaced it with an imaginary one; monasticism is to government like what kids having imaginary friends is to friendship).

Rather, the issue is to make sure the government you have is one that functions however is the way that's best for the people being governed, and *not* just for the interests of the people doing the governing.
Other than that there is a lot that bureaucrats simply do not need to know. Like my ethnic background, what, if any religion, how many pets. That is irrelevant to running the country.
Huh? That's *exactly* the kind of information governments need to know if they're going to be able to run the country; *especially* in a multicultural society... and by govern, I mean just manage things and maintain peace and order, which is what they're there for.

For example, suppose you've got a neighborhood that's become predominantly Sikh, who have a traditional mid-summer holiday where for one day per year they blow off lots of fireworks, and suppose the community they moved into has a historical ban on fireworks because their yahoo kids could only see fireworks as a fun way to make lots of trouble.

You could *wait* for the Sikhs to do their celebration and start shooting off fireworks, whereupon the surrounding community does a knee-jerk reaction by sending in the cops and triggering a riot, which is almost guaranteed if attacking someone's faith.

Riots are idiotically expensive. In terms of cost, they are not much different than fighting door-to-door combat in a war, regardless of which side predominates, and war is the second-most expensive thing we do.

(American studies say that the only thing more expensive than war is to maintain long-term embargoes. Evidently their own numbers say that it would have been economically cheaper for them to have launched an all-out invasion of Cuba than to have maintained their embargo, but they couldn't get around something that Fidel had on them, plus look at the cost of the Cold War - an embargo on the Soviet Union - it cost eight trillion 1982 US dollars, which was much more than if they'd just flown in with B-52 bombers and nuked the place and then rebuilt New York and LA and Chicago after those had been destroyed with the retaliating Soviet missiles... from the money-boys' point of view, that darned aspect of democracy that keeps putting the value of human life ahead of profits is a real nuisance.. but I digress...)

Anyway, after the riot, it leads to hatred on both sides, which means more expensive policing costs to keep everyone apart, plus it damages the economy, because now the communities are not doing business with each other. Just as it is healthier for assorted nations in a global economy to trade with each other, so also is it healthier to an internal economy for different communities to be doing business with each other, so the national economy is hurt, and it's costing more to maintain the peace.

OR someone at city-hall could have downloaded the latest reports from Stats Canada and said:

"Hey Tony, look at this. The percentage of Sikhs in the south-east has topped 23%, and trend-lines from the previous years show's it's growing at the rate of 4% per year.

"Tell Fred in public works to put enough in his budget for an extra fire-hall in that district, and tell him to make sure there's budget to cover the over-time cost for keeping those extra firemen on alert for the mid-summer Sikh fireworks show. It means fireworks sales are going to go up, so we'll cover the cost of the new fire-hall from the 5% tax on fireworks.

"It also says they're planning to build two new temples. Tell Shiela in Community Relations to find out who the temple leaders are going to be and have her explain to them the painful fines they're going to be hit with if we have to put out any fires started by their little party.

"The stats also say that Chinatown has upped the importation and retailing of fireworks 20% over their own needs, so tell Charlie to tell Chinatown to start adding Hindi to the labeling on the fireworks packaging or we're going to crank the tax on fireworks from 5% to 7%, and while he's at it have him explain how we're going to crank the tax on fireworks past the moon if they import any more of those cheep brands that blow up at the wrong time in order to cover the medical cost of stitching people's hands back on!"
[...] how many pets [...]
Staffer: "Hey Mr. Mayor, the south-central clinic is reporting a very strange jump in the number of people allergic to flea-bites coming in, and we're getting low on the medicine. Plus they're reporting three cases of buboes!"

Mayor: "Hmm, so that means there's a flea infestation down there, which means either there's an explosion in the rat population because nobody's keeping pets, or it means that people are keeping enough pets, but there's a shortage of pet flea-medication. Which is it?"

Staffer: "We don't know. Stats Canada stopped asking people how many pets they have."
Nor do they need to know how many bathrooms anyone has.
Staffer: "Mr. Premier, there's an ongoing outbreak of echoli infections in the east central part of the Island. It's killing the shell-fishery and we're getting low on antibiotics."

Premier: "Oh criminy, are those Lower Slobovian families and their twelve kids per family still doing that thing where instead of adding some plumbing and a new toilette they have their kids crap in the ocean when someone's monopolizing their one bathroom? If so, we can offer a home-owners incentive if they'll just upgrade their plumbing. That will be good for the hardware retailers who can speak Slobovian."

Staffer: "We don't know."

Premier: "Huh? Why not?"

Staffer: "Stats Canada stopped asking people their ethnic origin and how many toilette's they have."
The less information the government has on us the better.
Actually, if you're afraid of a Tyranny, they don't need to know *anything* about you. They just load up some flats with guys holding machine guns, and they drive around shooting anything on two legs. As Napoleon noted, the less the troops know about who they're shooting the better when it comes to using government force against its own people.
I don't trust any government employee not to sell all that info to some marketing firm either.
Penalties for a Stats Canada employee selling information to a marketing firm are the same as they are for a CCRA employee selling information, or using that information for personal gain.

Do you have *any* idea how serious breaches like that are taken, and how stiff the penalties are?

To put it as nicely as I can... Canada has the softest velvet glove on the planet, but if it ever pulls that glove off and starts clenching the titanium fist, you do *not* want to be the target... and violation of CCRA and/or Stats Canada privacy regulations are on that gloves-off level.
 
Last edited:

Tonington

Hall of Fame Member
Oct 27, 2006
15,441
150
63
I don't see any difference between weighting and manipulating data.

Well, there is no data to manipulate when you decide how to conduct your survey. It can't be manipulating data if there is no data available.

The data is still randomly selected. That bias is covered. And properly weighted data reduces bias in your sampling procedure. Every survey has a methodology. In the methodology you describe how you collected your data.

A famous example of not weighting data is the Presidential Election of 1936. A magazine called Literary Digest conducted their own straw poll. They mailed out surveys to subscribers (whose disposable income was well above the national average. They also based their polling on records from registered motor vehicles and telephones.

In 1936, having a telephone and an automobile made you upper class. So the poll that Literary Digest conducted was heavilly biased in their sampling methodology towards the upper class. Their poll results showed that Alf Landon (R) was going to win the Presidential election with 370 electoral votes. What actually happened, was that Franklin D. Roosevelt (D) won the most lopsided election in the history of the United states, by electoral college vote count. Alf Landon won only 8 electoral votes, while Roosevelt won 523.

This was the year that a household name (would become a household name) in polling, George Gallup, of Gallup polls had conducted a scientific survey (weighted appropriately) and he predicted correctly a Roosevelt win, as well as the spectacular failure of the Literary Digest poll.

It's not manipulating data to weight your sample to a known quantity, where possible when the quantity is known. It's ensuring the best quality data is collected.

To be accurate everyone would have to fill out the long form.

Not true. Accuracy improves as sample size increases. Twenty percent of Canadian households filled out the long form. That is sufficiently large enough to use in weighting for other samples.

Partly because of the disparity in population between cities and rural areas and their idea of regions.

The census is distributed randomly. It does distinguish between urban and rural.

Data collected in Victoria cannot be considered accurate when applied to the rest of Vancouver Island.

No, and it isn't.

The census itself remains the same.

No it doesn't. They've also removed questions. So that's not really the same. And since it's voluntary, the results won't be the same. It's going to cost more money, and it's going to produce poorer quality data.

Why on earth should we use the authorities to harass and threaten our citizens over a census.

You might have a point, but the short form is still mandatory, as is the agricultural census. So they haven't stopped harassing have they? They've only made one of the data products that businesses, researchers, municipalities, NGO's, and Federal departments use to make informed decisions.

Do you like inefficiencies in Ottawa with your money? I don't. If that's what you want, great, yell it loud and clear.

At tax payer expense no less.

Yes, and this census is going to cost more money...for less worth. That's a great bargain....

And a very, very important point to be made here is future governments can include or delete any questions they see fit.

They could do that before.

That means any future Federal governments be they Conservative, Liberal, NDP, Green or whatever the future holds politically can rejig the questions.

Which they could do all along. You're not the first one to notice; there's a movement to make Stats Canada arms length.

For what, good data?
For whom?

These groups, for starters:


  • Alberta Professional Planners Institute
  • Access Alliance Multicultural Health and Community Services
  • Alliance canadienne des personnes retraitées
  • Ancestry.ca
  • Anglican Church of Canada / Église anglicane du Canada
  • Anne Johnston Health Station
  • Association des statisticiennes et statisticiens du Québec
  • Association du Barreau canadien / The Bar Associate of Canada
  • Association canadienne d’économique / Association canadienne des économistes
  • Association des Soeurs du Canada
  • Association canadienne francaise pour avancement de science (ACFAS)
  • Association féminine d’éducation et d’action sociale (AFEAS)
  • Association francophone pour le savoir (Acfas)
  • Association of Canadian Archivists (ACA)
  • Association of Canadian Map Libraries and Archives (ACMLA) / Association des cartothèques et des archives cartographiques du Canada
  • Association of Educational Researchers of Ontario
  • Association of Municipalities of Ontario / Association des municipalités de l’Ontario
  • Association ontarienne des chercheurs et chercheuses en éducation au ministre Clement (AERO)
  • Association of Public Health Epidemiologists in Ontario (APHEO) / Association ontarienne d’épidémiologie et desanté publique
  • Association des statisticiennes et statisticiens du Québec (ASSQ)
  • Atlantic Provinces Economics Council / Conseil économique de province de l’atlantique
  • BC Chamber of Commerce
  • Black Creek Community Health Centre
  • BC Non Profit Housing Association
  • Bloc Québécois
  • Burnaby Family Life
  • C.D. Howe Institute
  • Caledon Institute of Social Policy / Institut Caledon des Politiques Sociales
  • Calgary and Red Deer City Planners
  • Canada Census Committee
  • Canadian Centre for Policy Alternatives (CCPA) / Centre canadien pour des politiques alternatives
  • Canada West Foundation
  • Canada Without Poverty Advocacy Network
  • Canadian Alliance of Student Associations (CASA) / Alliance canadienne des étudiants (CASA)
  • Canadian Anthropology Society / Société canadienne d’anthropologie (CASCA)
  • Canadian Association for Business Economics (CABE) / Association canadienne des économistes d’affaire
  • Canadian Association of Geographers / Association canadienne des géographes
  • Canadian Association of Journalists / Association canadienne de journalisme
  • Canadian Association of Midwifes (CAM)
  • Canadian Association of Public Data Users (CAPDU) / Association canadienne des usagers de données publiques
  • Canadian Association of Research Libraries (CARL) / Association canadienne des librairies de recherche
  • Canadian Association of Social Workers (CASW)
  • Canadian Association of University Teachers (CAUT) / Association canadienne des professeurs d’université
  • Canadian Chamber of Commerce
  • Canadian Conference of the Arts
  • Canadian Council on Social Development / Conseil canadien du développement social
  • Canadian Economics Association
  • Canadian Evaluation Society / Association canadienne d’évaluation
  • Canadian Federation of Demographers / Association canadienne des démographes
  • Canadian Federation of Francophone and Acadian Communities / Association canadienne des francophones et des acadiens
  • Canadian Federation of Humanities and Social Sciences / Fédération canadienne des sciences humaines et sociales
  • Canadian Federation of Independent Business / Fédération canadienne des entreprises indépendante
  • Canadian Historical Association / Société historique du Canada
  • Canadian Housing and Renewal Association
  • Canadian Institute of Actuaries / l’Association canadienne des actuaires
  • Canadian Institute of Transportation Engineers
  • Canadian Institute of Planners / Fédération canadienne des urbanistes
  • Canadian Jewish Congress / Congrès Juif Canadien
  • Canadian Labour Congress / Congrès canadien des travailleurs
  • Canadian Library Association
  • Canadian Marketing Association / Association canadienne de marketing
  • Canadian Medical Association Journal
  • Canadian Network of Metropolis Centers / Réseau canadien des centres Metropolis
  • Canadian Nurses Association / Société des infirmières du Canada
  • Canadian Population Society / Association canadienne de population
  • Canadian Public Health Association / Association canadienne de santé publique
  • Canadian Research Data Network Centre / Réseau des centres de données de recherche
  • Canadian Society for Epidemiology and Biostatistics (CSEB) / Société canadienne d’épidémiologie et de statistiques
  • Canadian Sociology Association / Association canadienne de Sociologie
  • Canadian Union of Public Employees (CUPE) / Syndicat Canadien de Fonction Publique (SCFP)
  • Canadian Urban Institute / Association canadienne de développement urbain
  • Canadian Women’s Foundation
  • Capital Regional District (in B.C.)
  • Mel Cappe, former Clerk of the Privy Council
  • Catholic Council of Bishops
  • Central Toronto Community Health Centres
  • Centre francophone de Toronto
  • Centre for Study of Living Standards / Centre de recherche pour niveau de vie
  • Centre interuniversitaire québécois des statistiques sociales / CIQSS-QICSS / Quebec Inter-University Centre for Social Statistics
  • Chinese Canadian National Council
  • Cities Centre – University of Toronto Research Institute
  • Citizens Engaging Democracy, Newmarket-Aurora
  • City of Calgary
  • City of Edmonton
  • City of Fredericton
  • City of North Vancouver
  • City of Ottawa
  • City of Red Deer
  • City of Toronto
  • City of Victoria
  • Co-operative Housing Federation of Canada / Fédération Canadienne des Coopératives de Logement
  • Committee of Presidents of Statistical Societies
  • Commission des droits de la personne et des droits de la jeunesse
  • Commissariat aux langues officielles
  • Community Development Halton (Ontario)
  • Community Development Council Durham
  • Community Foundations of Canada
  • Community Social Planning Council of Greater Victoria
  • Confédération des associations étudiantes de l’Université Laval (CADEUL)
  • Conseil consultatif sur la condition de femme du Nouveau-Brunswick
  • Concordia Student Union
  • Conference Board of Canada
  • Conference des Lecteurs et Principaux des University de Quebec / Conference of Rectors and Principals of Quebec Universities (Association of Universities in Quebec)
  • Conférence régionale des élus (CRÉ) de Laval
  • Conseil permanent de la jeunesse (CPJ)
  • Conseil des agences servant les immigrants
  • Conseil Québécois des Coopératives et des Municipalités
  • Conservative MP James Rajotte
  • Daily Bread Food Bank 9Toronto)
  • Davenport Perth Community Health Centre
  • Département de démographie of Université de Montréal
  • District of Nipissing Social Service Admin Board
  • Don Drummond; former chief economist of TD bank, former ADM of Finance
  • Edmonton Journal, Editorial
  • Environics Analytics
  • Evangelical Fellowship of Canada / Alliance Évangélique du Canada
  • Fair Share Peel
  • Family Service Association of Toronto
  • Research, Evaluation and Planning
    Family Service Toronto
  • Federation of Post-Secondary Educators of BC
  • Fédération canadienne de démographie
  • Fédération des associations étudiantes du Campus de l’Université de Montréal (FAÉCUM)
  • Federation des chambres de commerce du Quebec
  • Fédération des travailleurs et travailleuses du Québec
  • Federation of Canadian Municipalities / Fédération canadienne des municipalités
  • Fédération étudiante universitaire du Québec (FEUQ)
  • Fédération Québécoise des Professeurs et Professeures d’Universités
  • Dr. Robin Fitzgerald, Research Fellow, Key Centre for Ethics, Law, Justice and Governance, Griffith University, Canadian Scholar Downunder
  • Four Villages Community Health Centre
  • Ivan Fellegi, Former Chief Statistician, Statistics Canada
  • French Language Services Commissioner of Ontario / commissaire aux services en français de l’Ontario
  • Glendon School of International and Public Policy
  • Globe and Mail Editorials
  • Stephen Gordon, economist Université Laval
  • Government of Nunavut
  • Frank Graves, EKOS Research (polling)
  • Greater Halifax Partnership
  • Greater Victoria Community Indicators Network
  • Green Party of Canada / Parti vert du Canada
  • Grey County
  • Halifax Chronicle-Herald, Editorial
  • Halton, Region of
  • Hamilton’s Settlement and Integration Services Organization
  • Hamilton Roundtable for Poverty Reduction
  • Hamilton Community Foundation
  • The Hill Times editorial
  • Alex Himelfarb, former Clerk of Privy Council
  • Imagine Canada
  • Information and Communications Technology Council
  • Institut de statistiques Quebec / Statistical Institute of Quebec
  • Institute for Research on Public Policy
  • Inuit Tapiriit Kanatami / Association nationale Inuit (du Canada)
  • JJ McCullough
  • Kevin Milligan, economist at University of British Columbia
  • Laval, ville de
  • Liberal Party of Canada / Parti Libéral du Canada
  • Lumina Research Valuation and Advisory Services
  • Marketing Research and Intelligence Association (MRIA) / Association de la Recherche et de Intelligence Marketing (ARIM)
  • Roger Martin, Rotman School of Management
  • Martin Prosperity Institute
  • Medical Health Officers Council of Saskatchewan
  • Metcalf Foundation
  • Metropolis British Columbia
  • Milton, Town of
  • Mike Moffatt
  • Mississauga Council
  • Nanos Research (polling)
  • National Council of Women of Canada (NCWC)
  • National Post Editorial
  • National Specialty Society for Community Medicine
  • National Statistical Council ( acts in a consultative capacity for StatsCan) (French Statement) / Association statistique du Canada
  • Nature International Editorial
  • New Democratic Party of Canada / Nouveau Parti Démocratique du Canada
  • New Heights Community Health Centres
  • North Bay Parry Sound District Health Unit
  • North Western Ontario Municipal Association
  • Official Language Commissioner
  • Ontario Chamber of Commerce
  • Ontario Council of Agencies
  • Serving Immigrants
  • Ontario deputy finance minister Peter Wallace
  • Ontario Human Rights Commission (OHRC)
  • Ontario Non-Profit Housing Association (OPNHA)
  • Ontario-Municipal Social Services Association (OMSSA)
  • Ontario Professional Planners Institute
  • Ontario Public School Boards Association
  • Opportunities Waterloo Region
  • Ottawa Citizen Editorial
  • PARC (Toronto)
  • Peel, Regional Municipality
  • Peel Poverty Action Group (PPAG)
  • Peterborough’s medical officer of health
  • Planning Council of Cambridge and North Dumfries (Cambridge,Ont.)
  • Blake Poland, Associate Professor, Dalla Lana School of Public Health, University of Toronto
  • Poverty Free Halton
  • Valerie Preston, director of CERIS research centre on immigration and settlement issues York University
  • Prentice Institute at University of Lethbridge
  • Province of Manitoba
  • Province of New Brunswick
  • Province of Ontario
  • Province of Ontario – Office of Francophone Affairs
  • Province of Prince Edward Island
  • Province of Quebec
  • Public Service Alliance of Canada (PSAC)
  • Quebec Community Groups Network
  • Research Centre on Digital Inclusion / Centre de recherche et d’expérimentation sur l’inclusion numérique (CREIN)
  • Regent Park Community Health Centre
  • Regional Planning Commissioners of Ontario
  • Registered Nurses Association of Ontario
  • Richard Florida, University of Toronto
  • Residential and Civil Construction Alliance of Ontario (RCCAO) / Alliance de la construction résidentielle et civile de l’Ontario
  • Roman Catholic bishops
  • Royal Society of Canada
  • Rural Ontario Institute (ROI)
  • St. Joseph’s Health Centre
  • St. Stephen’s House
  • Saskatchewan Students’ Union (USSU)
  • Munir A. Sheikh, Former Chief Statistician of Canada
  • Martin Simard, laboratoire LERGA, Département des sciences humaines et CRDT, Université du Québec à Chicoutimi,
  • Social Planning Council of Kitchener-Waterloo
  • Social Planning Council of Sudbury
  • Social Planning Network of Ontario
  • Social Planning Toronto
  • Social Planning Council of Winnipeg
  • Social Policy in Ontario
  • Société franco-manitobaine
  • South Riverdale Community Health Centre
  • SPARC BC (Social Planning and Research Council of BC)
  • StarPhoenix [Saskatoon ] Editorial
  • Statistical Society of Canada
  • Statistics Canada Advisory Committee
  • Statistics Canada Advisory Committee on Demographic Statistics and Studies / Comité consultatif sur les études et les statistiques démographiques de Statistique Canada
  • Sudbury Star, Editorial
  • Surrey Board of Trade (BC)
  • Syndicat des employés internationaux unis (SEIU)
  • Table régionale des organismes communautaires autonomes en logement de Laval (TROCALL)
  • Tasha Kheirridin
  • Toronto Association for Business Economics
  • Toronto Board of Trade
  • Toronto Immigrant Employment Data Initiative (TIEDI)
  • Toronto Public Health / Directeur de santé publique de Toronto
  • Toronto Star Editorial
  • Toronto Social Research and Data Consortium (30 health, community and multi-service social agencies)
  • Toronto Women’s Housing Co-op
  • Town of Milton
  • Town of Smith Falls
  • Transportation Association of Canada / Conseil du transport urbain du Canada
  • Ukrainian Canadian Congress / Congrès Canadien Ukrainien
  • United Church of Canada
  • United Steelworkers
  • United Way of Canada
  • United Way of Kitchener-Waterloo and Area
  • United Way of Greater Simcoe County
  • United Way Toronto
  • Université de Toronto
  • Urban Futures
  • Vancouver Board of Trade
  • Volunteer Toronto
  • Darryl Walker, President of the BC Government and Service Employees’ Union
  • Wellesley Institute
  • West Hill Community Services
  • West Toronto Support Services
  • WoodGreen Community Services
  • Women’s Hands Community Health Centre
  • York Community Services

Research at Canadian Universities is Federally funded.

And they still have to pay to use the census data. Publicaly funded universities and publicly funded research grants spending money on data already collected by another government agency.

Back home in Canada the authorities went straight to the census data to round up the Japanese-Canadians and the German-Canadians.

So lets just make all gun ownership illegal while we're at it?

France, Germany and the UK are all at present debating getting rid of the mandatory census.

At least they're having debates.

Pretty much everyone agrees the harassment and enforcement issues do not justify the results.

According to you? Pretty much everyone?

And just how good is the data anyway?

Better than it will be with voluntary reporting.

Quite often the census data is just to dated to be very useful in this age of mobility,daily surveys and internet marketing.

Quite often? Over 7000 articles in the National Library of Medicine using census data. It's useful to a great many groups. Why do you think that list I put up top with all those organizations is so long? Just because you don't use it, don't be so arrogant as to call it useless.

But to make my tax dollars pay for it?

It's going to cost more.

And then to enforce mandatory compliance, again at tax payers expense?

See what happens if you don't fill out the short form.
 

Omicron

Privy Council
Jul 28, 2010
1,694
3
38
Vancouver
I never said they do manipulate data, I said that they would have to manipulate the data to remove the bias if they move to voluntary census forms. Weighting a sample is not manipulating data...
I heard a story that, about ten years ago, Nelson, as in the people who do the Nelson rating system, had a problem related to that.

The Nelson Rating System was (and probably still is) *so* key to the American economy that it could *not* be allowed to fail.

Their numbers determined viewership of TV programs, and who was watching those programs, which would determine everything from how much advertising on those programs would cost, and which retailers would advertise on which programs, which would ripple back to determine the cost of products on store shelves, plus their numbers would determine which programs would get run, which would determine the careers of Hollywood stars and the welfare of Hollywood, etc. etc. yadda yadda. The criticalness and value of the Nelson numbers was incalculable.

They were running on old iron, because they had something that worked, and, because the numbers generated by that system were *so* important, they couldn't afford to upgrade without being 100% certain that the new system would be valid, so they dragged their heals on upgrading, plus they outsourced maintenance of their computing iron to some third-party who were supposed to be experts at keeping old systems going.

They didn't. They got sloppy with their backups.

One day the old disk-drives containing recent Nelson Ratings data crashed, and there were no recent backups.

Lawsuits from the broadcasting networks and from all advertisers and from the Hollywood community would have bankrupt Nelson into the next universe, and passing the lawsuits onto the third-party outfit to whom the system maintenance had been outsourced would have been pointless, because they were a puny little fish that was *supposed* to provide diligent, quality service, but which had not.

A very, very, very confidential deal was made between parties at the core wherein the company that was supposed to maintain the systems would get them running again (which meant some serious work on their part running around the nation scrounging for old disk drives without raising eyebrows as to why anybody would want them) while Nelson put its statisticians to task for a very special project:

Specifically, perhaps they wouldn't know how many people and of what demographic had been watching TV program-X on channel-Y, but could they interpolate from the last twenty years of general over-all viewing patterns who-and-how-many would *probably* be watching those programs.

For six weeks - the amount of time it took to get the systems running again - Nelson and the economy of the US ran on interpolated data - data that ultimately was counting on changes in demographic patterns of TV viewing going more-or-less according to patterns they'd seen in the past - and with each passing week, the odds of that interpolated data being accurate got thinner and thinner.

Nelson knew they couldn't push it more than twelve weeks, after which the advertisers themselves might start noticing, from quarterly reports, inconsistencies in sales versus what Nelson had told them people were watching.

If corporate and media-entertainment America had ever found out what Nelson had been doing, Nelson would be *gone* today...

... and it's *that* kind of interpolation which Harper is telling Stats Canada to perform, without the long-form data.

No wonder the head of Stats Canada resigned over the decision, for it is more than an issue of professional integrity...

He *had* to resign, in order to cover his ass from future Inquires, which *will* happen, because someday an important decision is going to be made based upon post-Harper Stats Canada data, and that data is going to turn out to be utterly invalid, with disastrous consequences!
 
Last edited: