21,000 Jedi Knights:
The response to religious questions.

I support a compromise. Perhaps for different reasons than most. There is the myth of pure census data from the compulsory long form. The myth is substantiated by a kind of randomness that determines who gets the long form is evenly distributed among the populus. I think Stats Can got the randomness right. The problem is the corruption in the uptake of data.

Do we know how many Canadians intentionally lie in the census? Apparently, one group of francophone data was "highly suspect" to have been fraudulently answered for its pull of Federal dollars.

How many other deliberate untruths in the census data pool? Are people relying on that information aware of the degree of corruption?

This is an important question unanswered.

How much of the long form could be dropped in favor of Revenue Canada data. They know ages, wages, taxes, native or non-native status',locations, zip codes, employment status, disabilities,
business gross and net figures, investment levels, capital cost allowances and so on.

I wonder if Canadians in the hew and cry of this have missed this. I don't see it being discussed at any level.

Revenue Canada Data would be at least 5 times more pure than census data, for while the long form would hit 1/5 Canadians, the RevCan data would be 100% On top of that while perhaps some might lie about their taxes, the probability of coming up better data with tax returns of 21,000 Jedi Knights might be somewhat less...

Finally, Statscan does censuses once in 5 years. This data could be annual data, that stays current every single year, unlike census data quality which decays over the previous 5 years.

This sort of data should be managed by Stats Can, because they are able to assure that the information we give is entered into the data pool with no personal attachments. It would be good if the privacy commissioner looked at this as well.