Return Of Long Form Census

[Moonbox]

Funny how the Left wants us to copy other countries when it suits their issue but whines that "we're not like other countries" when it doesn't.

While this is definitely true, it's not just the "left" that would like to see the mandatory long-form census back. It may indeed be a tempest in a teapot, but its one of those things that didn't need abolishing in the first place and there didn't appear strong justification for the move.

Edited February 5, 2015 by Moonbox

[Black Dog]

Like I said - almost all countries have stopped collecting this kind of data.

That's completely false. They might not do it in this form, but they do it.

They limit their census to population and housing because other granular data is available through other sources or individual surveys.

That are also mandatory.

We're one of the last holdouts and we still do an excellent job in providing a lot of free data. Tempest in a teapot. Funny how the Left wants us to copy other countries when it suits their issue but whines that "we're not like other countries" when it doesn't.

Perhaps the government should propose an alternative then.

[On Guard for Thee]

Then where are the studies to support your assumption? The only information I have seen is a naive assumption that if the forms are filled out they must have good data. This is a not a reasonable assumption to make.

I would say its only naive to assume that a lot of people purposely falsify such documents. What would be the point unless youre an idiot. I have a lot more faith in people than you do it would seem.

[jacee]

TimG said

" The government still has the data. "

Unh ... no Tim. It wasn't collected, remember?

.

[TimG]

I've seen no evidence the data is bad, so it's unclear why your unsupported belief should carry the day.

It could be bad - it could be good. We don't know. If it could be bad you can't argue that a mandatory census will necessarily provide more reliable data which makes hard to justify forcing people to fill it out.

Though I suspect people who actually use this data have a sense of how reliable it is, but what do the experts know?

I have seen many examples of experts using junk data to draw conclusions because that data is the only data available. It takes more than an expert claiming that they want more data to establish that more data is actually better data. Edited February 5, 2015 by TimG

[TimG]

It may indeed be a tempest in a teapot, but its one of those things that didn't need abolishing in the first place and there didn't appear strong justification for the move.

FWIW, I didn't see the mandatory census as a problem that need solving, however, the partisan posturing over this issue is way over the top. They make it sound like the world will end if we don't force people to fill out the long form even though we have no data that will tell us whether forcing people to fill out the form actually produces more reliable data. It is quite possible a volunteer census sent to a larger number of households would provide better data. Edited February 5, 2015 by TimG

[Black Dog]

It could be bad - it could be good. We don't know. If it could be bad you can't argue that a mandatory census will necessarily provide more reliable data which makes hard to justify forcing people to fill it out.

And if it could be good etc etc.

Look the only person making data reliability a sticking point is you, so the onus is on you to show there's a problem. If you can't do that, you're pissing in the wind.

I have seen many examples of experts using junk data to draw conclusions because that data is the only data available. It takes more than an expert claiming that they want more data to establish that more data is actually better data.

I wonder if you recognize the contradiction in these two statements.

[Black Dog]

FWIW, I didn't see the mandatory census as a problem that need solving, however, the partisan posturing over this issue is way over the top. They make it sound like the world will end if we don't force people to fill out the long form even though we have no data that will tell us whether forcing people to fill out the form actually produces more reliable data. It is quite possible a volunteer census sent to a larger number of households would provide better data.

Which is what they've been doing at a higher cost ($22 million more than the long form). According to the people who should know these things, it's not adequate. And again, the issue was never the reliability of the data or the efficacy of the survey.

[overthere]

The govt is well known (past and present) for going to homes and pointing a gun at the kids/pets/Grannies head to get the long form census completed.

The census takers are very persistent and use a variety of begging and threats. I've experienced both.

I tell them right off the hop that there are x number or persons living in the home, and that's it. I won't answer any questions regarding my ancestry, religion, race etc. . I answer questions about money every year on my tax form, and won't discuss it or anything else personal with a stranger on my doorstep.

They threaten prosecution but never follow through.

[TimG]

Look the only person making data reliability a sticking point is you, so the onus is on you to show there's a problem.

The law has been changed already. You are asking for it to be changed back. The onus for proof is on the person asking for change.

I wonder if you recognize the contradiction in these two statements.

Unlike you I don't take experts at their word. I ask to see the evidence that supports their claims.

[TimG]

And again, the issue was never the reliability of the data or the efficacy of the survey.

Where is the evidence that anyone has ever looked at the issue? I suspect the experts are like many others who just want data and don't want to look too deeply at whether people forced to fill out the forms will do so accurately. Edited February 5, 2015 by TimG

[TimG]

They threaten prosecution but never follow through.

I suspect the social dynamics have been permently changed by the debate and if they made it mandatory again there would be a much large number of people who, like you, tell the census people to pound salt with the knowledge of actually getting charged is pretty small.

[guyser]

54 people were referred for prosecution .

[TimG]

54 people were referred for prosecution .

http://www.tbnewswatch.com/news/302801/Not-guilty-Census-protestor-acquitted-for-not-filling-out-form

TORONTO - A Toronto judge has acquitted an 89-year-old peace activist charged after she refused to fill out the 2011 census.

In 2011, StatsCan received 13 million completed census forms, a 98 per cent response rate.

Overall, it referred 54 people for prosecution for failing to complete the mandatory census form.

2% of 13 million is 260,000. So the chance of prosecution is about 0.02%

[dre]

Out of curiosity: do you believe that information gained from torture is reliable?

ROFLMAO

[guyser]

http://www.tbnewswatch.com/news/302801/Not-guilty-Census-protestor-acquitted-for-not-filling-out-form

2% of 13 million is 260,000. So the chance of prosecution is about 0.02%

Yup

And I have a hunch why so low.......a 98 per cent response rate , meaning they dont need to go after anyone.

[-1=e^ipi]

Very dumb decision to abolish it. A few months ago I ended up having to use data from the 2006 census rather than the 2011 National Household Survey due to differences in data. Needless to say, time difference from 2006 to my other data made all my results were inconclusive (I performed many tests as to why things were inconclusive and it was due to having to use the 2006 census).

I personally cannot see why any statistician would put more weight on data that was gathered under threat than data that was volunteered.

Because it's no longer as representative of a sample..

Also, there are ways to test changes in the quality of data; you can look at changes in the variance of the responses.

How often do they just use U.S. data anyway ? Because it exists and is easier to find.....

Oh yes. I know this so well when I was trying to collect various comparable data sets from both countries in a comparative study last semester. Literally takes 20x longer to find equivalent data on Canada, if it exists.

Canada, why do you make data so hard to find?!

[TimG]

Also, there are ways to test changes in the quality of data; you can look at changes in the variance of the responses.

Can you explain how that works? How can a statistical test tell how much data is bad because people choose to enter inaccurate responses? Find it hard to believe such errors are detectable without a set of 'known to be correct' data. Edited February 6, 2015 by TimG

[-1=e^ipi]

I disagree. The long form census has many questions on things like the # of hours worked or the income earned. If someone does not have a full time work actually filling out the form correctly would take considerable effort to do the calculations. It is a lot easier to "make numbers up" or simply claim that no work was done. Also, one can omit household members to save time and the census people would be none the wiser. To get reliable data I think you need people filling out the form because they believe it is right thing to do.

Tim, I know researchers that deal a lot with income data from the census and have worked with it a bit myself. They know of these issues and there are ways of trying to account for them and to deal with them. They also acknowledge the limitations of the data.

In any case, inaccurate data is far superior to no data, provided you understand the limitations of the data.

Out of curiosity: do you believe that information gained from torture is reliable?

It is better than no information.

Edited February 6, 2015 by -1=e^ipi

[Moonlight Graham]

No doubt the stats gleamed from the census are extremely useful in making sound public policy. The long-form census is yet another dilemma in the classic political question of individual privacy rights vs the collective good. Anyone's answer to that question will come down do what they personally value more.

If the results from the census I fill out could be kept completely anonymous at all points in the process, thus my privacy secured, then I would favour doing the census and others concerned with privacy wouldn't have much of an argument. It would also need to be made sure that people fill out the forms accurately.

[-1=e^ipi]

Can you explain how that works? How can a statistical test tell how much data is bad because people choose to enter inaccurate responses? Find it hard to believe such errors are detectable without a set of 'known to be correct' data.

I don't want to go into too much detail, but I'll give you a simple example. Let's say data on # of hours worked. There is variation and uncertainty on this data due to people lying. If you take this data and control for factors that can explain hours worked (say race, gender, location, education, profession, experience, etc.) and you find that the variance of the residual is much higher for the 2011 NHS than the 2006 Census, then is a strong indication that more people are lying or people are providing less accurate data.

[TimG]

In any case, inaccurate data is far superior to no data, provided you understand the limitations of the data.

In this case, you still have data - just based on a smaller sample of the population but the data should be more accurate since volunteers have no incentive to lie or take short cuts. It is not clear to me why a smaller sample makes the data useless especially when the long form only goes to a subset of people as it is.

[Ash74]

Why are the people that are complaining (rightfully so) about the new terror bill being to invasive in our lives now complaining about not being forced to send private information to the government?

[-1=e^ipi]

In this case, you still have data - just based on a smaller sample of the population but the data should be more accurate since volunteers have no incentive to lie or take short cuts. It is not clear to me why a smaller sample makes the data useless especially when the long form only goes to a subset of people as it is.

It may be more accurate, but your sample size is smaller and it is less representative. If the non-response rate is correlated with a factor you are studying, then that can severely limit or skew results of a study.

As for smaller sample, it is very simple. If you are trying to measure say the mean income of a population of people, then the standard deviation of your sample mean is roughly inversely proportional to the squareroot of the population. So larger sample -> more confident results.

[TimG]

the variance of the residual is much higher for the 2011 NHS than the 2006 Census, then is a strong indication that more people are lying or people are providing less accurate data.

IOW - you do need a "known to be good" reference to detect bad data.