Census thoughts on strategy

[Heather Morrison-2]

Tracey - impressive list of organizations advocating for reinstatement of the long form, and certainly many that I support.

From the perspective of someone who might be required to fill out the long form, it would be easier for me to understand the issues and support this cause if people were to speak about these good works.  Being told to fill out a form because "the experts" say so elicits a negative gut reaction from me, and I suspect this is the case for the average Jedi as well :).  On the other hand, if people tell me the reason for this exercise is so that people can understand the needs of people in our society so that we can start to figure out how to improve people's lives, this makes sense to me.

When charitable organizations say that they need this information, does this need speak to the very core of conservative philosophy, i.e.don't ask for government help, do it yourselves?

As for privacy concerns, I am wondering whether there could be additional effective arguments.  For example, while I would argue that there are privacy concerns with the census, I would also argue that these concerns are in no way addressed by making the long form voluntary rather than mandatory.  The key, to me, is breaking links between individuals and their responses, and this needs to happen with the short form.

If the government is concerned with enforcement, then devise penalties that are not excessive; also Jennifer's lottery idea is a good one.

It is good to see people gathering around this issue, just wondering how to get those without a love of data or stats per se onboard.

Best,

Heather Morrison
[hidden email]

[Tracey P. Lauriault]

see inline!

On Fri, Jul 23, 2010 at 5:07 PM, Heather Morrison <[hidden email]> wrote:

Tracey - impressive list of organizations advocating for reinstatement of the long form, and certainly many that I support.

Cheers! This is also very useful info - http://datalibre.ca/2010/07/19/uses-of-census-long-form-data-question-justification/

>From the perspective of someone who might be required to fill out the long form, it would be easier for me to understand the issues and support this cause if people were to speak about these good works.  Being told to fill out a form because "the experts" say so elicits a negative gut reaction from me, and I suspect this is the case for the average Jedi as well :).  On the other hand, if people tell me the reason for this exercise is so that people can understand the needs of people in our society so that we can start to figure out how to improve people's lives, this makes sense to me.

Thanks for sharing that!  It is helpful to understand that.  maybe I can get you to read some stuff I write and comment?

When charitable organizations say that they need this information, does this need speak to the very core of conservative philosophy, i.e.don't ask for government help, do it yourselves?

Can you help me better understand what you mean here?

As for privacy concerns, I am wondering whether there could be additional effective arguments.  For example, while I would argue that there are privacy concerns with the census, I would also argue that these concerns are in no way addressed by making the long form voluntary rather than mandatory.  The key, to me, is breaking links between individuals and their responses, and this needs to happen with the short form.

That is already done in the Census and they are very good at ensuring that that link is broken.  I do not have worries about Statcan and maintaining privacy.
 

If the government is concerned with enforcement, then devise penalties that are not excessive; also Jennifer's lottery idea is a good one.

 

That makes total sense.

It is good to see people gathering around this issue, just wondering how to get those without a love of data or stats per se onboard.

I think it is starting, but the ridiculous disinformation campaign is having some success!  And I think that is StatCan's fault for not demonstrating in a more tangible way why they are so important to canadians.

cheerio
t

Best,

Heather Morrison
[hidden email]
_______________________________________________
CivicAccess-discuss mailing list
[hidden email]
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

--
Tracey P. Lauriault
613-234-2805

[Heather Morrison-2]

responding to bits also inline


On 23-Jul-10, at 3:19 PM, Tracey P. Lauriault wrote:
>
>
> Cheers! This is also very useful info - http://datalibre.ca/2010/07/19/uses-of-census-long-form-data-question-justification/

Good stuff.  If people can write concrete examples - maybe the  
journalist group that signed on - illustrating what can, or has, been  
done with the census information, that would be even better.

For example, are the questions on mobility limitations used by anyone  
for planning services for the disabled?

>
> From the perspective of someone who might be required to fill out  
> the long form, it would be easier for me to understand the issues  
> and support this cause if people were to speak about these good  
> works.  Being told to fill out a form because "the experts" say so  
> elicits a negative gut reaction from me, and I suspect this is the  
> case for the average Jedi as well :).  On the other hand, if people  
> tell me the reason for this exercise is so that people can  
> understand the needs of people in our society so that we can start  
> to figure out how to improve people's lives, this makes sense to me.
>
> Thanks for sharing that!  It is helpful to understand that.  maybe I  
> can get you to read some stuff I write and comment?

sure
>
> When charitable organizations say that they need this information,  
> does this need speak to the very core of conservative philosophy,  
> i.e.don't ask for government help, do it yourselves?
>
> Can you help me better understand what you mean here?

Note that I am not an expert on conservative philosophy, but for what  
it's worth:

Here I am thinking of the United Way.  Is it not part of the  
conservative agenda that services should be provided by charitable  
groups like this, not the government?  If this is the case, how does  
it make sense to take away a valuable tool for these groups at a time  
of fiscal difficulty when their help will certainly be needed?  I am  
not familiar with the roles of all the supporting groups - if there  
are other groups like this that are directly providing services,  
particularly with non-governmental funds, this is what strikes me as a  
conflict with conservative philosophy.

Accountability is one thing the census can do, and that also fits with  
conservative philosophy.

>
> As for privacy concerns, I am wondering whether there could be  
> additional effective arguments.  For example, while I would argue  
> that there are privacy concerns with the census, I would also argue  
> that these concerns are in no way addressed by making the long form  
> voluntary rather than mandatory.  The key, to me, is breaking links  
> between individuals and their responses, and this needs to happen  
> with the short form.
>
> That is already done in the Census and they are very good at  
> ensuring that that link is broken.  I do not have worries about  
> Statcan and maintaining privacy.

Let's agree to disagree.

My two points on this:

1. With respect to making information available to current users of  
the database, you are correct.  However,  they have an option to make  
your information available after what - 90 years?  In order to make  
information available about individuals at any point in the future,  
they have to be maintaining information in such a way that the links  
between the answers and the answers can be re-connected.  Compare this  
with what happens at the ballot box:  they have lots of information  
about you as an individual, where you live, etc., in order to make  
sure that only those who are eligible vote, and that we only get to  
vote once.  However, we do not collect information about who voted  
which way.  This information could never be put back together.

2. My (non-expert) understanding is that at least some of the census  
work is outsourced to an American company.  No doubt the contract  
covers confidentiality, but I am not confident that it is possible to  
guarantee privacy under these circumstances (particularly in light of  
my concern above).

cheers,

Heather Morrison
[hidden email]

[Shawn Simister]

Heather Morrison wrote:

> On 23-Jul-10, at 3:19 PM, Tracey P. Lauriault wrote:
>>
>> When charitable organizations say that they need this information,
>> does this need speak to the very core of conservative philosophy,
>> i.e.don't ask for government help, do it yourselves?
>>
>> Can you help me better understand what you mean here?
>
> Note that I am not an expert on conservative philosophy, but for what
> it's worth:
>
> Here I am thinking of the United Way.  Is it not part of the
> conservative agenda that services should be provided by charitable
> groups like this, not the government?  If this is the case, how does
> it make sense to take away a valuable tool for these groups at a time
> of fiscal difficulty when their help will certainly be needed?  I am
> not familiar with the roles of all the supporting groups - if there
> are other groups like this that are directly providing services,
> particularly with non-governmental funds, this is what strikes me as a
> conflict with conservative philosophy.
>
> Accountability is one thing the census can do, and that also fits with
> conservative philosophy.
>

This is a pretty interesting angle to the debate that I haven't seen
anywhere else. It seems pretty logical that people who oppose "big
government" would support stronger charitable organizations to take the
burden off of the federal government. Unfortunately, this debate hasn't
always been bound by logic. :-)

One of the arguments that I've seen against collecting the sort of
information found on the long-form is that it is used by "special
interest groups" (which is mostly just code for immigrants and poor
people) to demand hand-outs from the federal government. I don't really
have anything pleasant to say about people who hide behind that sort of
argument so I won't bother debunking it but that doesn't change the fact
that those sentiments are still out there. Having a comprehensive list
of which organizations use the census and why they rely on it seems like
a valuable tool to help more people see the big picture.

Shawn

[Jennifer Bell]

The more I think about it, the more I think the Jedi issue is where the Conservative's heads are.  

 

It seems in the very recent US census, an anti-federalist group tried to invalidate it by getting people to abstain:

http://www.newser.com/story/85722/tea-partys-census-rebellion-is-a-bust.html

To my mind, encouraging innacurate submissions would have been a much better tactic b/c it's harder to quantify & show it didn't work.  (And it seems whether or not the campaign had an impact depends right now on what you read, I didn't find anything official/authoritative in a quick check.)

If you consider that in Canada, our anti-federalist groups tend to be region-based, and throw in the fact that some of them have an additional language barrier... the potential for misinformation campaigns is huge.

I think it's very possible that twitter *could* kill the census in Canada.

re: Alternatives to jail time... What about with-holding the GST rebate if the census isn't completed?  That would motivate the right group of people, without much additional cost or paperwork.

I agree with the line of thought that a democracy where people trust all decisions to 'experts', and don't bother to think things out for themselves, isn't a democracy [1].  So maybe that's why I'm so willing to second-guess a long list of expert agencies. 

Whatever you think of the current govt. they're a) not stupid, and b) extremely good at differentiating an 'elite' issue from a 'mass' issue, to the chagrin of the elites.

Jennifer

[1] As discussed in 'The Unconscious Civilization', by John Ralston Saul

--- On Fri, 7/23/10, Heather Morrison <[hidden email]> wrote:

From: Heather Morrison <[hidden email]>
Subject: [CivicAccess-discuss] Census thoughts on strategy
To: "civicaccess discuss" <[hidden email]>
Received: Friday, July 23, 2010, 5:07 PM

Tracey - impressive list of organizations advocating for reinstatement of the long form, and certainly many that I support.

From the perspective of someone who might be required to fill out the long form, it would be easier for me to understand the issues and support this cause if people were to speak about these good works.  Being told to fill out a form because "the experts" say so elicits a negative gut reaction from me, and I suspect this is the case for the average Jedi as well :).  On the other hand, if people tell me the reason for this exercise is so that people can understand the needs of people in our society so that we can start to figure out how to improve people's lives, this makes sense to me.

When charitable organizations say that they need this information, does this need speak to the very core of conservative philosophy, i.e.don't ask for government help, do it yourselves?

As for privacy concerns, I am wondering whether there could be additional effective arguments.  For example, while I would argue that there are privacy concerns with the census, I would also argue that these concerns are in no way addressed by making the long form voluntary rather than mandatory.  The key, to me, is breaking links between individuals and their responses, and this needs to happen with the short form.

If the government is concerned with enforcement, then devise penalties that are not excessive; also Jennifer's lottery idea is a good one.

It is good to see people gathering around this issue, just wondering how to get those without a love of data or stats per se onboard.

Best,

Heather Morrison
hgmorris@...
_______________________________________________
CivicAccess-discuss mailing list
CivicAccess-discuss@...
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

[Heather Morrison-2]

On 23-Jul-10, at 4:58 PM, Shawn Simister wrote:

>
> One of the arguments that I've seen against collecting the sort of  
> information found on the long-form is that it is used by "special  
> interest groups" (which is mostly just code for immigrants and poor  
> people) to demand hand-outs from the federal government. I don't  
> really have anything pleasant to say about people who hide behind  
> that sort of argument so I won't bother debunking it but that  
> doesn't change the fact that those sentiments are still out there.  
> Having a comprehensive list of which organizations use the census  
> and why they rely on it seems like a valuable tool to help more  
> people see the big picture.
>

Interesting - to me, what springs to mind with respect to "special  
interest groups" is for-profit corporations protecting their profits,  
or at least people simply protecting their own jobs.  Certainly not  
provinces, cities, or groups like the United Way, or poverty  
organizations, so the more information people have about who is  
supporting what, the better.

The various peoples and needs that may be served through census data  
are likely to often cross conservative priorities, I think.  It is not  
unusual for immigrants to be socially and fiscally conservative, for  
example, and I suspect that all political parties would like to have  
the newcomers with no established loyalties to Canadian political  
parties to consider them favorably.

best,

Heather Morrison
[hidden email]

[Tracey P. Lauriault]

Here is a growing list of groups that use the Census:

http://datalibre.ca/census-watch/

I was once told that women are a special interest group as are children. 

On Fri, Jul 23, 2010 at 8:12 PM, Heather Morrison <[hidden email]> wrote:

On 23-Jul-10, at 4:58 PM, Shawn Simister wrote:

One of the arguments that I've seen against collecting the sort of information found on the long-form is that it is used by "special interest groups" (which is mostly just code for immigrants and poor people) to demand hand-outs from the federal government. I don't really have anything pleasant to say about people who hide behind that sort of argument so I won't bother debunking it but that doesn't change the fact that those sentiments are still out there. Having a comprehensive list of which organizations use the census and why they rely on it seems like a valuable tool to help more people see the big picture.

Interesting - to me, what springs to mind with respect to "special interest groups" is for-profit corporations protecting their profits, or at least people simply protecting their own jobs.  Certainly not provinces, cities, or groups like the United Way, or poverty organizations, so the more information people have about who is supporting what, the better.

The various peoples and needs that may be served through census data are likely to often cross conservative priorities, I think.  It is not unusual for immigrants to be socially and fiscally conservative, for example, and I suspect that all political parties would like to have the newcomers with no established loyalties to Canadian political parties to consider them favorably.

best,

Heather Morrison
[hidden email]



_______________________________________________

CivicAccess-discuss mailing list
[hidden email]
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

--
Tracey P. Lauriault
613-234-2805

[Shawn Simister]

Jennifer Bell wrote:

I think it's very possible that twitter *could* kill the census in Canada.

Technology doesn't kill censuses. People kill censuses. :-)   Whatever apprehensions you may have about Twitter there would still need to be hundreds of thousands of actual people who decided to undermine their own government. If that happens, the long-form census will be the least of our worries.

re: Alternatives to jail time... What about with-holding the GST rebate if the census isn't completed?  That would motivate the right group of people, without much additional cost or paperwork.

The "right group of people" here would be all Canadians. I can't emphasize enough how important it is for the census to count everyone equally otherwise the results are useless. Every G20 nation holds a census. I really don't think its too much to ask for people to act selflessly once every ten years.

I agree with the line of thought that a democracy where people trust all decisions to 'experts', and don't bother to think things out for themselves, isn't a democracy [1].  So maybe that's why I'm so willing to second-guess a long list of expert agencies. 

Whatever you think of the current govt. they're a) not stupid, and b) extremely good at differentiating an 'elite' issue from a 'mass' issue, to the chagrin of the elites.

Whether it's proroguing parliament during the Olympics, changing the lyrics of the national anthem on the same day as the budget or trying to slip census reform under the noses of Canadians over the summer holidays, this government isn't half as clever as they think they are.

Shawn

[Shawn Simister]

Jennifer Bell wrote:

The more I think about it, the more I think the Jedi issue is where the Conservative's heads are.  

 

It seems in the very recent US census, an anti-federalist group tried to invalidate it by getting people to abstain:

http://www.newser.com/story/85722/tea-partys-census-rebellion-is-a-bust.html

To my mind, encouraging innacurate submissions would have been a much better tactic b/c it's harder to quantify & show it didn't work.  (And it seems whether or not the campaign had an impact depends right now on what you read, I didn't find anything official/authoritative in a quick check.)

If you consider that in Canada, our anti-federalist groups tend to be region-based, and throw in the fact that some of them have an additional language barrier... the potential for misinformation campaigns is huge.

I think it's very possible that twitter *could* kill the census in Canada.

I thought a little bit more about how this would play out from a statistical standpoint and I actually don't think that your theory holds any water whatsoever.

If we estimate that there are 12 million households in Canada and 20% or 2.4 million households get sent the long-form census. That means that 24,000 households would need to list Jedi as their religion to make just a 1% difference in the results. However, since there's no way to tell beforehand which 2.4 million out of the 12 million households will be sent the long-form census, the Jedi would actually need 1% of all households or 120,000 households to be in on the prank in order to give themselves a good shot at skewing the census.

What I missed the first time around was that this is all based on the rules of the old mandatory census. Under the new voluntary census the Jedi, Twitter users, separatists or special interest groups would actually have disproportionately MORE influence over the results.

The government has announced that they will be sending out 4.5 million forms to compensate for the fact that less people will respond to a voluntary census. Since the mandatory census would have given us 2.4 millions results we can infer that the government estimates that only about 53% of the people who receive the voluntary long-form census will fill it out. The catch is that since the Jedi are all actively trying to skew the results, their response rate should be pretty close to 100% which effectively gives them twice as much influence over the results of a voluntary census than what they would normally have in a mandatory census.

Shawn

[Jennifer Bell]

My point is that a mandatory census will likely get more Jedis than a voluntary one.

To counter your argument that it's not signifigant: if the Jedis were evenly distributed across the population, you'd be right, but that's not going to be the case.  Call me a non-statistician, but due to the nature of social networks, inaccurate responses would be concentrated in particular demographics and areas & to my thinking that seems to make the situation worse.  How can you talk about changes to the youth in Newfoundland if 10% of your youth in Newfoundland are Jedis?  How can you compare that to youth in other areas?

Heather is quite right that a lot of people don't like the commercial applications of the census data.  It was a big talking point for the Tea Party in the US:  'Don't fill out your census -- it's going to be used by marketers to send you junk mail.'  ... Which is hard to counter because it's true.  And lots of people like neither marketers nor junk mail.  Throw in the idea of incarceration if you don't answer the questions, and it seems like a public relations nightmare.

Let's think about what's changed in the last 5 years:

 1. ascendancy of social media, blogs and twitter

 2. increased weakness of central media & increased difficulty for govt's to get their perspective out there. (Apparently, not even Obama can do it. [1])

 3. increase in sensitivity of general public to 'privacy' issues due to media coverage of general privacy concerns in Facebook, etc.

 

Supporters of a mandatory long form census will have to overcome the combination of the above. 

I'm assuming lots of things here b/c I have no time right now to do proper research (and, as a new mom, I'm pretty bleary-eyed), but I still think this angle is interesting & relevant...  Feel free to point out where you think I'm wrong. 

... re: demographic based incentives -- I don't think how people are motivated to answer the questions matters, from a statistical standpoint, it just matters that enough ppl from a group answer the questions to represent that group accurately.  Though... I admit I could be really wrong on this.  If so, a $2M lottery would probably do a lot to increase participation in problem groups, and would be open to everyone.  

Jennifer

[1] recent New Yorker article I can't find on how a blog called 'Politico' sets the agenda for daily political coverage in the major papers... and it seems Obama's team has a lot of trouble getting their views in.  (I wish I could find it because it's very Noam Chomsky.)

--- On Sat, 7/24/10, Shawn Simister <[hidden email]> wrote:

From: Shawn Simister <[hidden email]>
Subject: Re: [CivicAccess-discuss] Census thoughts on strategy
To: "civicaccess discuss" <[hidden email]>
Received: Saturday, July 24, 2010, 5:54 PM

Jennifer Bell wrote:

The more I think about it, the more I think the Jedi issue is where the Conservative's heads are.  

 

It seems in the very recent US census, an anti-federalist group tried to invalidate it by getting people to abstain:

http://www.newser.com/story/85722/tea-partys-census-rebellion-is-a-bust.html

To my mind, encouraging innacurate submissions would have been a much better tactic b/c it's harder to quantify & show it didn't work.  (And it seems whether or not the campaign had an impact depends right now on what you read, I didn't find anything official/authoritative in a quick check.)

If you consider that in Canada, our anti-federalist groups tend to be region-based, and throw in the fact that some of them have an additional language barrier... the potential for misinformation campaigns is huge.

I think it's very possible that twitter *could* kill the census in Canada.

I thought a little bit more about how this would play out from a statistical standpoint and I actually don't think that your theory holds any water whatsoever.

If we estimate that there are 12 million households in Canada and 20% or 2.4 million households get sent the long-form census. That means that 24,000 households would need to list Jedi as their religion to make just a 1% difference in the results. However, since there's no way to tell beforehand which 2.4 million out of the 12 million households will be sent the long-form census, the Jedi would actually need 1% of all households or 120,000 households to be in on the prank in order to give themselves a good shot at skewing the census.

What I missed the first time around was that this is all based on the rules of the old mandatory census. Under the new voluntary census the Jedi, Twitter users, separatists or special interest groups would actually have disproportionately MORE influence over the results.

The government has announced that they will be sending out 4.5 million forms to compensate for the fact that less people will respond to a voluntary census. Since the mandatory census would have given us 2.4 millions results we can infer that the government estimates that only about 53% of the people who receive the voluntary long-form census will fill it out. The catch is that since the Jedi are all actively trying to skew the results, their response rate should be pretty close to 100% which effectively gives them twice as much influence over the results of a voluntary census than what they would normally have in a mandatory census.

Shawn

-----Inline Attachment Follows-----

_______________________________________________
CivicAccess-discuss mailing list
CivicAccess-discuss@...
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

[Tracey P. Lauriault]

It has been mandatory for 150 years jen!

On Sun, Jul 25, 2010 at 5:07 PM, Jennifer Bell <[hidden email]> wrote:

My point is that a mandatory census will likely get more Jedis than a voluntary one.

To counter your argument that it's not signifigant: if the Jedis were evenly distributed across the population, you'd be right, but that's not going to be the case.  Call me a non-statistician, but due to the nature of social networks, inaccurate responses would be concentrated in particular demographics and areas & to my thinking that seems to make the situation worse.  How can you talk about changes to the youth in Newfoundland if 10% of your youth in Newfoundland are Jedis?  How can you compare that to youth in other areas?

Heather is quite right that a lot of people don't like the commercial applications of the census data.  It was a big talking point for the Tea Party in the US:  'Don't fill out your census -- it's going to be used by marketers to send you junk mail.'  ... Which is hard to counter because it's true.  And lots of people like neither marketers nor junk mail.  Throw in the idea of incarceration if you don't answer the questions, and it seems like a public relations nightmare.

Let's think about what's changed in the last 5 years:

 1. ascendancy of social media, blogs and twitter

 2. increased weakness of central media & increased difficulty for govt's to get their perspective out there. (Apparently, not even Obama can do it. [1])

 3. increase in sensitivity of general public to 'privacy' issues due to media coverage of general privacy concerns in Facebook, etc.

 

Supporters of a mandatory long form census will have to overcome the combination of the above. 

I'm assuming lots of things here b/c I have no time right now to do proper research (and, as a new mom, I'm pretty bleary-eyed), but I still think this angle is interesting & relevant...  Feel free to point out where you think I'm wrong. 

... re: demographic based incentives -- I don't think how people are motivated to answer the questions matters, from a statistical standpoint, it just matters that enough ppl from a group answer the questions to represent that group accurately.  Though... I admit I could be really wrong on this.  If so, a $2M lottery would probably do a lot to increase participation in problem groups, and would be open to everyone.  

Jennifer

[1] recent New Yorker article I can't find on how a blog called 'Politico' sets the agenda for daily political coverage in the major papers... and it seems Obama's team has a lot of trouble getting their views in.  (I wish I could find it because it's very Noam Chomsky.)

--- On Sat, 7/24/10, Shawn Simister <[hidden email]> wrote:

From: Shawn Simister <[hidden email]>
Subject: Re: [CivicAccess-discuss] Census thoughts on strategy

To: "civicaccess discuss" <[hidden email]>

Received: Saturday, July 24, 2010, 5:54 PM

Jennifer Bell wrote:

The more I think about it, the more I think the Jedi issue is where the Conservative's heads are.  

 

It seems in the very recent US census, an anti-federalist group tried to invalidate it by getting people to abstain:

http://www.newser.com/story/85722/tea-partys-census-rebellion-is-a-bust.html

To my mind, encouraging innacurate submissions would have been a much better tactic b/c it's harder to quantify & show it didn't work.  (And it seems whether or not the campaign had an impact depends right now on what you read, I didn't find anything official/authoritative in a quick check.)

If you consider that in Canada, our anti-federalist groups tend to be region-based, and throw in the fact that some of them have an additional language barrier... the potential for misinformation campaigns is huge.

I think it's very possible that twitter *could* kill the census in Canada.

I thought a little bit more about how this would play out from a statistical standpoint and I actually don't think that your theory holds any water whatsoever.

If we estimate that there are 12 million households in Canada and 20% or 2.4 million households get sent the long-form census. That means that 24,000 households would need to list Jedi as their religion to make just a 1% difference in the results. However, since there's no way to tell beforehand which 2.4 million out of the 12 million households will be sent the long-form census, the Jedi would actually need 1% of all households or 120,000 households to be in on the prank in order to give themselves a good shot at skewing the census.

What I missed the first time around was that this is all based on the rules of the old mandatory census. Under the new voluntary census the Jedi, Twitter users, separatists or special interest groups would actually have disproportionately MORE influence over the results.

The government has announced that they will be sending out 4.5 million forms to compensate for the fact that less people will respond to a voluntary census. Since the mandatory census would have given us 2.4 millions results we can infer that the government estimates that only about 53% of the people who receive the voluntary long-form census will fill it out. The catch is that since the Jedi are all actively trying to skew the results, their response rate should be pretty close to 100% which effectively gives them twice as much influence over the results of a voluntary census than what they would normally have in a mandatory census.

Shawn

-----Inline Attachment Follows-----

_______________________________________________
CivicAccess-discuss mailing list
CivicAccess-discuss@...
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

_______________________________________________
CivicAccess-discuss mailing list
[hidden email]
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

--
Tracey P. Lauriault
613-234-2805

[Russell McOrmond-3]

Note: please don't assume that Jedi is an answer put in for someone not wanting to give an answer any more than "Christian" in Canada.   For much of my life it is the closest answer I would have answered,  given agnostic or athiest aren't quite correct.

--
Sent from my Google Nexus One
http://flora.ca

[Jennifer Bell]

Yes, but now groups who oppose individual aspects of a mandatory census (and there are many individual aspects to oppose*) suddenly have much better tools for mobilizing.  Many of these folks are not going to read any of the enlightened common-good arguments that supporters manage to get in the Globe and Mail, etc., b/c they don't read the Globe and Mail, or know anyone that does.  

If the general demographics of who answers the census shifts in 2011 because of the above, isn't that the same negative result as the switch to a non-mandatory census?

This is all a big thought experiment.  It will be interesting to see what the Conservatives actually say tomorrow.  

Best,

Jennifer

* Here's a well-expressed one, from the second comment on a CBC story:

It doesn't matter how much "greater good" can be achieved from the information if the information collected has forcibly violated anyone's privacy. If it is unethical to collect this information forcibly then it should not be done. 

Read more: http://www.cbc.ca/politics/story/2010/06/29/census-longform-scrapped.html#ixzz0upNpufjF

--- On Sun, 7/25/10, Tracey P. Lauriault <[hidden email]> wrote:

From: Tracey P. Lauriault <[hidden email]>
Subject: Re: [CivicAccess-discuss] Census thoughts on strategy
To: "civicaccess discuss" <[hidden email]>
Received: Sunday, July 25, 2010, 5:12 PM

It has been mandatory for 150 years jen!

On Sun, Jul 25, 2010 at 5:07 PM, Jennifer Bell <jenniferlianne@...> wrote:

My point is that a mandatory census will likely get more Jedis than a voluntary one.

To counter your argument that it's not signifigant: if the Jedis were evenly distributed across the population, you'd be right, but that's not going to be the case.  Call me a non-statistician, but due to the nature of social networks, inaccurate responses would be concentrated in particular demographics and areas & to my thinking that seems to make the situation worse.  How can you talk about changes to the youth in Newfoundland if 10% of your youth in Newfoundland are Jedis?  How can you compare that to youth in other areas?

Heather is quite right that a lot of people don't like the commercial applications of the census data.  It was a big talking point for the Tea Party in the US:  'Don't fill out your census -- it's going to be used by marketers to send you junk mail.'  ... Which is hard to counter because it's true.  And lots of people like neither marketers nor junk mail.  Throw in the idea of incarceration if you don't answer the questions, and it seems like a public relations nightmare.

Let's think about what's changed in the last 5 years:

 1. ascendancy of social media, blogs and twitter

 2. increased weakness of central media & increased difficulty for govt's to get their perspective out there. (Apparently, not even Obama can do it. [1])

 3. increase in sensitivity of general public to 'privacy' issues due to media coverage of general privacy concerns in Facebook, etc.

 

Supporters of a mandatory long form census will have to overcome the combination of the above. 

I'm assuming lots of things here b/c I have no time right now to do proper research (and, as a new mom, I'm pretty bleary-eyed), but I still think this angle is interesting & relevant...  Feel free to point out where you think I'm wrong. 

... re: demographic based incentives -- I don't think how people are motivated to answer the questions matters, from a statistical standpoint, it just matters that enough ppl from a group answer the questions to represent that group accurately.  Though... I admit I could be really wrong on this.  If so, a $2M lottery would probably do a lot to increase participation in problem groups, and would be open to everyone.  

Jennifer

[1] recent New Yorker article I can't find on how a blog called 'Politico' sets the agenda for daily political coverage in the major papers... and it seems Obama's team has a lot of trouble getting their views in.  (I wish I could find it because it's very Noam Chomsky.)

--- On Sat, 7/24/10, Shawn Simister <shawn@...> wrote:

From: Shawn Simister <shawn@...>
Subject: Re: [CivicAccess-discuss] Census thoughts on strategy

To: "civicaccess discuss" <civicaccess-discuss@...>

Received: Saturday, July 24, 2010, 5:54 PM

Jennifer Bell wrote:

The more I think about it, the more I think the Jedi issue is where the Conservative's heads are.  

 

It seems in the very recent US census, an anti-federalist group tried to invalidate it by getting people to abstain:

http://www.newser.com/story/85722/tea-partys-census-rebellion-is-a-bust.html

To my mind, encouraging innacurate submissions would have been a much better tactic b/c it's harder to quantify & show it didn't work.  (And it seems whether or not the campaign had an impact depends right now on what you read, I didn't find anything official/authoritative in a quick check.)

If you consider that in Canada, our anti-federalist groups tend to be region-based, and throw in the fact that some of them have an additional language barrier... the potential for misinformation campaigns is huge.

I think it's very possible that twitter *could* kill the census in Canada.

I thought a little bit more about how this would play out from a statistical standpoint and I actually don't think that your theory holds any water whatsoever.

If we estimate that there are 12 million households in Canada and 20% or 2.4 million households get sent the long-form census. That means that 24,000 households would need to list Jedi as their religion to make just a 1% difference in the results. However, since there's no way to tell beforehand which 2.4 million out of the 12 million households will be sent the long-form census, the Jedi would actually need 1% of all households or 120,000 households to be in on the prank in order to give themselves a good shot at skewing the census.

What I missed the first time around was that this is all based on the rules of the old mandatory census. Under the new voluntary census the Jedi, Twitter users, separatists or special interest groups would actually have disproportionately MORE influence over the results.

The government has announced that they will be sending out 4.5 million forms to compensate for the fact that less people will respond to a voluntary census. Since the mandatory census would have given us 2.4 millions results we can infer that the government estimates that only about 53% of the people who receive the voluntary long-form census will fill it out. The catch is that since the Jedi are all actively trying to skew the results, their response rate should be pretty close to 100% which effectively gives them twice as much influence over the results of a voluntary census than what they would normally have in a mandatory census.

Shawn

-----Inline Attachment Follows-----

_______________________________________________
CivicAccess-discuss mailing list
CivicAccess-discuss@...
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

_______________________________________________
CivicAccess-discuss mailing list
CivicAccess-discuss@...
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

--
Tracey P. Lauriault
613-234-2805

-----Inline Attachment Follows-----

_______________________________________________
CivicAccess-discuss mailing list
CivicAccess-discuss@...
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss