Login  Register

Re: Freeing Census Data vs Linux Access

Posted by aph809 on May 19, 2006; 3:22pm
URL: http://civicaccess.416.s1.nabble.com/Freeing-Census-Data-vs-Linux-Access-tp602p627.html

If I may respond to my own posting!

I published an article about 10 years ago, which I also put
on the internet:

Andrew Hubbertz,"Response to Bernie Gorman".

http://library.usask.ca/~hubbertz/eip96.html

This has the data I had in mind and thought I had only in some
files at home....


Here is the salient passage.  It deals with total Statistics
Canada budget, not the Census alone:

"In 1994-95, Statistics Canada had an operating budget of $319 million. Of this,
some $44 million was revenue credited to the vote (i.e. revenues generated from
sales of products and services). According to information obtained through the
Access to Information Act, these revenues may be broken down as follows:


  Federal entities             $23,902,916
        Other levels of government     4,703,482
        Other                         15,386,351

        Total                        $43,992,749

In other words, $28.6 million in sales were to other government institutions.
Revenues generated from non-governmental sources amount to $15.4 million, or
only 4.8% of the operating budget, before the costs of marketing, legal fees
associated with license agreements, etc."


Cheers,

Andrew Hubbertz








Quoting [hidden email]:

> The principal published source for this kind of information
> is the Estimates, available on the Treasury Board Secretariat
> site at http://www.tbs-sct.gc.ca/tb/estimate/estime.html.
>
> The site seems to be down this morning.  I assume this will
> be rectified shortly.
>
> I compiled this information for the 1996 Census, relying upon
> the Estimates and information obtained using the Access to Information
> Act.  I have the details at home, but I have a pretty good
> recollection that it runs something like this:
>
> Cost of the 1996 census:  $310 million
>
> Revenues from sales:  $45 million
>
> Using the AIA, I found that only $15 million of that $45 million came from
> non-governmental sources, i.e. 'new' money.  Health Canada, for example,
> might
> pay Stat Can to put a few questions into the long form of the
> Census.  Other government institutions purchase data products,
> or customized tabulations.
>
> Don't quote those numbers.  I will check on them over the weekend.
>
> Keep in mind the added costs entailed in selling:  marketing,
> monitoring security to prevent 'data leakage', etc.  If the data
> is free, these costs are not incurred.
>
> I have not compiled this data for the 2001 census, and of course
> data on the 2006 census won't be available for several years.
>
> Regards,
>
> Andrew Hubbertz
>
>
>
>
> Quoting "Tracey P. Lauriault" <[hidden email]>:
>
> > Anyone have leads on how we can find out how much stat can earns from
> > selling data?
> >
> > Could we estimate if the cost of selling is more expensive than just
> > giving it away?
> >
> > TpL
> >
> >
> > Olivier Charbonneau wrote:
> > >> due to the $1000 per mini-set price tag.
> > >>
> > >
> > > Does anyone know how much revenue StatCan generates from its access
> > licenses ?
> > >
> > > I'm ready to bet that Access Copyright and Copiebec generate more profit
> > from
> > > the *interest* running on the unallocated reproduction licenses collected
> > from
> > > Canadian Universities, schools and gvmt :)
> > >
> > > Maybe some of that money could go to "liberate" StatCan data and help
> > finance
> > > Opne Access initiatives... see:
> > > http://www.fedcan.ca/english/advocacy/openaccess/
> > >
> > > We should use the interest on these amounts (which sould not have been
> > created
> > > if the market were efficient) to fix the market's market
> > failures(unallocated
> > > repro fees means that we paied for content and the money has not found an
> > > author). Besides, the idea is that if authors have an incentive to give
> > works
> > > away in the first place, we should use money left over to set this
> process
> > up.
> > >
> > > Any thoughts ?
> > > Olivier
> > >
> > >
> > > Quoting Cory Horner <[hidden email]>:
> > >
> > >
> > >> There seems to be a little bit of confusion here... 2 different
> > >> discussions regarding the Census happening simultaneously.
> > >>
> > >> 1) Making Census data more freely available
> > >> 2) Census 2006 on-line submission form for Linux users
> > >>
> > >> We're effectively beating a dead horse on the Linux issue... People
> > >> complained, StatsCan initially denied their claims were relevant, but
> > >> soon gave in.  Horray!  Threads about 1) seem to keep getting hijacked
> > >> by 2)...  The petition proposed is related to 1)... not 2) !
> > >>
> > >> The issue actually under discussion is getting the *results* from the
> > >> 2006 Census out to the public -- not the generic "the population of
> > >> nunavut is x", but the raw anonymized data, with which an unknown wealth
> > >> of knowledge exists but sits unused due to the $1000 per mini-set price
> > >> tag.
> > >>
> > >> Glancing at the Statistics Act, it seems the StatsCan policy on cost
> > >> recovery has no basis in the legislation.  I'd like to see us aim to get
> > >> a new section (3f) added to the act, which would add a responsibility
> > >> for StatsCan to disseminate all data freely.
> > >>
> > >> Cory.
> > >>
> > >> _______________________________________________
> > >> CivicAccess-discuss mailing list
> > >> [hidden email]
> > >>
> http://civicaccess.ca/mailman/listinfo/civicaccess-discuss_civicaccess.ca
> > >>
> > >>
> > >
> > >
> > > _______________________________________________
> > > CivicAccess-discuss mailing list
> > > [hidden email]
> > > http://civicaccess.ca/mailman/listinfo/civicaccess-discuss_civicaccess.ca
> > >
> > >
> >
> >
>
>
> Andrew Hubbertz
> Librarian Emeritus
> University of Saskatchewan Library
>
> 613 692 2709
> [hidden email]
>
>
>


Andrew Hubbertz
Librarian Emeritus
University of Saskatchewan Library

613 692 2709
[hidden email]