Questions
I downloaded the March 2006 PCCF from the DLI directory, and have some questions.
In the directory structure, there is a "corrected-postal-codes" subdirectory, with two files:
1) pc_corr_cp.xls - which is a list of 550 postal codes --- Is this the list of postal code corrections mentioned in the second bullet on page 5 of the codebook? Are we supposed to do something with this file, and if so, what? Is there a list of the corrections made that could advise our users, or do we simply tell them "these postal codes got fixed somehow - we don't know what was wrong, or what was corrected"?
2) problemsdpl_problemesld.txt - which has 45 lines of data, each line having a "PCODE", "DPL", and "Incorrect_DPL"
--- Is this the file of "unnecessary Designated Place - postal code linkages" mentioned in section 4.4 of the codebook?
When I look at the records in the data file, it appears that if we make the changes recorded in problemsdpl_problemesld.txt, we will have identical duplicate records for the postal codes - so I assume we are supposed to delete the records with the incorrect DPL?
I ran the duplicate checking procedures I created against the non-retired SLI=1 postal codes (theoretically the "best match" file), and came up with 48 records that were duplicated. My list of 48 postal codes includes all 45 of the ones listed in the file problemsdpl_problemesld.txt, plus postal codes V9B0A4 (DPLs 0010 and 9959), V9B6X4 (DPLs 0010 and 9959), and V9B6X5 (DPLs 0010 and 9959).
What about these 3 postal codes that seem to be duplicated - which is the "wrong" DPL (and, if I'm right in assuming, a candidate for deletion)?
In the short term, a "readme" file in the corrected-postal-codes directory would be quite useful, I think, to instruct us in the use of these two files.
3) Why isn't it possible to get this file corrected by Statistics Canada, either at the source division or at DLI, so that each DLI institituiton doesn't (or shouldn't) have to make the same corrections?
Answers
1) This file is explained on page 5 as you mentioned, it is a list of postal codes that were linked to incorrect geographic units but have now been corrected. You do nothing with this file, these corrections are in the Postal Code Conversion File (PCCF) .
2) Yes, there were some duplicate DPL linkages created during the automated geocoding process and we have discovered a few more which include the postal codes you have mentioned below.
3) - EAC) The issue with the DLPs results from the automated geocoding process. The Postal Code Project Team is working on redesigning the geocoding system in order to reduce errors and increase data quality of the Postal Code Conversion File (PCCF). The January 30, 2007 PCCF release will contain the same DPL duplicates because we are not switching over to the new geocoding system until after this last release (based on 2001 geographic units). Once we release the first PCCF based on 2006 geographic units the DPL duplicates should no longer exist because we will be using the new geocoding system.
Wednesday, November 29, 2006
APS 2001 Off-Reserve Data and Chi-Square
Question
I have a question regarding the use of the Chi-square test, this time when using the Aboriginal Peoples Survey.
The researcher notes that "One way to correct the Chi-square for a complex sample is to divide the Chi-square produced, say by SPSS, by the design effect for the sample being used. The result is only approximate, but it works well enough for sorting out what are likely significant relationships."
The answer received for the previous question suggested how to do this for the 2001 Census PUMF of Individuals. Now, he is using the APS 2001 Off-Reserve data set but cannot find a table of design effects, (or conversion factors, or quality factors or anything that looks like values that they can turn into design effect values) in the documentation tha he has for this complex sample. Has he missed something or might there be missing documentation? Do you know where this information might be, or there some other way to do the Chi-square test using the APS?
Answer
I am assuming that the researcher is referring to the PUMF, which does not contain the bootstrap weights and so it is not possible to do this kind of test using that dataset.
My best suggestion would be for the person to get access to the analytical file(s) in the RDC. Failing that, it might be possible to do it through the custom requests route.
I have a question regarding the use of the Chi-square test, this time when using the Aboriginal Peoples Survey.
The researcher notes that "One way to correct the Chi-square for a complex sample is to divide the Chi-square produced, say by SPSS, by the design effect for the sample being used. The result is only approximate, but it works well enough for sorting out what are likely significant relationships."
The answer received for the previous question suggested how to do this for the 2001 Census PUMF of Individuals. Now, he is using the APS 2001 Off-Reserve data set but cannot find a table of design effects, (or conversion factors, or quality factors or anything that looks like values that they can turn into design effect values) in the documentation tha he has for this complex sample. Has he missed something or might there be missing documentation? Do you know where this information might be, or there some other way to do the Chi-square test using the APS?
Answer
I am assuming that the researcher is referring to the PUMF, which does not contain the bootstrap weights and so it is not possible to do this kind of test using that dataset.
My best suggestion would be for the person to get access to the analytical file(s) in the RDC. Failing that, it might be possible to do it through the custom requests route.
Tuesday, November 28, 2006
Provincial Expenditures on Social Services and Health
Question
A political science graduate student is looking for historical data (back to the 1970s) for provincial governments expenditures in social services and health. CANSIM Table 385-0008 provides data from 1989 onward and Historical Statistics of Canada (H332-344. Provincial governments, gross general expenditure by function) presents data from 1965 to 1975.
Where can we find statistics for the missing period, 1976 to 1988?
Answer
Time series data prior to 1989 were removed from Statistics Canada's CANSIM database because the statistical concepts and universe coverage employed when these terminated data were produced, differed significantly from SNA 93 international guidelines. SNA 93 guidelines were implemented during the 1997 Canadian System of National Accounts (CSNA) Historical Revision and the data which Public Institutions Division currently produces and supports covering the 1989 to present period, subscribe to these new guidelines.
Unfortunately, data prior to 1989 are unsupportable and unavailable.
A political science graduate student is looking for historical data (back to the 1970s) for provincial governments expenditures in social services and health. CANSIM Table 385-0008 provides data from 1989 onward and Historical Statistics of Canada (H332-344. Provincial governments, gross general expenditure by function) presents data from 1965 to 1975.
Where can we find statistics for the missing period, 1976 to 1988?
Answer
Time series data prior to 1989 were removed from Statistics Canada's CANSIM database because the statistical concepts and universe coverage employed when these terminated data were produced, differed significantly from SNA 93 international guidelines. SNA 93 guidelines were implemented during the 1997 Canadian System of National Accounts (CSNA) Historical Revision and the data which Public Institutions Division currently produces and supports covering the 1989 to present period, subscribe to these new guidelines.
Unfortunately, data prior to 1989 are unsupportable and unavailable.
Monday, November 27, 2006
Updated Products - GSS 19
General Social Survey, Cycle 19: Time Use (2005)
The core content of time use repeats that of cycle 12 (1998), cycle 7 (1992) and cycle 2 (1986), and provides data on the daily activities of Canadians. Question modules were also included on unpaid work activities, cultural activities, social networks and participation in sports. The target population of the General Social Survey consisted of all individuals aged 15 and over living in a private household in one of the ten provinces.
FTP: /ftp/gss/cycle19-2005
The core content of time use repeats that of cycle 12 (1998), cycle 7 (1992) and cycle 2 (1986), and provides data on the daily activities of Canadians. Question modules were also included on unpaid work activities, cultural activities, social networks and participation in sports. The target population of the General Social Survey consisted of all individuals aged 15 and over living in a private household in one of the ten provinces.
FTP: /ftp/gss/cycle19-2005
Friday, November 24, 2006
Birth and Death Rates for Sackville, N.S.
Question
I have a student looking for birth and death rates for Sackville, Nova Scotia. This is a "Designated Place" according to the Census people. I can't seem to get to birth and death rates for anything other than provinces when using CANSIM. Sackville, N.S. is part
of the Halifax Regional Municipality and is served by the Cobequid Community Health Board. She can't seem to get information from HRM or the Health Board.
Any and all suggestions are welcome. I suspect it is another one of the situations where the raw data is collected, but no calculating is done or at least published on this level of geography.
Answers
1. Birth and death statistics were included in the 1996 Community Profiles. If you go to the STC website and select "Community profiles" from the left sidebar menu, you will find a link to the 1996 Community profiles in the bottom right of this page. A search for Sackville, NS results in two choices: both Subdivision C for Halifax. Taking this link will allow you to pick "Births and Deaths" from the following page.
2. Another alternative is to use Annual demographic statistics, which contains births, deaths, and population by county (=census division). From births/deaths and population, one can compute birth and death rates. Of course, It requires assuming that the birth/death rates for the county in also hold for Sackville, but it might be faster than waiting for birth/death stats to show up in the 2001 Community profiles.
3. Birth and death rates are available by county in the Nova Scotia Vital Statistics Annual Report
(http://www.gov.ns.ca/snsmr/vstat/annualreports/) .
Totals are also given for all incorporated areas (over 1000) but Sackville is not one of them.
I have a student looking for birth and death rates for Sackville, Nova Scotia. This is a "Designated Place" according to the Census people. I can't seem to get to birth and death rates for anything other than provinces when using CANSIM. Sackville, N.S. is part
of the Halifax Regional Municipality and is served by the Cobequid Community Health Board. She can't seem to get information from HRM or the Health Board.
Any and all suggestions are welcome. I suspect it is another one of the situations where the raw data is collected, but no calculating is done or at least published on this level of geography.
Answers
1. Birth and death statistics were included in the 1996 Community Profiles. If you go to the STC website and select "Community profiles" from the left sidebar menu, you will find a link to the 1996 Community profiles in the bottom right of this page. A search for Sackville, NS results in two choices: both Subdivision C for Halifax. Taking this link will allow you to pick "Births and Deaths" from the following page.
2. Another alternative is to use Annual demographic statistics, which contains births, deaths, and population by county (=census division). From births/deaths and population, one can compute birth and death rates. Of course, It requires assuming that the birth/death rates for the county in also hold for Sackville, but it might be faster than waiting for birth/death stats to show up in the 2001 Community profiles.
3. Birth and death rates are available by county in the Nova Scotia Vital Statistics Annual Report
(http://www.gov.ns.ca/snsmr/vstat/annualreports/) .
Totals are also given for all incorporated areas (over 1000) but Sackville is not one of them.
Mining FDI in Latin America
Question
A researcher here would like to get Canadian mining foreign direct investment into Latin American countries. The following CANSIM table comes closest:
Table 376-0053
International investment position, Canadian direct investment abroad and foreign direct investment in Canada, by industry and country, annual (dollars)
The closest industry is "energy and metallic minerals", but the real problem is the level of geographic detail; the only categories are US, UK, other EU countries, Japan, OECD, Japan and OECD countries, and the wonderfully evocative "all other foreign countries". Is it possible to get more detail from Statistics Canada? Alternatively, can anyone suggest a source that might provide a little more detail, especially for South America?
Answer
One could try the NRCan? (http://www.nrcan.gc.ca/mms/hm_e.htm).
They have statistics, a form to request information, and a link to industry associations (http://www.nrcan.gc.ca/mms/lien/mac_e.htm).
The Mining Association of Canada puts out Facts and Figures
(http://www.mining.ca/www/media_lib/MAC_Documents/Publications/
English/2006_FF_Eng.pdf)
which does mention investments in general - they may have more detailed information if you contact them directly.
A researcher here would like to get Canadian mining foreign direct investment into Latin American countries. The following CANSIM table comes closest:
Table 376-0053
International investment position, Canadian direct investment abroad and foreign direct investment in Canada, by industry and country, annual (dollars)
The closest industry is "energy and metallic minerals", but the real problem is the level of geographic detail; the only categories are US, UK, other EU countries, Japan, OECD, Japan and OECD countries, and the wonderfully evocative "all other foreign countries". Is it possible to get more detail from Statistics Canada? Alternatively, can anyone suggest a source that might provide a little more detail, especially for South America?
Answer
One could try the NRCan? (http://www.nrcan.gc.ca/mms/hm_e.htm).
They have statistics, a form to request information, and a link to industry associations (http://www.nrcan.gc.ca/mms/lien/mac_e.htm).
The Mining Association of Canada puts out Facts and Figures
(http://www.mining.ca/www/media_lib/MAC_Documents/Publications/
English/2006_FF_Eng.pdf)
which does mention investments in general - they may have more detailed information if you contact them directly.
SLID 2000 Immigrant Variable
Question
In the 2000 Survey of Labour and Income Dynamics there is a variable "immst15" (immigrant flag). Marginals are as follows:
1 yes 3,641
2 no 11,608
7 don't know 42,192
We're assuming that the "don't know" category should actually be "not applicable" and also that it refers to persons born in Canada. Is that a valid assumption?
Answer
Immigration status is available only for persons living in urban size of 500,000 and higher all other individuals are set to don't know for confidentiality reasons.
Follow up
Is there not some way that another code could be used that could be specifically for "confidentiality protection"? A code of "don't know" should be reserved for persons who indicate that as their response. This undermines confidence in coding.
In the 2000 Survey of Labour and Income Dynamics there is a variable "immst15" (immigrant flag). Marginals are as follows:
1 yes 3,641
2 no 11,608
7 don't know 42,192
We're assuming that the "don't know" category should actually be "not applicable" and also that it refers to persons born in Canada. Is that a valid assumption?
Answer
Immigration status is available only for persons living in urban size of 500,000 and higher all other individuals are set to don't know for confidentiality reasons.
Follow up
Is there not some way that another code could be used that could be specifically for "confidentiality protection"? A code of "don't know" should be reserved for persons who indicate that as their response. This undermines confidence in coding.
Subscribe to:
Posts (Atom)