Friday, February 9, 2018

KLEMS

Question:
The KLEMS data is pretty historic now.  Will the database ever be updated?

Answer:
Subject matter directed me to the following CANSIM tables with the message “KLEMS is now available in CANSIM:

http://www5.statcan.gc.ca/cansim/a26?lang=eng&id=3830032&p2=33
http://www5.statcan.gc.ca/cansim/a26?lang=eng&id=3830021&p2=33

Thursday, February 8, 2018

Missing Values in NHS Individuals PUMF


Question:
For the WAGES variable in the NHS individuals PUMF, why are "not available" and "not applicable" not declared as missing values? The codebook states that:

"The value 8,888,888 stands for not available. The value 9,999,999 stands for not applicable and is applied to all persons aged less than 15 years"

Leaving those values (which constitute over 17% of the cases) subject to calculation would seem to skew the results rather significantly, would they not?


Answer:
B Estimation
Note: Users must refrain from publishing unweighted estimates and from conducting analyses based on unweighted data from the file because the unweighted results do not represent the population but
only describe the sample. They must also make sure to exclude values of study variables that are not applicable or not available from their calculations because those values might be considered as valid observed values by the statistical software when they are not. For example, values such as 9,999,999 or 8,888,888 for a numeric (or quantitative) variable would be interpreted as valid observed values but should be considered as nominal values indicating these values are not usable in estimation.

Example 4:

We want to estimate the average total income of women aged 15 years and over living in Ontario who
have an income. In the calculation of the numerator, WEIGHT is multiplied by the value of the 'total income' variable for individuals with an income (where TOTINC ^= 8,888,888, TOTINC ^= 9,999,999, TOTINC ^= 0) whose gender is female (SEX = 1) and who are aged 15 or over (AGEGRP ≥ 6, AGEGRP ^= 88) in the province of Ontario (PR = 35); the results are then totalled. To estimate the average, the numerator (or estimated total income) is divided by the sum of WEIGHT for individuals satisfying the same conditions on TOTINC, SEX, AGEGRP and PR.
The result obtained is: $179,154,359,345 / 5,072,260 = $35,320, which means the average total income of women aged 15 and over living in Ontario who have an income is around $35,320.
B.2.b.3 Estimator of a ratio
A ratio can be defined as the division of two amounts, which could be two totals or two averages

Example these counts are included
8888 Not available (unweighted) 13,676 - (Weighted) 483,013
9999 Not applicable Canadian citizens by birth and non-permanent residents (unweighted) 697,600 – (Weighted) 26,060,226

PUMF for Survey on Disability 2017

Question:
A researcher has asked when (or if) there will be a public use microdata file for this product (http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=3251).


Answer:
CSD will not be getting any PUMFs in the future – The reason is due to the fact that the majority of the variables would need to be suppressed and/or aggregated for confidentiality reasons and would not be of analytical value. The geography would likely have to be at the highest level (Canada).

Wednesday, February 7, 2018

Number of Employees


Question:
I have a student who is interested in # of employees (full time, part time or self-employed) for: City of North Vancouver, District of North Vancouver, District of West Vancouver and the Arbutus/UBC corridor in the smallest possible geography. He's not interested in category of employee or employer characteristics--just the number of employees.

He would like the data in a format that could easily be exported in ArGIS or QGIS.

Answer:

The 2016 Census offers “Number of employment income recipients”, that is the total number of people who received employment income during the year. It does not break them out by full, part-time, self employed. But using Vancouver CMA Census tracts seems like it might meet your requirement. You should be able to identify the Census Tracts you need for Arbutus and West.

Follow-Up Answer:

This is a custom table request that has to be assessed by the nearest regional office. I am not sure of this the Arbutus/UBC corridor as a geography or a format that exports to ArGIS or QGIS.

They can download Canada, provinces, territories, census divisions, census subdivisions and dissemination areas (British Columbia only) from the Census profile page

or

Census Profile, 2016 Census
http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/prof/search-recherche/results-resultats.cfm?Lang=E&TABID=1&G=1&Geo1=&Code1=&Geo2=&Code2=&SearchText=Vancouver&SearchType=Begins&wb-srch-place=search

This table has the variable Class of Worker (Employee –Self employed).
http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/dt-td/Rp-eng.cfm?LANG=E&APATH=3&DETAIL=0&DIM=0&FL=A&FREE=0&GC=0&GID=0&GK=0&GRP=1&PID=110647&PRID=10&PTYPE=109445&S=0&SHOWALL=0&SUB=0&Temporal=2017&THEME=124&VID=0&VNAMEE=&VNAMEF=

Friday, February 2, 2018

Local Government Expenditure Data

Question:

A student here is working on a project related to charitable giving using Canadian data, and interested in government tax expenditures on education, environment, social service, health, recreation and culture at the local level with particular year of 2004, 2007, 2010, and 2013. She's found CANSIM table 3850001 (1989-2009) and 3850041 (2008-present), but the two tables use different accounting methods to calculate the figures, and therefore can't be joined.

Are definitions of what is included under education, environment, social service, health, recreation and culture available?

Is there any chance of getting the data calculated on a consistent basis for all the years in question? Assuming there is, is it available at some sub-annual frequency?

Another problem is that provincial and local data are always consolidated; any chance of getting the data only for local governments?

Another question re: local government: does it always include aboriginal governments, or not?

Answer:

Are definitions of what is included under education, environment, social service, health, recreation and culture available?
For table 385-0001, references can be found under the archived Financial Management System documentation (http://www5.statcan.gc.ca/olc-cel/olc.action?objId=68F0023X&ObjType=2&lang=en&limit=0) 
For table 385-0041, PSSD does not currently have a list of the definitions your client is looking for. She may be able to find the definitions she is looking for in the 2014 GFS Manual found here: https://www.imf.org/external/Pubs/FT/GFS/Manual/2014/gfsfinal.pdf

Is there any chance of getting the data calculated on a consistent basis for all the years in question? Assuming there is, is it available at some sub-annual frequency?
As your client pointed out, the CANSIM tables that she was looking at use two different accounting methods. The FMS was founded on a modified-cash based system of accounting (eg., CANSIM Table 385-0001). Statistics Canada has moved from a modified-cash based accounting system to an internationally accepted accrual accounting framework for government finance statistics (eg., CANSIM Table 385-0041). For this reason, FMS data is not comparable to GFS data, therefore CANSIM Table 385-0001 is not directly comparable to CANSIM Table 385-0041.               

Another problem is that provincial and local data are always consolidated; any chance of getting the data only for local governments?
Your client can refer to CANSIM Table 385-0037 for local government data that is not consolidated.

Another question re: local government: does it always include aboriginal governments, or not?
Local government does not always include aboriginal governments.

Wednesday, January 24, 2018

Dissemination Area Around Bedford, NS

Question: 
I got a good question today! A student noticed that “the stats Canada definition for Dissemination Areas is supposed to be between 400 and 700 people, but some of the DAs have populations over 1,000 and upwards of 6,000 (DA: 12090839) for example.”

I told the student that the Census Dictionary definition for DAs says that "Dissemination areas are uniform in terms of population size, which is usually targeted from 400 to 700 persons to avoid data suppression. DAs with lower population counts (including zero population) may result in order to respect the boundaries of census subdivisions and census tracts. DAs with higher population counts may also result.”


I looked at the map of the DA he mentioned, and it did seem to me like there might not be many options to subdivide it further because of how the streets kinda curl into a giant neighbourhood. But I promised I would check. Is it possible for a DA to have a population of over 6000? Is the census dictionary definition the best explanation we have on that topic?

Answer:
Here is the general reasoning I was able to find behind the situation here. It goes a little bit more in-depth than was is provided in the Census Dictionary, but basically covers the same ground:

“The introduction of the national block program in 2001 allowed for the de-coupling of collection and dissemination geographies.  Prior to 2001 the enumeration area (EA) was the smallest unit for the collection and dissemination of census data.  However, the EA was designed to be small enough for census collection and therefore was not always large enough for census dissemination – resulting in some data suppression for small area geography.  The dissemination area (DA) was introduced in 2001 – new DAs (replacing EAs) were delineated using an automated delineation tool in all census tract areas of the country; however, outside of census tracted areas the DAs continued to follow the 1996 EA limits (collection and dissemination geographies were not de-coupled in these areas).  The DAs in the non-tracted parts of the country were completed re-delineated using an automated delineation tool for the 2006 census – at this time the DAs in tracted parts of the country were split as a result of high level boundary changes or where growth dictated. 

Current situation:

DAs are designed to respect both census tract (CT) and census subdivision (CSD) boundaries with an optimal population size of between 400 and 700.  However, in densely populated portions of the country  (e.g. areas with high concentrations of high rise apartments) DAs can be quite large (over 1000 dwellings) but may be composed of just one or two dissemination blocks, thus making it impossible to reduce the population of the DA.  Conversely, in rural parts of the country with very low population density and with the constraint that DAs must respect CSDs (some CSDs have less than 100 people or are made up of two or more discontiguous polygons with very low population counts) DAs can be quite small (50 – 100 dwellings), thus making it impossible to create a larger DA that would contain a population within the ideal 400-to-700 range.”

Monday, January 8, 2018

Annual deaths by hospital unit in Canada

Question:

Do we have any data on hospital deaths by unit in Canada?

Answer:
I think that it would be possible to derive that from the master file of the Discharge Abstract Database from CIHI – I think that that is accessible through RDC? You still wouldn’t necessarily know which “unit” it was (e.g., cardiology / surgery), but you would be able to distinguish emergency admissions/transfers from “routine” one, I think.

It doesn’t go down to the unit level, but CIHI has a product called hospital standardized mortality ratio – tech notes at https://www.cihi.ca/sites/default/files/document/hsmr_tech_notes_en.pdf. HMSR is one of the elements available online, publicly – from the CIHI web site - https://yourhealthsystem.cihi.ca/hsp/indepth?lang=en#/

Once you bring up a hospital, go to the bottom and “see  other indicators” – Hospital Deaths (HSMR)  (e.g., https://yourhealthsystem.cihi.ca/hsp/indepth?lang=en#/indicator/005/4/O5142/) -- but I think you need to look hospital by hospital / city by city / health region by health region…