Finding Statistics Canada open data

Jeremy Buhler
Data Librarian, UBC Library Research Commons
jeremy.buhler@ubc.ca
https://researchcommons.library.ubc.ca/

Learning objectives

Understand what data is available
Know where to find it
Access public Use Microdata Files (PUMFs)
Learn where to get help

slides at https://bit.ly/find-statcan-data

Use the Zoom toolbar to engage

Participants window

To annotate in a Zoom presentation

Often use Statistics Canada resources

New to Statistics Canada resources

New to working with data

Often work with data

About Statistics Canada

Serving Canada with high-quality statistical information that matters

"Statistics Canada produces statistics that help Canadians better understand their country - its population, resources, economy, society and culture."

from https://statcan.gc.ca/about/about

Statistics Canada Open Licence

https://statcan.gc.ca/reference/licence

Grants permission to

use
reproduce
publish
freely distribute
sell

Statistical overview

counts, percentages, etc. for certain regions or groups
calculated for a purpose, but not very flexible

Underlying data can answer follow-up questions

How does that differ by province? By age group?

Public Use Microdata Files (PUMFs)

Public data modified to protect respondent privacy

Some variables suppressed
Some variables aggregated
May only be a sample of responses

That's where the Public Use Microdata Files, or PUMFs, come in. Please use the reactions button to indicate yes or no, whether you've looked at PUMF data before. Even if you've just opened a PUMF to see what's inside, answer "yes"; if you you've never accessed a PUMF, answer "no". This workshop doesn't assume any prior experience with PUMFs, but it will help me to understand what experience you have.

PUMFs are files that contain case-level data from the Census and from Statistics Canada surveys, meaning each row contains data for one individual. There's a lot you can do with this level of detail, but researchers should be aware that this public data is modified in a few ways to protect respondent privacy. Sensitive variables or those that could identify individuals may be removed from the PUMF or aggregated. For example, PUMFs don't usually contain sexual orientation data, even where that data is collected, and age might be reported in age groups instead of an exact age. For the Census, the PUMF represents a 2.7% sample of the Canadian population - not responses for every individual. This is still plenty for of data for many kinds of research, but those who need data not included in the PUMFs may need to apply for access to a Research Data Centre, as the more detailed data is not freely available.

In a moment we'll look at some tables on the Statistics Canada website, then explore PUMFs available from Abacus, UBC Library's data repository. Before we do, I'd like to hear from you. What kind of data are you working with, or do you expect to work with? Are there specific Statistics Canada surveys that interest you? Please respond in the chat.

Where to access tables and PUMFs

Statistics Canada Data search https://www150.statcan.gc.ca/n1/en/type/data

Abacus Open Data collection https://abacus.library.ubc.ca/dataverse/open

Steps for Statistics Canada Data search. make sure window is sized so subjects appear on left, not top

Orient, show subject list
Filter by subject "Labour", note large number of results
Search keywords "employment"
Switch to "Tables" tab, select "Labour force characteristics by province..." in new tab
Point out the 7.1% unemployment rate from article
Scroll back up, show filters and add/remove data; but what if you need even more detail?
Scroll to bottom, show source of data (Labour Force Survey)
Return to search results
Switch to PUMF tab and select LFS (second in list)
Note data availability, indicated by links (older data not available)
Two reasons to go to Abacus: PUMF exists but is not accessible from site, or PUMF is there but want more documentation and related files, or data in a different format (LFS is only in CSV on Statistics Canada site, but SPSS file in Abacus)

After Statistics Canada, ask about presentation speed

Steps for Abacus Open Data collection

Orient, show filter options
Note that most recently updated datasets appear first
Open Labour Force Survey 2021, orient
Note large number of files, note the tags. Try not to scroll past the tag chooser so you don't need to scroll back up.
Explain importance of documentation, open first file, show TOC
Return to record, apply August tag to filter
Show August files, show tab file (what to look for to find the data, the tags and the file descriptions
Download the Aug LFS .sav file, note the license (which you mentioned earlier)
Open in SPSS, go to variable view
Switch to data view, make it clear that this freely available data includes the individual responses from the nearly 87K people who completed the August 2021 Labour Force Survey
Pause for questions
If no questions, return to Abacus open data and search for CCHS, browse files and point out documentation, etc.

Selected PUMFs in Abacus

Census of Canada
Labour Force Survey
General Social Survey
Canadian Community Health Survey
COVID-19 Crowdsourcing series
...and many more

Additional support

Data analysis tools and methods

Online workshops https://researchcommons.library.ubc.ca/workshops

Consultations https://researchcommons.library.ubc.ca/consultation-requests

General questions

Jeremy Buhler, Data Librarian jeremy.buhler@ubc.ca