Imdb La Confidential
Serafin Sonnier
The Longitudinal Immigration Database (IMDB) includes administrative immigration data for all immigrants since 1952 and non-permanent residents since 1980. Outcomes for this population are available from tax files since 1982. The IMDB provides detailed and reliable information on socioeconomic outcomes of immigrants after their admission, such as employment income and mobility. It connects short- and long-term outcomes with characteristics at admission, such as immigrant admission category, source country and knowledge of official languages. The database also provides information on pre-admission experience in Canada, such as work or study permits as well as refugee claims. It now includes data about wages and salaries, immigrant children's family outcomes, settlement and resettlement services (information available since 2013), as well as citizenship acquisition (information available since 2004).
The database is managed by Statistics Canada on behalf of a federal-provincial consortium led by Immigration, Refugees and Citizenship Canada.
Collection period: Annual. Immigration data: summarized annually. Tax data (T1 files): filed in the spring following the year of reference.T1 files for income year "y" are received from Canada Revenue Agency during year "y+2".
The Longitudinal Immigration Database includes, all immigrants admitted to Canada since 1952 and all non-permanent residents since 1980. Outcomes for this population are available from tax files since 1982.
The Longitudinal Immigration Database (IMDB) brings together immigration information from Immigration, Refugees and Citizenship Canada, taxation data from the Canada Revenue Agency, and the date of death from the Canadian Mortality Database.
Each year the IMDB is updated with information about the newest immigrant and non-permanent cohorts. Also, data about citizenship, settlement services and taxation are updated annually. Tax information from previous years may be added later on if individuals' data are subsequently linked.
The immigration microdata go through a series of edits prior to linkage to ensure that there are no duplicate records and that no data mandatory for matching is missing. Logical edits, range edits and content edits are used at this stage to improve the data quality of the files. After the linkage, extreme values and missing information necessary for derivations are identified. SAS software is used for these processes.
A series of diagnostic tables is produced after each update to evaluate the quality of the match. These provide historical trend comparisons of counts with previous years data by key characteristics. Coverage rates are produced to assess the success of the matching process.
Statistics Canada is prohibited by law from releasing any data which would divulge information obtained under the Statistics Act that relates to any identifiable person, business or organization without the prior knowledge or the consent in writing of that person, business or organization. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. As necessary, data are suppressed or rounded to prevent direct or residual disclosure of identifiable data.
The use of the Longitudinal Immigration Database is subject to the normal privacy and confidentiality constraints to prevent the release of personal information. Micro-records are not released to users and all aggregate statistics are subject to suppression or rounding.
The Longitudinal Immigration Database (IMDB) combines administrative immigration data with T1 Family Files (T1FF) and T4 supplementary file through probabilistic record linkages. Of immigrants admitted between 1980 and 2021, 85.7% were linked to at least one T1FF record from tax year 1982 to 2021. Details on data accuracy for information from the T1FF may be consulted under the T1FF entry (record number 4105). Of immigrants admitted between 1980 and 2021, 70.8% were linked to at least one T4 record from tax year 1997 to 2021.
The immigration files go through a validation process for accuracy of code sets and comparisons with the previous year's files.
No coefficients of variation are produced for estimates as the IMDB is considered a census of all immigrants within the reference years.
As per usual it is a challenge to try and capture what is happening fromsuch a broad and engaged community. I can only imagine the challenge forthe other regional secretaries. I thank all those who submitted contentand I apologize if I missed something. These are in no specific order.
Just one year ago, the project, building a data portal designed tofacilitate data exploration and extraction, received funding from theOntario Council of University Libraries (OCUL)and broader Public Sector (BPS) Supply Chain Secretariat, OntarioMinistry of Finance(OntarioBuys). Sincethen, much has been accomplished: Paula Hurtibise, the project managerwas hired, co-op students have marked-up over 400 datasets includingdemographic data from Statistics Canada and polling data from Gallup andon May 4, 2008 odesi.ca was launched on the highlysuccessful Scholars Portal website.
Key to the success of is the DDI, Best Practices Document (BPD) createdby the project team. The BPD is widely used by Canadian institutions formarking-up files in DDI format and is available throughodesi.ca in both French and English.
The western Canadian library data services group (ACCOLEDS) isundertaking a pilot project with the editors of a couple of open accessjournals to mark-up statistical tables. The objective is to mark-upcontent about tables that will allow better searching for statisticalinformation in these tables and that will enable linking between tablesand the data sources from which they were derived. Maxine Tedesco (anIASSIST member from the University of Lethbridge) will be spearheadingthis pilot while on study leave in 2008.
Second, Library and Archives Canada concluded at the end of 2007 apublic consultation to develop a Canadian Digital Information Strategy.This process resulted in a document detailing the principles andactivities needed in Canada to produce, preserve and provide access todigital content arising in the heritage, scientific and governmentsectors( -1000.01-e.html).Research data are clearly identified in this strategy document asvaluable digital resources that Canadians need to preserve and toprovide access.
CARL formed a new Data management Working Group after a 2007 survey theyconducted revealed that most of their member libraries are interested inholding researcher-generated data, but few have a formal datapreservation policy. This working group consists of five directors frommember libraries.( -abrc.ca/about/working_groups/data_mgt_mandate-e.html)
Discussions in CANDDI tapered off with the emergence of the DLI DDI andprojects. While the CANDDI template helped facilitate discussion aroundthe best practices document that was subsequently developed inconjunction with , all other activity pretty much disappeared in CANDDI.Recent interest, however, has surfaced to revitalize CANDDI to addresssome of the other issues for which it was originally organized,including shared variable group names and control vocabulary.
In Canada, there have been some positive developments in addition tosome continuing concerns. The group working on the 2006 Census PUMFs hascome up with a compromise that would see two files developed. The firstwould be an individual file (2% sample); the second would behierarchical in nature (1% sample). This represents a move from theoriginal position which would have eliminated the individual file. Thecommittee is to be commended for it extensive consultation andcreativity in the face of competing demands.
The Canadian Household Panel Survey has been working on producing alongitudinal, synthetic file that would allow users to create modelsthat mimic the actual confidential data without risk of disclosure.While not PUMFs, these will be public and available for preliminaryresearch. Both the above innovations have been driven by internationaldemand.
On the other hand, there is a committee studying the future of PUMFs.The fact that the value of PUMFs greatly outweighs their cost may not befully appreciated. With 74 post-secondary institutions and projects suchas serving to make PUMFs ever-more accessible, the cost of using thesefiles is approaching zero. Curtailing the production of PUMFs wouldresult in the same sort of situation we were in pre-DLI.
In addition, DLI has budgeted for two R&D projects focusing on makingmetadata more compatible. The first is to develop a two-way crosswalkbetween the DDI and the IMDB. The second is to investigate thefeasibility of creating a wizard to take Beyond 20/20 files and markthem up as DDI-compliant cubes. That would mean that we could augmentthe metadata and move the B20/20 files out of their proprietary formatand into a preservable mode.
We are now creating data about data. :-) The second survey on theStatistics Canada DLI program has been conducted in March-April 2008.The participation of the Canadian data librarians has been fantastic.Final response rate: Overall 82% (97/118) DLI contacts 92% (66/72)Designates 67% (31/46). Gaetan Drolet and folks are now cleaning thedata to produce a SPSS file. The data will be analysed this summer byWendy Watkins and Gaetan Drolet. They plan to split the data by DLIregions for the DLI training coordinators. They will present a reportabout the training information collected to the DLI training committeein September 2008 and a report to the EAC for future planning of the DLIprogram. They also plan to disseminate the data to the DLI communitylater in a DDI format with NESSTAR and possibly make presentations toDLI workshops and eventually IASSIST.
During the year the Laurier Institute for the Study of Public Opinionand Policy (LISPOP) received a generousdonation from IPSOS- Reid. This includes many public opinion surveys,election surveys and other types of surveys. LISPOP is working closelywith ODESI to mark-up this rich collection, with a target of making thispublicly available.
c80f0f1006