On this page
Getting started
Record linkage proposals involve a significant amount of initial analysis and discussion to develop a formal project contract. Information is provided to Statistics Canada so that project feasibility can be assessed before the formal record linkage application for approval can begin. To facilitate the experience through all project steps, applicants are encouraged to familiarize themselves with the requirements for approval, understand their data sources and what is needed for record linkage, as well as giving thought early on about desired outputs and data access protocols.
Here are some questions to consider before starting a record linkage project at Statistics Canada:
For approval
- Do you have a clear research question?
- Do you have a research protocol? Do you have an ethics approval?
- Does your study meet the expected results from Statistics Canada's Directive on Microdata Linkage? Can you clearly demonstrate how the public interest is served by your project and why a record linkage is the best means to achieve this public benefit? Are you sure that the data you expect from linkage are not otherwise available?
Feasibility of linkage
- Have you examined the available source data file documentation to ensure that the variables of interest will serve your needs and that the sample (if applicable) is adequate for your study?
- If applicable, does your microdata file have personal identifiers that would enable record linkage?
- If your project requires an external microdata file, have you received permission from the data owner to provide it to Statistics Canada?
Final deliverables and access
- Have you thought about the structure of your final linked analysis file? Are derived variables needed? Will you need to person-orient records? How will different reference periods be treated? Will you be using longitudinal data?
- What are your plans to identify and address the data quality issues associated with record linkage and with the specific sources that you have in mind?
- How do you intend to access the linked microdata? Have you considered access through the Research Data Centres (RDC) Program?
Phase 1: Feasibility of the study and calculating time and cost estimates
In Phase 1 of the record linkage application process, the applicant needs to provide information about the proposed study so that Statistics Canada can assess the feasibility and prepare a preliminary cost and time estimate for conducting the record linkage.
SDLE Proposal Form
The SDLE Proposal Form collects contact information, description of the study, details on data files required, and expected timelines and deliverables. The client is required to identify the personal identifiers that are available on the cohort file to be linked.
Feasibility
The feasibility of the study will be assessed against the conditions of Statistics Canada's Directive on Microdata Linkage, and whether the content and coverage of the data files to be used in the linkage would meet the needs of the study's objectives. In some cases, this may require a more in-depth analysis (feasibility study) which would be conducted on a cost-recovery basis.
Initial cost estimate
If the study is deemed to be feasible, Statistics Canada will provide a preliminary cost estimate so that the applicant can secure appropriate funding. Preliminary cost estimates may be subject to change based on changes to the initial timelines and assumptions. Statistics Canada provides custom services, including the Social Data Linkage Environment (SDLE), on a cost recovery basis (fee-for-service). This means that clients are responsible for covering the full costs of the work required to deliver their project. Each SDLE project is unique. Fees are determined by the scope, technical complexity, data requirements, and level of effort associated with the proposed work. As a result, costs vary from project to project and reflect the specific requirements of the client.
Phase 2: Formal approval
Upon funding being secured for the proposed study, phase 2 will commence. It involves preparation of the application to link proposal, review and approval of the record linkage. During this process, applicants may be contacted with comments or questions for the purpose of clarifying the submission.
Approval by the Chief Statistician of Canada
When the application is considered complete after the review by the Social Data Linkage Environment (SDLE) team, subject-matter experts and Information Management Division, it is submitted to the Chief Statistician of Canada for review and approval. If approved, a summary will be published on Approved record linkages. These are posted to demonstrate transparency and to inform the Canadian public how information on individuals is used. The same summaries are included in the Annual Report to Parliament on Privacy.
Contract(s)
A letter of agreement (contract) will be prepared that will outline the terms and conditions of the work to be undertaken by Statistics Canada in order to complete the proposed linkage. It will be sent to the applicant for signature. The anticipated start time and duration of the record linkage project will be negotiated with the applicant and will be part of the letter of agreement.
Phase 3: Record linkage and deliverables
Once all necessary approvals are in place and a contract is signed by all parties, the record linkage will be conducted and a microdata file will be constructed and will reside at Statistics Canada.
Deliverables
The output of the record linkage project will be a microdata file without any personal identifiers. In other words, a file containing records on individuals but will all of the personal details (names, addresses, phone numbers, etc.) removed. This file will be deemed sensitive statistical information (SSI) and will be subject to the confidentiality requirements of the Statistics Act. SSI is any information from surveys or other sources about a person, business or organization held by Statistics Canada – with or without name or other direct identifiers – as well as any aggregation of that information that could directly or indirectly identify a person, business or organization.
The deliverables to be generated from the SDLE need to be considered in phase 1 and specified in the contract. These can take many forms, depending on the applicant's requirements. The outputs options are:
- Linked analysis file - This type of file contains analytical variables from both the cohort and outcome file(s) (flat file or relational file format) with supporting documentation.
- Linked outcome file(s) - This type of file contains cohort unique ID with specified analytical file variables. Additional merge with the cohort file is required to generate a linked analysis file. Supporting documentation is provided.
- Linked keys file(s) - This file contains only the unique associated identifiers of the source micro-records resulting from the linkage of two or more source data setsof the cohort and outcome files(s) that were linked. Additional merge of cohort file variables with outcome file(s) variables is required to generate a linked analysis file.
- Feasibility study - For example, when research requires extensive explorative analysis to determine feasibility of linkage and costing.
The deliverables will be supported with documentation.
Access
The applicant can apply to access the microdata file in a Research Data Centre (RDC). RDCs provide access, in a secure university setting, to microdata from population and household surveys, as well as from linked microdata files. The centres are staffed by Statistics Canada employees. They are operated under the provisions of the Statistics Act in accordance with all the confidentiality rules and are accessible only to applicants with approved projects who have been sworn in under the Statistics Act as 'deemed employees' of Statistics Canada. As is the case with employees of Statistics Canada, a deemed employee does not get access to all information held by Statistics Canada. Access to confidential information is strictly on a need-to-know-basis. The applicant will need to sign a microdata research contract which will specifically indicate the microdata files that can be consulted for the project. Having access to the files in the RDC allows the applicant to perform inferential statistical analysis on the confidential microdata. A Statistic Canada analyst will review all outputs to ensure they meet the confidentiality requirements of the Statistics Act before they are taken out of the RDC.
It is recommended that applicants contact Microdata Access Division well before delivery of the linked data outputs to start the access process.
Consult the RDC program website to get more information on the application process and guidelines.
Phase 4: Retention of linked files, keys and identifiers
As stated in Statistics Canada's Directive on the Management of Statistical Microdata Files, the linked files are retained until no longer required. The linkage keys and direct identifiers are to be retained until the possibility of re-use for linkage purposes is considered low.