The RDC Data Access Division (DAD) only considers research projects that are not descriptive in nature. They must be analytic with a focus on modelling.

Where possible, researchers are expected to use publicly available data tables to support analytic modeling outputs.

If you wish to obtain custom tabulations, you will need to contact the appropriate subject-matter division within Statistics Canada, or contact infostats@statcan.gc.ca for more information.

All researchers who intend to include First Nations people, Métis or Inuit in their study population are encouraged to engage with Indigenous organizations and communities.

Resources have been developed by First Nations, Métis and Inuit organizations that explain ethical considerations, the role of research in Indigenous self-determination and the imperative for respectful research practices.

Examples

• First Nations Information Governance Centre
• Inuit Tapiriit Kanatami
• Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans – TCPS 2 (2022)

1. Project title:

2. Project description, rationale and objectives:

• Provide a brief summary of the current literature and identify the knowledge gap(s) to be addressed (approximately 1 page)
  
  
• Clearly identify the specific questions or objectives of the project

3. Data requirements:

• Justification for using microdata (as opposed to alternative data access methods such as public use microdata files, remote access or custom tabulations)
  
  
• Dataset(s) required
  
  
• Time period: List years/cycles required
  
  
• Sample description: Identify the unit of analysis and population of interest
  
  
• Geography: Identify the level(s) of geography for which estimates will be produced (e.g., national, provincial or lower)
  
  
• Variable list: When possible (i.e., if sufficient documentation is available for use outside of the RDC), provide a list of variables expected to be used in the analysis
  
  
• Will datasets be merged or pooled together?
  • If yes, explain how the data from each source will be combined and utilized in this analysis. For example, you may pool data from the same source to increase your sample size, or merge contextual data to the micro-records.
    
    
  • Record linkages are not permitted in the data centres. For more information on the process to request a microdata linkage, see: Social Data Linkage Environment (SDLE)

4. Methodology:

• Descriptive statistics: When possible, use publicly available tables as opposed to producing descriptive statistics in the RDC. Summary statistics produced in the RDC must support the corresponding analytical output. Describe the tables expected to be produced.
  
  
• Is the expected sample size sufficient to complete the analysis as well as respect the confidentiality of the respondents? Explain.
  
  
• Modelling: Describe the type of modelling being used (OLS, Logistic regression, etc.) Note: Applicable weights must be applied when using survey data.

If the project utilizes machine learning techniques, include the following information:

• Describe in detail the type of machine learning techniques being used (e.g., supervised/non-supervised, neural networks, support vector machines (SVM) and tree-based methods including random forests, variable section such as lasso, ridge, and elastic nets, etc.)
  
  
• List the software requirements to carry out the analysis. See below regarding the use of software programs and packages.

Note: Key considerations for confidentiality vetting, such as establishing the minimum unique observations and testing are outlined in the Vetting Handbook. To obtain the most up to date copy of the handbook, contact your local RDC or FRDC Analyst.

Refer to these additional guidelines for proposals requesting the following data:

• Census and National Household Survey (NHS)
• Longitudinal and International Study of Adults (LISA)
• Survey of Household Spending (SHS)

5. Describe requirements for any external data:

• Statistics Canada recognizes that in some cases, combining data from different sources has the potential to strengthen research by increasing sample size and/or providing contextual information. For example, environmental data may be merged to Census geography in order to add an environmental component to an analysis (e.g., organization data collected through the Canadian Urban Environmental Health Research Consortium – CANUE).
  
  
• External data refers to information that is collected by other organizations, departments of government and individuals for their own purposes.
  
  
• Publicly available data (e.g., data or information that is available on the Internet or that can be obtained by anyone from other sources) can be added to a research project once approved.
  
  
• Describe any external datasets that you plan to use, and how you will incorporate them into your analysis.
  
  

6. Software requirements: What software will be required? Not all software programs and packages are available for use in the FRDC/RDC. In some instances, packages cannot be used.

All software packages and add-ons that are not currently approved for use are required to go through an approval process, which may cause delays. See frequently asked questions, or check with your local RDC or FRDC Analyst.

7. Expected output: Describe the products that will result from the proposed analysis (e.g., working paper, peer-reviewed journal article, book or chapter, thesis or dissertation, conference presentation, or commissioned report (e.g., government report).

8. Expected start date and duration of the project:

9. Source(s) of funding for this project: List all sources of funding, or indicate n/a if not applicable.

10. Location of work:

• If the location of work is the FRDC, is the project covered by a departmental seat or through a specific funding arrangement? If yes, specify.

11. Literature references:

• CRDCN
• Research Data Centres (StatCan)
• Upcoming Training (CRDCN)
• Data Available (StatCan)
• FAQ (StatCan)
• Application Process & Guidelines
• Publications