Our Methodology

Overview

The Charity Insights Canada Project (CICP) conducts ongoing survey research with a panel of Canadian registered charities to generate timely insights on sector trends, challenges, and emerging issues. Our methodology is designed to balance speed, consistency, and broad sector representation, while maintaining transparency about the strengths and limitations of panel-based research.

Data Source

The panel is constructed using publicly available data from the Canada Revenue Agency (CRA), specifically the T3010 Registered Charity Information Returns.

• The project is independent and is not affiliated with the CRA or the Government of Canada.
• This dataset represents the full population of registered charities in Canada.
• CICP uses unaltered copies of this publicly available dataset.

Sample Development

Sampling Frame
The initial population includes approximately 84,400 registered charities in Canada, based on CRA T3010 Registered Charity Information Returns. To focus analysis on organizations with comparable governance, funding structures, and operational models, several core exclusions are applied.

Core Inclusion Criteria

• Designated as Type C (charitable organizations)
• Reported at least 1 full-time employee (FTE)
• Reported more than $1 in annual expenditures (in year of panel selection)

Core Exclusions

• Hospitals
• Schools
• Most religion‑related charities, defined as charities whose primary classification reflects religious identity (including Christianity, Islam, Judaism, Other Religions, Support of Religions, Ecumenical and Inter‑faith Organizations, and Foundations Advancing Religions).

An exception is made for religion‑related charities that operate as ongoing, service‑delivering organizations comparable to other operating charities (for example, faith‑based counselling services, spiritual or educational libraries, facilitator organization supporting and enhancing the work of groups involved in the advancement of religion, etc.).

Panel Composition
From this structured sampling frame, a randomized main sample was drawn, supplemented by seven targeted sub-samples (SS1-SS7)

This resulted in a panel of approximately 1,300 charities, with around 1,100 active participants as of March 2026. The panel reflects diversity across:

• Geography
• Organization size
• Sector of activity
• Operational structure

Supplementary Sub-Samples

To ensure broader representation, additional sub-samples were constructed to capture groups not fully represented in the main frame:

• Subsample 1 (SS1): Includes only charities with the designation A and B (public and private foundations) that had expenditures of at least $1.
• Subsample 2 (SS2): Includes only charities with the designation C (charities) that have no full-time employee (FTE) or that left the field blank and that had expenditures of at least $1. These are what we think of as volunteer run organizations.
• Subsample 3 (SS3): Includes designation C charities with at least 1 FTE, $1 in expenditures, ensuring representation from the province of British Columbia.
• Subsample 4 (SS4): Includes designation C charities with at least 1 FTE, $1 in expenditures, ensuring representation from all geographical regions of the country.
• Subsample 5 (SS5): Includes designation C charities with at least 1 FTE, $1 in expenditures, ensuring representation from the province of Québec.
• Subsample 6 (SS6): Includes charities that have spent at least $1 internationally (with the additional criteria of including designation C charities with at least 1 FTE in CICP’s first two years, but have been excluded since CICP’s Year 3 – 2025)

Additional Religion‑Related Subsample
Beginning in Year 2 (2024), an additional subsample was introduced to capture religion‑related charities that were largely excluded from the core sample.

• Subsample 7 (SS7): Includes Type C charities with at least one FTE and at least $1 in annual expenditures that are classified under all religion‑related categories, including Christianity, Islam, Judaism, Other Religions, Support of Religions, Ecumenical and Inter‑faith Organizations, and Foundations Advancing Religions.

Recruitment & Participation

• Organizations were recruited via email and phone outreach
• Participants completed a digital onboarding form to validate organizational details and provide additional workforce data
• Participation is voluntary, with respondents typically representing senior leadership (e.g., Executive Directors)
• The CICP operates with approval from Carleton Universities Reseasrch Ethics Board (REB)

Survey Process

• Surveys are distributed weekly (Wednesdays mornings)
• Respondents are given:
  • An initial 24-hour response window
  • A reminder on Thursday mornings if not yet completed
• Outputs include:
  • Weekly rapid reports
  • Quarterly briefs
  • Thematic research publications

This high-frequency model enables near real-time tracking of sector trends.

Governance & Oversight

Survey design is led by the CICP’s core team, with guidance from an Advisory Board composed of sector representatives.

The Advisory Board:

• Identifies emerging issues
• Provides input on survey topics
• Ensures relevance to sector priorities

Interpreting the Results: Strengths and Limitations

Strengths

• Large, engaged panel (~1,100 active organizations)
• High response rates (~75-85% weekly)
• Consistency over time, enabling trend analysis
• Direct input from organizational leadership
• Rapid data collection on emerging issues

Limitations and Considerations

1. Digital Engagement Bias

All surveys are conducted electronically. As a result:

• Participating organizations are more likely to be digitally engaged

2. Respondent Knowledge

Survey respondents may not always have direct knowledge of all operational processes.

• Some responses may reflect best estimates rather than confirmed practices
• This is particularly relevant for administrative processes

3. Sample Restrictions

The panel excludes or underrepresents certain types of organizations. As a result:

• Findings are most representative of operational charities with staff and expenditures in the current operating year
• And cannot provide information about or should not be interpreted as fully representative of the entire charitable sector

4. Panel-Based (Not Fully Random)

While the initial sample is randomized:

• Participation is voluntary
• The panel is not a fully random probability sample of all charities

This introduces potential:

• Self-selection bias
• Overrepresentation of organizations more willing to engage in research

5. Sub-Sample Integration

Targeted sub-samples are included to improve coverage, but:

• These groups may be overrepresented relative to their population size
• Results are typically reported without weighting (something which we plan to address going forward)

On Representativeness

CICP results should be interpreted as: Highly informative of trends and perspectives among engaged, operating charities in Canada, rather than precise population estimates.