Stakeholder engagement: risk and equity matrix

Purpose: To review how equitably risks and benefits are distributed amongst stakeholders, researchers and others involved in any research project.

Description: The aim of the matrix is to stimulate systematic consideration of potential impacts for stakeholders, researchers and others involved in a research process, so that risks and benefits can be equitably distributed.

For each party involved in the research, this involves completing a matrix, such as the one shown below, that considers:

  • Costs
  • Benefits
  • Risks
  • Mitigation.


MIT Governance Lab (2020b)

Costs involve “any material or intangible items that must be given up in order to achieve the study goals. This might include payments such as time, personnel, or spent political or social capital. Costs can’t be directly mitigated” (MIT Governance Lab, 2020a).

Benefits cover any expected or potential positive outcomes for the different parties involved in the research, “including material or intangible gains like compensation for time, new knowledge, hard resources (services, hardware, etc.), capacity-building (skills), or reputation-building” (MIT Governance Lab, 2020a).

Risks are “possible harms including a comprehensive list of expected and unexpected events that may happen” (for example, in the field) and “that could negatively impact any actors involved. The risks can also be annotated with analysis of levels of likelihood and severity (low to high)” (MIT Governance Lab, 2020a).

Mitigation “references plans to alleviate any potential risks identified for specific actors. For example, addressing threats to personnel safety and well-being in the field includes both designing research protocols to be carried out in teams of two, using low-cost tech, and having regular check-ins to address any unexpected problems that may arise; as well as having a communication and escalation action plan ready in case” (MIT Governance Lab, 2020a).

These matrices should be filled out for everyone involved in the research, including:

  • all stakeholders, including practitioners, civil society partners, government actors, informal and traditional authorities (eg., “community leaders, religious, traditional, or tribal chiefs,” MIT Governance Lab, 2020a), businesses
  • all researchers, from lead researchers to research assistants and study managers (some risks and their mitigation may be covered in institutional work health and safety reviews)
  • “research subjects,” noting that these are generally covered in institutional ethics reviews
  • Other actors, such as donors and the media. (Adapted from MIT Governance Lab, 2020a.)

Process issues

As far as stakeholder engagement is concerned, filling out the matrix is ideally a joint exercise between the academic leads and each stakeholder group. Conversation prompts include:

  • “Look at the different categories, are any of them lopsided? Do risks seem equitably distributed?
  • If it is lopsided, is there a way to shift the intervention or research design to change this?
  • Are the mitigation plans specific enough to implement? Is there an accountability mechanism or fail safe in case something goes wrong?
  • Have mitigation measures been budgeted for sufficiently?” (MIT Governance Lab, 2020a).
    There should also be follow-up to double-check that stakeholders are still comfortable with risks they have accepted, along with the process and decision-making that led to that acceptance.

The matrix is useful at three different stages of the research:

  1. “in the planning phase of the collaboration, before major decisions on the research or intervention design, and budget, are set in stone”
  2. at “a mid-point check-in to revisit the matrix and see if anything needs to be reconsidered or updated”
  3. at the end of the project, when it is important to “document critical lessons: What went according to plan? What changed or was unexpected? What takeaways can be noted to improve outcomes for next time?” (MIT Governance Lab, 2020a).

The following questions can help assess how well the relevant issues have been reviewed. The parties involved ask themselves if they have been:

  • “Specific: Are your responses tailored to the actor and clear about the extent to which people are being affected (e.g. ‘10 hours per week’ instead of ‘time costs’)?
  • Comprehensive: Do your answers include a range of considerations beyond time, money, and physical safety? For example, political, reputational, professional, and emotional factors.
  • Realistic: Do researchers have the time, ability, and authority to implement mitigation strategies? Are the strategies feasible and measurable (e.g., ‘weekly calls to facilitate communication’)?
  • Balanced: Conduct assessment at various time periods to understand immediate and longterm effects of intervention to inform larger programming goals.”
  • “Vague: Are your answers too brief, non-specific, or not concrete enough to evaluate (e.g., ‘no greater risks than everyday risks’ or risks broadly categorized as high, medium or low without explaining what was at risk and why)?
  • Repetitive: Are you using generic answers, or copying and pasting answers across the different actors?
  • Inconsistent: Are you giving much more attention to one category or actor than another? For example, sometimes people give much more consideration and detail to describe benefits to various actors than costs, risks, or mitigation” (MIT Governance Lab, 2020a).


    • MIT Governance Lab (MIT GOV/LAB, Authors: Alisa Zomer and Selmah Goldberg). 2020a. “Risk and Equity Matrix”. Version 1. Engaged Scholarship Tools. Massachusetts Institute of Technology Governance Lab, Boston, United States of America.
    • MIT Governance Lab (MIT GOV/LAB, Authors: Alisa Zomer, Varja Lipovsek, and Selmah Goldberg). 2020b. “Workbook”. Version 1. Engaged Scholarship Tools. Massachusetts Institute of Technology, Boston, United States of America.
    • A guide and workbook for using the matrix are available at:

Related tools on this website:

Related tools on the i2Insights blog: N/A

Related topics on Wikipedia: N/A

Posted: February 2021
Last modified: February 2021