By Gabriele Bammer
Given that most research projects will not have the funding or time to involve all stakeholders who have been identified as potential contributors, what criteria are useful for selecting those to be invited to participate? How can those identified be assessed against the criteria?
Four criteria for selecting stakeholders are:
- their legitimacy
- their real and potential power
- the urgency they assign to the problem
- practical considerations.
Legitimacy
Legitimacy can usefully be examined along the following four dimensions:
- Potential value of the stakeholder contributions.
Stakeholders who bring to bear extensive experience; insights into central aspects of the problem, its context or ways of dealing with it; and/or analytical details have greater legitimacy than those with shallow or peripheral viewpoints.
- Whose interests the stakeholders serve.
Stakeholders who work for societal good or the good of a group have greater legitimacy than those who primarily serve their own interests.
- How representative the stakeholders are.
Formal representatives of stakeholder groups have more legitimacy than informal representatives or self-appointed representatives.
- Legitimacy conferred on the stakeholders through participation.
The invitation to participate in research can confer legitimacy on stakeholders and their perspectives. This can be used deliberately by researchers to increase the legitimacy of marginalised groups.
For any particular stakeholder, legitimacy may not be uniform across each of these dimensions and there is no easy way of weighting the importance of different aspects. For example, a stakeholder may have extensive experience to bring to bear, but may be self-appointed. This was the case in my research on the feasibility of prescribing heroin to treat heroin dependence, where much was learnt from drug users, service providers and police who volunteered to help me understand their worlds.
Real and potential power
Stakeholders may already have power or may have the potential to be powerful, and it is useful to consider both groups along three dimensions:
- Utilitarian power
Stakeholders with utilitarian power contribute (or could contribute) resources or practical support to the research or to implementing its outcomes. This may include access to funding, data sources and equipment.
- Normative power
Stakeholders with normative power contribute (or could contribute) symbolic resources such as prestige, esteem, and acceptance. This can include public support for, and endorsement of, the research from community leaders or professional associations.
- Coercive power
Stakeholders with coercive power use or could use force or influence to direct the research. One way this can play out is when stakeholders with vested interests set out to halt or discredit research that threatens those interests.
Urgency
Urgency or the need for immediate attention applies to stakeholders for whom understanding and/or acting on the problem is of major importance. Importance can be compounded by time sensitivity.
Time sensitivity can be associated with problems that have an irreversible element, for example they are life-threatening, associated with significantly reduced quality of life, threaten economic security, threaten ecosystem biodiversity, or threaten the destruction of artefacts.
Time sensitivity can also be associated with a narrow window for action. In a policy context this is often referred to as a “window of opportunity,” when political and other circumstances mean that there is a good possibility for action to be taken and to be successful.
Practicalities
Selecting stakeholders based on legitimacy, power and urgency ensures that the most relevant stakeholders are involved, which is generally not the case if they are selected on practicalities alone. Nevertheless, practicalities cannot be completely ignored. In working within the constraints of limited budgets and time, practical considerations include:
- How easy is it to access the stakeholders?
- Have relationships with the stakeholders already been built?
- How willing are the stakeholders to participate?
- What can realistically be achieved?
- Are particular skills or other resources to enable participation required and available, eg., expert facilitation, recompense to stakeholders for their participation?
- Are there any risks?
The selection process
Stakeholders can be mapped according to how they rate on each of the first three attributes, using a Venn diagram as shown in the figure below.
Stakeholders can then be categorised as having:
- High legitimacy, power and urgency
- High legitimacy and power
- High legitimacy and urgency
- High legitimacy
- High power and urgency
- High power
- High urgency.
In general, priority for selection would be given to stakeholders with high legitimacy (groups 1-4) over the other groups.
In the heroin trial feasibility research, for example, police commissioners, the police union and the peak medical association were considered to have high legitimacy and power. Dependent heroin users and their parents were considered to have high legitimacy and urgency.
At the other extreme, the research team was occasionally approached by members of the public who were very concerned about the issue who had advice, but little concrete experience, to offer; they could be assessed as meeting the criterion of urgency, but as having neither legitimacy nor power.
The four attributes provide an aid to decision making rather than hard and fast rules. Importantly there are a number of constraints in assessing stakeholders according to the attributes, including that they are:
- dynamic and can change over time
- socially constructed
- difficult to assess.
Assessment of stakeholders against the criteria of legitimacy, power and urgency may require input from a diverse group of advisers, who have different insights into the various stakeholders, including from direct knowledge, observations and their own research.
Assessments become particularly important when selecting stakeholders with opposing views on a problem, bearing in mind that including stakeholders in research confers legitimacy on them and their perspectives. In such cases a balance has to be found between what the stakeholders can offer the research and the legitimacy that being invited to contribute confers.
Anything to add?
What has your experience been with stakeholder selection? Are there other criteria that you have brought to bear? What processes have you used to decide which stakeholders to invite to join the research? How have you managed stakeholders with opposing views?
Sources and references:
The power, legitimacy and urgency framework was adapted from research in a business context by Mitchell and colleagues. The three dimensions of power were adapted from the research of Etzioni cited by Mitchell and colleagues. The challenges in the selection process were adapted from the work by Durham and colleagues on stakeholder engagement. All of these authors, in turn, drew on the work of others, which is cited in the references below.
Durham E., Baker H., Smith M., Moore E. and Morgan V. (2014). BiodivERsA Stakeholder Engagement Handbook. ERA-NET BiodivERsA: Paris, France. (Online – open access): http://www.biodiversa.org/702
Etzioni, A. (1964). Modern organizations. Prentichttps://i2insights.org/2021/12/02/generating-ideas-and-reaching-agreement/e-Hall: Englewood Cliffs, New Jersey, cited in Mitchell at al., (1997).
Mitchell, R. K., Agle, B. R. and Wood. D. J. (1997). Toward a Theory of Stakeholder Identification and Salience: Defining the Principle of Who and What Really Counts. Academy of Management Review, 22, 4: 853-886. (Online) (DOI): http://dx.doi.org/10.2307/259247
Biography: Gabriele Bammer PhD is Professor of Integration and Implementation Sciences (i2S) at the National Centre for Epidemiology and Population Health at The Australian National University in Canberra. She is also a member of blog partner PopulationHealthXchange.
The Stakeholder Engagement Primer comprises the following blog posts:
Published:
1 a. Why a primer? b. Defining stakeholders (October 14, 2021)
2. Identifying stakeholders (October 21, 2021)
This blog post:
3. Selecting stakeholders (October 28, 2021)
Still to come:
4. Options for engagement (November 4, 2021)
5. Choosing engagement options (November 11, 2021)
6. Making engagement effective (November 18, 2021)
7. Listening and dialogue (November 25, 2021)
8. Generating ideas and reaching agreement (December 2, 2021)
9. Evaluating engagement (December 9, 2021)
10. Advanced skills (December 16, 2021)
Thanks, Gabriele.
It was very interesting.
My question is about those who are not able to talk or participate in research or discussions but are affected. Which representative is more legitimate for these stakeholders?
How can we be sure that the representative is properly representing the interests of those stakeholders?
That’s an important question. Can you give some examples of the kinds of stakeholders you have in mind?
An example I can think of is those engaging in criminal behaviour. In my experience you have to work with self-appointed volunteers. The value of such volunteers is that they can provide different perspectives and help you question your assumptions. But they are probably only presenting part of the picture for that stakeholder group – so you also have to be appropriately cautious. It can also help to get as many different views as you can, including reading books and other accounts that may have been published. It is important never to think that you know everything and to be open to learning. Would be great to get other views.
You mentioned a very good point
I also try to make my questions clearer. Imagine a situation where people and the environment are in front of a company or organization. Although it is very useful to know the opinions of both groups, the company’s stakeholders are much more persistent in expressing their positions. On the other hand, representatives related to people, future generations, the environment and animals do not have the patience and strong will to stand against them.
I want to know how to achieve the desired result in such a situation?
In such cases, I try to quantify the answers obtained by different groups by correctly recognizing the factors and metrics.
Many thanks for the clarification. You have highlighted a challenging situation where there is both a major power imbalance and stakeholder groups that cannot be easily represented.
As you mention (and describe in your blog post at https://i2insights.org/2021/07/20/system-redesign-for-shared-value/ and the paper it is based on), one way to help overcome the power imbalance is to use structured decision making rather than relying on dialogue alone.
It would be great to hear from others who have been involved in such cases, both for ways to tackle major power imbalances among stakeholders and how to effectively represent the “common good” (which is how I interpret “people” in your comment), future generations, the environment and/or animals.
Another excellent read. Thank you Gabriele. As an natural scientists who often works with / strays into the world of social sciences in my work I am often asking myself and the teams I work with what is it that we need to do to ensure effective and efficient representation of a stakeholder group and how should this be defined and documented in the research process to ensure non bias or false / misrepresentation? How can you REALLY ensure you have identified and captured all of the groups effectively and equitably.
Good point and question. All researchers can do is give it their best effort and be open to continual learning. There are no sure-fire, fail-safe processes for stakeholder engagement, so all researchers can do is make an honest attempt. It helps if the researchers are genuinely curious and keen to understand, constantly question their own assumptions and challenge their own biases. This can be hard when they are working under time and other resource pressure. Stakeholders will generally recognise and appreciate when a genuine attempt is being made.
Hi Gabriele, thank you for your insightful posts. Another consideration in the identification and selection of stakeholders and partners that we have found, relates to their role at the different stages of the research process. For example, while some may be involved as project partners and collaborators throughout the whole process, some may have a greater role to play at the earlier phases e.g. problem identification, whereas others will have a greater role in the sharing of results throughout their networks. We have used this tool: https://www.beyondresults.co.nz/tools/impact-planning-tool to help identify and make explicit these roles from the beginning. When used in a participatory way with stakeholders and the research team, this increases the understanding of how different stakeholders and partners can contribute to impact and also helps identifies any gaps in stakeholder involvement at an early stage.
Thanks for that important point and for alerting readers to the impact planning tool. I touch on the issues of different stakeholder involvement at different stages of the research, as well as different kinds of involvement at different stages, in Primer #4. I’d suggest that stakeholders involved early in the research are still kept informed as the research progresses, especially of the outcomes – it can be easy to miss that in the busy day-to-day of conducting the research. Is that in line with your experience?
Yes, certainly agree Gabriele to keep everyone informed along the way, although sometimes easier said than done without deliberate processes in place.
Thanks, Gabriele, for this clearly structured approach to stakeholder selection. One additional criterion that I deem relevant, but which is not explicitly mentioned, falls in the category of “Practicalities”: the number of stakeholders to involve. Obviously, there is no “optimal” number for all subjects and processes, but I would imagine there is an upper bound of how many stakeholders you can involve from a) a mere cost perspective, and (maybe more importantly) b) a process perspective (with too many stakeholders making the process hard to manage). Given your ample experience, I would be interested whether you have ever faced a situation where there would have been “too many” stakeholders to include based on the three criteria of legitimacy, power, and urgency, and how you dealt with such a situation?
Thanks for this important question and I hope others will contribute experiences. Processes certainly have to be tailored to the number of stakeholders involved, along with other issues, such as whether there is conflict between some stakeholders. Important considerations for researchers include whether to bring different stakeholders together or to engage them one-on-one, when to bring them together and how to bring them together. In Primer #4 (to be published this week), I review different options for engaging stakeholders and in primer #5 I review the considerations for choosing options. In Primer #10, I touch on working effectively with multiple stakeholders at the same time, as well as effective one-on-one engagement.
Thank you – I have just read this week’s primer and I agree that it helps to look at the question from a different angle. When writing the question above, I was still in the mindset of “involving all stakeholders at all stages”, which in reality will happen only in rare cases, I assume. Being able to spread stakeholders across project components and across time sounds like a reasonable strategy to balance the dilemma of ensuring both diverse participation and practical feasibility. Thanks again, also for the primer series in general!
An interesting organisational framework for managing stakeholder ‘selection’. Two issues that I believe need to be added to assist the robustness of any study based on these criteria are managing researcher worldview bias and diversity of stakeholder knowledge. It is not difficult to imagine that knowledge about the ‘way things are’ rather than the ‘way they could be’ would be favoured by the process described. For me diversity of engagement with the study issue is paramount. Strongly held opposing views can make the study more difficult to manage but it is more likely to reveal new insights.
Thanks Bruce for raising those important considerations. They are one of the reasons why I suggest putting a lot of effort into identifying stakeholders in Primer #2 (https://i2insights.org/2021/10/21/identifying-stakeholders/). Your comment here adds weight to adding a tool such as Werner Ulrich’s critical systems heuristics (https://i2s.anu.edu.au/resources/critical-systems-heuristics) – as you suggested in your comment on that blog post – to ensure that the issues of researcher bias and stakeholder diversity are adequately addressed.