«Abstract. The paper explores what exactly it is that users participate in when being involved in participatory design (PD), relating this discussion to ...»
Even when choices are open to participation and all voices are heard, not all positions can be equally reflected in the design. […] within one and the same project there may be different depths of participation, depending on the role and particular expertise of participants but also on the types of issues. […] (p. 106f).
Our analysis suggests thinking about how to assess the depth of participation.
This is not a new idea. Probably best known is Arnstein’s (1969) ‘ladder of participation’, which is based on a distinction between empowering forms of participation, ‘tokenism’ (in which she includes consultation, informing and placation) and nonparticipation. Pretty’s (1995) typology reaches from ‘manipulative participation’, where ‘participation is simply a pretense’, to selfmobilization. He also introduces a distinction between participation as a means (functional participation) and participation as a right (interactive participation), where the process is a joint effort. Other examples are Rocha’s (1967) ‘ladder of empowerment’ or Cornwall’s (2008) ‘typology of interests’, which focuses on who participates and where their agency and their interests take them. Hyysalo (2015) maps various use-oriented approaches according to the ‘agency given to designers and users’, including recent developments where users drive the design process.
Our analysis of decision-making in PD projects allows us to be much more specific. We specify a set of criteria for assessing how participatory a PD project is and even suggest a hierarchy of these criteria.
At the top of our list of criteria is the participatory design result, which we have defined as increasing users’ ‘power to’, since this is what PD is all about.
The IT artifacts that came out of the four projects empowered their intended users in many different ways: supporting their ways of working or giving them a voice in and influence on processes they would otherwise find inaccessible. The Florence project’s particular achievement was an IT artifact that followed the nurses’ ‘logic’ of working; Sisom implemented a system that reflected a child’s ‘logic’ of talking about symptoms as well as a playful and stimulating way of recording them; IPCity created a ‘new language’ for building urban scenes that enabled citizens to express and develop their ideas. Although the design result of Desarte was not as useful as envisioned, it enriched architects’ thinking about inspirational material. Not one of these outcomes would have been possible without the contributions of users.
But how do we come to know that a project has achieved a participatory design result? And how much does it depend on the participatory process? PD researchers would argue that the quality of a design result has to be assessed in ‘real use’ after design. This was clearly the case in Florence and Sisom where the design became embedded in the routines of hospital work. A question that is more difficult to answer is how users could see their contributions in the design result.
Our analysis of the projects came to the conclusion that this was the case, without explicitly having asked the users. We think that an explicit focus on how users may recognize their contributions to the design result may help make these more visible; or even assist in documenting the important design moves in ways that makes them open to scrutiny. The third open question concerns the quality of the process: do users have to participate in all stages of a design?
Our answer is more nuanced here, leading to the next step of assessing the depth of participation: user participation in creating choices. This is one of the strengths of PD. Many of its techniques support users in contributing to defining the problems that a design project should address and also indicate possible solutions. There is a difference between setting the stage for design and merely deliver ideas. Hence, apart from seeking to systematically widen the space of possible choices, participatory designers also need to consider the type of choices they invite users to contribute to. Are these just about the design of the user interface (as Sisom seems to be) or do they concern the problems to be considered, as in the wish list the nurses in Florence had prepared?
Next in importance is the ‘see’/evaluate part of designing, which is an essential part of making design moves. Probing an emerging design in use or in a situation that comes as close as possible to real use gives the participating users some real influence. It offers the designers the possibility to see and experience if what they have built meets the defined aims; and eventually correct, modify, and add to the design. We suggest that more systematically planning the ‘seeing’ part of a project may contribute to increasing users’ influence on the design result. The evaluation method should allow the users to probe their own design moves. This requires a certain level of openness of the artifact they are supposed to test. The design result of IPCity, for example, was relatively open: ‘We did not implement any ‘rules’ or ‘constraints’ beyond the technical limitations of the tools, and with this made an explicit step away from simulation tools. This moved decisions away from the technology into the responsibility of the participants’ (Wagner et al.
2009, p. 193).
We have seen that user participation in selecting among choices is not always possible (or even desired). In the Florence project selecting among choices was organized in a participatory way. The designers carefully listened to the nurses’ voice and translated the nurses’ sketches into a system. Desarte was small and everybody involved was a ‘maker’. Much of the decision-making was embedded in the joint design moves, many of which included the architect-users. Sisom and IPCity were different in this respect, with power issues playing a much more significant role. Decision-making in both projects was framed from the start: by a strong vision, by institutional constraints, the composition of the project team, which from the beginning emphasized certain properties of the design solution.
The children in Sisom were simply not held capable of deciding by the adults ‘who knew best’; in IPCity some of the selecting was imposed and much persuasion was needed for the urban architect team to consent.
In the context of an IT project concretizing the choices is mainly in the hands of the designers. The possibilities for users to participate in this process depend on the technical ambition and complexity of the project. The PD literature mostly describes users as contributing in the form of mock-ups and scenarios of use, which help the designers learn about users and future use, serve as sources of inspiration or are eventually also ‘translated’ into a working prototype. It seems important to emphasize and facilitate non-technical ways of ‘making’ in order to strengthen users’ influence on the technical implementation. We have given several examples of users concretizing design choices in the form of sketches.
We contend that carefully examining the different facets of participation in a PD project may help more consciously balance the tension between an ideal of participation and the different ‘constraining’ forces. The hierarchy of criteria we introduced mirrors the ‘reflection-on-action’ we undertook when analyzing the four PD projects. It is a ranking of opportunities and not intended to be prescriptive in any way.
7 Revisiting the decision-making framework Our main aim in this paper has been to explore the question what it is that users participate in. We have proposed a method for studying the dynamics of a PD project that, ideally, leads to a participatory result. We have argued that we can observe different types of decision-making: phases in which the design proceeds through a series of ‘design moves’ (Schön), alternate with or are complemented by phases of ‘phantasizing and projecting’ (Schütz). We have also looked into issues of power – the ‘power to’ of the different stakeholders in a project (in some cases only analytically separable from ‘power over’), as well as structural aspects of power that are to do with the context, in which a PD project is embedded. We propose this ‘decision-making framework’ as a tool for ‘reflection-on-action’ in
the sense Schön (1987) described:
We may reflect on action, thinking back on what we have done in order to discover how our knowing-in-action may have contributed to an unexpected outcome. We may do so after the fact, in tranquility, or we may pause in the midst of action to make what Hannah Arendt (1971) calls a "stop-and-think" (p. 26).
It is a conceptual tool to apply during a design project but also when looking back. It can also be appropriated by a researcher engaged in an observational study of design practice. We do not, however, propose to do observational studies of decision-making ‘as such’ (or argue in favor of the development of tools in support of the decision-making process). The notion of ‘design moves’ indicates that coming up with choices, reflecting and pursuing some of them is inherent in the work of designing and, in many cases the act of decision-making is not directly observable. This is why we stress the importance of identifying choices as they evolve: are there alternative paths of action and what are these; do they get taken up or dropped; in acts of deliberation or as part of ongoing design moves?
In practical terms we also suggest to document decision-making in a PD project or other types of design project we are engaged in. In the Desarte project this was done in the form of a diary that told the story of the 3D Wunderkammer with all its twists and turns. In IPCity the designer team very early on decided to document all the decisions concerning token and interaction design that resulted in different versions of the ColorTable (Maquil et al. 2008). The Florence project produced a number of status reports documenting the progress of the project, including the slow progress in the project before the designers identified a hardware bug in the system that made the programs fail. In the Sisom project large parts of the design process were videotaped.
Our second aim was to frame the question of participation and decision-making in a CSCW context. We think that the decision-making framework we suggest applies to design work in general and may help better understand the particular design results that are achieved in a project. It would be an interesting example to revisit, for example, the electronic patient record project Martin et al. (2007) describe, applying the lens of decision-making. This may be difficult without an intimate knowledge of the different design moves and deliberate choices, in particular in a large-scale project that involves multiple stakeholders in different places (see e.g. Grisot and Vassilakopoulou 2015). In architectural planning, another example of design work, we can see how design decisions may ‘travel’ through the whole building, affecting many interlinked parameters.
An intriguing part of this kind of analysis is the notion of decision linkages. In studying design practice several authors have forwarded the notion of linkages (e.g. Cross 1979). Wang and Habraken (1982) examined designers’ decisionmaking process, identifying a ‘critical path’ that led to the design result.
Goldschmidt and Tatsa take up of the notion of ‘linkography’, with the intent to
evaluate how good design ideas are:
The first kind are those moves (or ideas) designated because of their backlinks, i.e. links to previous moves/ideas. The second kind of critical moves earn their designation due to their forelinks, that is, links that posterior moves ‘make’ to them (these links cannot be determined by judgment; they are derived only once the analysis is complete). (p. 595).
These authors are ultimately interested in modeling the design process or in making judgments about creativity, which we are not. However, we think that analyzing the interdependencies that are created by particular choices is crucial to understanding how a not yet circumscribed design space is narrowed down. We have given examples of how design decisions influence each other: evoking new problems that need to be resolved due to unforeseen consequences; opening up alternative choices or preempting them; several small ideas snowballing into a large one, and so forth. Some CSCW research focuses on interdependencies at work and how practitioners deal with them. While some of these interdependencies result from the complexity of the object of work, others have to do with the fact that collaborative work involves multiple actors that bring different perspectives and types of expertise with them, with the result that ‘control’ is distributed (Schmidt 2002). The vocabulary offered by Langley et al.
(1995) helps us talk about some of these interdependencies, focusing on the implications of choices in a dynamic way.
However, we do not think that a decision-making perspective makes sense in all work domains. It captures a significant aspect of collaborative work in areas, where space for and competence in generating and further developing choices is critical to the work. These are domains of work where the design space is rather large and the process of arriving at a solution necessarily open (Wagner 2004). In these domains, understanding where the (alternative) choices come from, how and by whom they are selected, and evaluated, is important. In a health care setting, for example, decision-making is crucial, given the complexities and uncertainties of diagnostics, the nature of patients’ illness, which may involve dramatic changes requiring urgent interventions. Advocates of participatory medicine may face
similar kinds of challenges as the ones we have described, having to ask:
participation in what? The idea is that patients, their families or friends should assume an important role in diagnostic and therapeutic processes and that a ‘participatory result’, involving patients’ active participation, may in many cases be a better one.
PD is characterized by a particular organization of the design in that users are seen as co-designers in many – if not all – design activities (Bratteteig et al 2012).
The particular organization of design is supported by methods and techniques that enable users to take that role. The way of practicing participation, inviting users and designers to collaborate, also has to do with the actual people involved and the collaborative spirit they create together (see e.g. Light and Akama 2012).
Controversies and conflicts – if they exist – must be dealt with as the final design result often implements one perspective. A participatory design result is also the result of a social work process where the participants managed to share (some) power between them.
Our last point concerns the concept of power. Our analysis has focused on the ‘power to’ of the different participants in a project, which is based on skills, experience but also authority and position. We have examined (institutional) givens that frame the design space: from commitments frozen in contracts, budgets and temporal constraints to rules and expectations of the outside world of funding agencies, project partners or participating organizations. Power is an explanatory concept: it helps see why some things are done in a certain way and not otherwise. In parts this ‘why’ points at particular people, their skills (which we may think of in terms of power/knowledge), or their authority; in parts at structural constraints and opportunities; in parts at the ‘logics’ of a practice. The notion of power suggests that the design result inter alias depends on a specific constellation of actors and resources.
8 References Arendt, Hannah (1970). On Violence. Boston: Houghton Mifflin Harcourt.
Arnstein, S. R. (1969). A ladder of citizen participation. Journal of the American Institute of Planners, vol. 35, no. 4, pp. 216–224.
Balka, Ellen; Pernille Bøjrn; and Ina Wagner (2008). Steps Toward a Typology for Health Informatics. In CSCW`08. Proceedings of the 2008 ACM Conference on Computer supported cooperative work. San Diego, CA, USA, November 08 - 12,
2008. New York: ACM Press, pp. 515-524.