Monday, November 2, 2009

Participatory Design

As a usability/user experience professional working for the e-commerce team of a telecommunications company, I had heard of participatory design, but never knew much about it or worked with anyone who had experience in it. It sounded logical to me to involve end-users in the software design process. However, I had heard that it more specifically meant involving users in the design process directly, something that seemed problematic to me.

I guess it’s natural to me that involving end-users in the design process is essential. Doing so will help you to better define your requirements, to understand your end-users’ goals, needs and environmental contexts and to verify that what you are creating is on track. However, it was clearly noted in my mind that end-users (in our markets) are rarely software/website designers and thus do not have the training or advanced techniques/understanding that these professionals have.

For example, end-users may struggle with something and not have the words or knowledge to specify what the problem is exactly. However, a designer can observe their struggle and understand that a subtle fix can make a significant improvement. Still, it seems reasonable that obstacles like that can be challenged and domains can be bridged.

In fact, Michael Muller discussed the idea of a third space, a “hybrid of space between software professionals and end-users”. In this space where the two groups come together, assumptions can be questioned, knowledge can be exchanged, and a common language can be established to the betterment of both groups. He described Participatory Design as creating this third space which in turn can benefit the field of Human Computer Interaction. He then examined a variety of Participatory Design methods to see how well each contributed to the goals of hybridity.

Within his Participatory Design chapter, Muller summarizes the claims of third spaces, which include, among other attributes:
  • Questioning and challenging of assumptions
  • Mutual learning
  • Synthesis of new ideas
  • Negotiation and (co-)creation of Identities, working language, working assumptions and dynamics, understandings, relationships
  • Dialogue across and within differences
The elements of Participatory Design are intriguing – a dialogue between people from separate domains facilitated by hands-on activities. Issues that come to my mind as I consider how I might apply this to my work are practical ones such as, how many Participatory Design sessions in a design process, where in the lifecycle does this fit, how many end-users, how to recruit/screen, how to analyze results and turn them into actionable outcomes, how could I make a case for this at my company. Muller identifies a few as well, one of which is the concern for universal usability. He points out that nearly all of the Participatory Design methods that he examined are “highly visual and require hands-on manipulation of materials”, making them not accessible to many people with limited visual or motor skill abilities.

A recent paper by Loebbecke and Powell (2009) discusses Distributed Participatory Design and the need for it to evolve by learning from other collaborative design methods. They point out that Participatory Design was initially developed for groups of software professionals and end-users to meet physically in the same location, which was appropriate for the time. With the common practice today of virtual teams who may be in separate countries and time zones, Distributed Participatory Design has developed. The authors compared Distributed Participatory Design with other collaborative design methods (Distributed Action Research and Distributed Design Science) to discover similarities between approaches that might be beneficial to all.

They identified several issues that practitioners have with Distributed Participatory Design, physical distribution (separation of people and resources), organizational distribution (work structure, differences in skill levels, knowledge levels) and temporal distribution (limited time teams can meet virtually given differences in time zones).

Loebbecke and Powell describe the Action Research method as one in which some change is introduced to complex social processes followed by observation of those processes and any effect(s) of the change. The steps are part of an iterative process and involve (1) Understanding and diagnosis of the situation and its underlying dynamics, (2) action planning, (3) intervention, (4) evaluation, and (5) reflection.

The authors then describe the Design Science method as concerned with creating something new and innovative to enhance HCI often to solve a business problem/need. They present seven principles suggested by Hevner et al (2004): (1) ‘Design as an artifact’, i.e., producing a viable artifact (construct, model, method, instantiation), (2) ‘problem relevance’, i.e., searching for important solutions for the business world, (3) ‘design evaluation’, i.e. assuring quality and utility of an artifact, (4) ‘research contributions’, i.e., reflecting upon the design (how did she/he contribute to the body of knowledge he used?), (5) ‘research rigor’, i.e., rigorously applying methods along the process, (6) ‘design as a search process’, i.e. stressing the dual imperative between solutions and environmental constraints, and finally (7) ‘communication of the research’ underlining the double audience of stakeholders and research community.

After conducting a textual analysis of the three approaches looking at publications and projects, Loebbeck and Powell found many commonalities (research focus, outcomes, and research process) and differences (terminology, references, and audiences). They were concerned with the separation of these areas, referring to them as “walled gardens”.

They suggest that these approaches may be better considered as paradigms than methods. They also recommend that researchers of Distributed Participatory Design would benefit from looking to Distributed Action Research and Distributed Design Research for inspiration.
“This research suggests that there is a lack of cross-fertilization from approach to approach, and that walled gardens exist or are starting to emerge. Thus, there is a need for the walled gardens of different methods to at least have windows so that knowledge can be exchanged and ultimately the walls need to be removed. However, as demonstrated by this research, the walls are high at present and, though there is some evidence of cracks in the windows, distributed PD may miss potential sources of enrichment”.

Although I find that I still have many unanswered questions about this approach or paradigm, I am inspired by the potential that it and other collaborative methods have to offer.

Muller, M. J. (2007). Participatory design: The third space in HCI. In Sears, A. & Jacko, J. (Eds.). The Human-Computer Interaction Handbook: Fundamentals, Evolving Technologies and Emerging Applications, 2nd Edition. (pp. 1061-1082). Lawrence Erlbaum.

Loebbecke, C. & Powell, P. (2009). Furthering Distributed Participative Design. In Scandinavian Journal of Information Systems, 21:1 (pp. 77-106).

No comments: