The Methodology of Participatory Design
Total Page:16
File Type:pdf, Size:1020Kb
APPUED THEORY SUMMARY • Provides the historical and methodological grounding for understanding participatory design as a methodology • Describes its research designs, methods, criteria, and limitations • Provides guidance for applying it to technical communication research The Methodology of Participatory Design CLAY SPINUZZI INTRODUCTION to design rather than a rigorous research methodology. echnical communicators have begun writing In this article, I discuss participatory design as a re- quite a bit about participatory design, sometimes search methodology, characterizing it as a way to under- with a fervor that rivals that with which we used stand knowledge by doing, the traditional, tacit, and often Tto write about T-units or think-aloud protocols. invisible (in the sense of Nardi and Engestrom 1999; Muller The terms participatory design and user-centered design 1999) ways that people perform their everyday activities are being broadly applied in the philosophical and peda- and how those activities might be shaped prociuctively. I gogical work of technical communication (Blythe 2001; first define and describe if ^af participatory design research Henry 1998; Johnson 1998; Salvo 2001; Spinuzzi 2003); is. I describe participatory design research in terms of its methods associated with those terms are being applied in paradigm, methodology, research design, and methods. technical communication research (Mirel 1988, 2003; Smart With this definition and description as a framework, I next 2003; Smart and Whiting 2002; Smart, Whiting, and discuss why we should pursue participatory design studies. DeTienne 2002; Spinuzzi 2002a, 2002c, in press; Wixon and In this section, I discuss the benefits of knowledge by Ramey 1996); and prototypes in particular are often pre- doing and provide evaluative criteria to use as guidelines sented as a vital part of iterative usability (see, for example, for creating and assessing participatory (design research. Barnum 2002, Chapter 4; Smart and Whiting 2002). But that Finally, I explore the implications of understanding partic- breadth of application has often come at the price of ipatory design as a research methodology, and I discuss imprecision. It's hard to find a good methodological expU- some practical applications. nation of participatory design. That lack of a strong methodological explanation is not WHAT IS PARTICIPATORY DESIGN RESEARCH? just technical communication's problem, though. Participa- Participatory design is research. Although it has sometimes tory design is often discussed in human-computer interac- been seen as a design approach characterized by user tion, computer-supported cooperative work, and related involvement Qohnson 1998), participatory design has its fields as a research orientation or even a field (see Muller own highly articulated methodological orientation, meth- 2002, p. 1,052) rather than a methodology. The distinction ods, and techniques, just as does participatory action re- may be important in principle, but in practice, it has be- search, the approach on which it is based (Glesne 1998). come an escape hatch that allows practitioners to label Implementations of participatory design do vary in their their work "participatory design" without being account- attention to rigor and validity (Spinuzzi in press), but they able to established, grounded prece(dent. all reflect a commitment to sustained, methodical investi- By looking at that established precedent, I argue, we gation according to grounded methodological principles, can define participatory design as a methodology, even if as we'll see below. it's a loose one. And I believe it's time we did: Without such Participatory design is rather different from most re- a definition, we can't hold ourselves accountable to partic- search conducted by technical communicators, though it ipatory design or build on a coherent body of knowledge. Consequently, we have trouble applying participatory de- sign rigorously to our technical communication projects, Manuscript received 11 August 2004; reviseci 24 November and we tend to think of participatory design as an approach 2004; accepted 28 November 2004. Volume 52, Number 2, May 2005 • TechnicalCOMAlNCATION 163 APPLIED RESEARCH The Methodology of Partioipatory Design Spinuzzi turns out to be a good match for the work we do. As the research. That's especially true in studies of workers, for name implies, the approach is just as much about design— which participatory design was initially designed, but also producing artifacts, systems, work organizations, and prac- in studies of end users and students. tical or tacit knowledge—as it is about research. In this methodology, design is research. That is, although partici- History patory design draws on various research methods (such as Participatory design originated in Scandinavia in the 1970s ethnographic observations, interviews, analysis of artifacts, and 1980s. This early Scandinavian work was motivated by and sometimes protocol analysis), these methods are al- a Marxist commitment to democratically empowering ways used to iteratively construct the emerging design, workers and fostering democracy in the workplace. This which itself simultaneously constitutes and elicits the re- avowedly political research aimed to form partnerships search results as co-interpreted by the designer-researchers with labor unions that would allow workers to determine and the participants who will use the design. the shape and scope of new technologies introduced into Like member checks in ethnographic research, partic- the workplace. Up to that point, labor unions had little ipatory design's many methods ensure that participants' experience with computer technologies and had been interpretations are taken into account in the research. Un- forced to accept systems developed by management, sys- like member checks, however, these methods are shot tems that represented a sharp break from workers' tradi- through the entire research project; the goal is not just to tional ways of working; exerted a greater and greater con- empirically understand the activity, but also to simulta- trol over increasingly fine details of their work; and neously envision, shape, and transcend it in ways the automated large swathes of the workflow, putting people workers find to be positive. In participatory design, partic- out of work (see Ehn, 1990; Zuboff, 1989). ipants' cointerpretation of the research is not just confirma- Since they did not know how to design computer tory but an essential part of the process. technologies themselves, workers were put into the posi- Participatory design started in Scandinavia through a tion of accepting these disempowering technologies or partnership between academics and trade unions. Since simply rejecting them. Some Scandinavian researchers set that time it has worked its way across the Atlantic, becom- out to develop a third way, an approach that provided a set ing an important approach for researchers interested in of "language games" (Ehn and Kyng 1991, pp. 176-177) human-computer interaction, computer-supported cooper- that would allow software developers and workers to col- ative work, and related fields. From there, it has begun to laboratively develop and refine new technologies—allow- influence writing studies, particularly through technical ing workers to retain control over their work. communication as well as computers and composition (for These researchers turned to action research, in which example, Sullivan and Porter 1997; Johnson 1998; see ethnographic methods are linked to positive change for the Spinuzzi 2002b for an overview). research participants (see Glesne 1998 for an overview). Participatory design has undergone many changes— Clement and van den Besselaar explain that for instance, later variations have moved away from the Marxist underpinnings of the earlier work—but its core has Untike conventional research, which is directedprimar- remained more or less constant. It attempts to examine the ity at producing resutts of interest to those heyond the tacit, invisible aspects of human activity; assumes that these immediate research site, an essential goat of action aspects can be productively and ethically examined research is to achieve practical or political improve- through design partnerships with participants, partnerships ments in the participants' Hves (e.g., less routine work, in which researcher-designers and participants coopera- greater autonomy, more effective tools). The researcher tively design artifacts, workflow, and work environments; hecomes directly involved in the ongoing work and feeds and argues that this partnership must be conducted itera- results back to the participants. (1993, p- 33) tively so that researcher-designers and participants can develop and refine their understanding of the activity. The Action research involves alternating between practical result of the research typically consists of designed artifacts, work to support changes (such as design activities) on one work arrangements, or work environments. hand, and systematic data collection and analysis on the As Pelle Ehn suggests, participatory design attempts to other hand (p. 33; see also Bertelsen 2000). In early par- steer a course "between tradition and transcendence" ticipatory design studies, workers tried to describe com- (1989, p. 28)—that is, between participants' tacit knowl- puter systems that could automate work while still valuing edge and researchers' more abstract, analytical knowledge. their craft