David Siegel, Susan Dray
As our profession (or any other for that matter) evolves and people enter it from related but different disciplines, it naturally differentiates into subspecialties. Simultaneously, roles in organizations evolve and differentiate. This tends to produce increased richness and complexity, but can also lead to unfortunate barriers that fragment knowledge. Sometimes in organizations, this type of evolution leads beyond fragmentation to actual schism.
We are now seeing a potential schism in our field that threatens to undermine the ability of UCD to introduce a holistic and comprehensive focus on the user into all phases of technology design. The potential schism we are focused on in this column arises from two things: the institutionalization in large companies of usability as a "testing function," which defines it as being downstream in the design process, and the arrival on the scene of ethnography as a separate user-centered design discipline.
Ethnography and usability focus on aspects of information about people that must be integrated for design to be successful. This is evident when you consider that both utility and usability are interdependent and necessary for product success. Utility requires usability, and usability is meaningless unless it is usability for something worthwhile. Ethnography should provide insights into what people really need and how it will fit into the dynamics of their lives, and usability should help ensure that people can successfully obtain the intended benefits of the technology. Thus, it seems that these two disciplines should be fully intertwined.
However, there are indications that this integration may be threatened. This issue came to our attention as we have heard conversations suggesting that there is potential for ethnography and usability to develop a contentious relationship, including comments denigrating the other camp. For example, ethnographers can be accused of providing input that is too abstract and general to be useful. They can be viewed as overly reliant on anecdotal information and too lacking in rigor. Ethnographers can see usability people as too tied to evaluating specific features without questioning whether the feature should exist in the first place.
Both of these sets of charges can be turned into negative stereotypes. Ethnographers are supposedly "fuzzy in their thinking," and "don't understand design or technology." Conversely, usability people are "so focused on technology features that they are hardly better than technocentric designers and developers," and "are incapable of stepping back and seeing the big picture." We have also heard ethnographers and usability people defend the barriers around their respective parts of the UCD continuum. This includes ethnographers saying that deeper knowledge of features would bias their observations in the field and take them away from their focus on high-level dynamics. Conversely, we have heard usability people limiting their focus to feature usability and leaving consideration of feature value and utility to others.
Only a minority of companies is large enough to have differentiated ethnography and usability roles. But we don't think these should just be dismissed as turf battles particular to any one company or set of personalities. We think the risks are worth noting for UCD practitioners in general, because the potential for the split is inherent in the nature of the subdisciplines and the general dynamics of professional differentiation we mentioned earlier.
For one thing, schism can arise from the tendency of related subdisciplines to accentuate their differences as they stake out their turf, claim legitimacy, and emphasize their unique contributions. It is natural for them to fight for recognition and to protect what is unique about them. Each of them has to justify its existence (and budget) separately and explain why the other camp does not meet the need.
Second, both groups are subject to pervasive forces in the larger organization that threaten to dilute their concept of quality. These include budgetary and commercial realities; organizational politics; the fact that their audiences and cross-disciplinary team members may have only a superficial understanding of what goes into good ethnographic or usability research; and other factors. Therefore, both groups have to defend the integrity of their own methods. Sometimes an unfortunate byproduct of this can be as one subdiscipline defends its own skills and expertise, it tries to define limits around the skills and expertise of related subdisciplines, and seemingly natural allies end up undermining each other.
Third, there are fundamental paradigm differences between ethnography and usability that are almost bound to produce clashes. These differences are all traceable to the part of the UCD continuum that the two groups occupy. Let's look in more detail at the underlying differences between ethnography and usability that can drive the split. Table 1 presents one way of summarizing and organizing these. It is an obvious oversimplification, and like any such table intended to emphasize contrasts, it risks veering into caricature. Hopefully, it will be instructive even though it cannot be comprehensive or adequately nuanced.
First, there are a number of obvious differences in what the two fields are attempting to do. They have contrasting focuses, and these drive different approaches to data collection, different styles of reasoning, and different types of outputs. Second, these approaches call for different skills and orientations on the part of the practitioners.
Next, there are differences in what these approaches produce as an input to the design process, and how and when the information is introduced. Ethnography provides insights that are most useful in product planning and high-level conceptual design. Usability provides input into narrowing and refining design choices, and ideally, it should also play a role in generating design options. One consequence of this is that the two groups may have different audiences. Ethnographers may be talking more to product planners early in the process and usability people to designers later in the process, when the basic design is a fait accompli. Even though ethnography and usability are partners in UCD, they may not collaborate as much as they should.
These differences in organizational roles can accentuate and rigidify the differences in perspective. The fact that the methodological limitations (and complementarity) of the two approaches has to be understood and managed is nothing new. However, this becomes harder when separate camps in the organization "own" the different methods. Once a professional distinction has been reified by the organization, people who want to work in a more integrative manner, such as usability people who want to do field studies, may be viewed as stepping on someone else's turf.
A schism such as this can be damaging to both usability and ethnography, because of their complementarity. In fact, the two sides can potentially undermine each other. More important, this is not useful to the overall goals of UCD. Even though there are paradigm differences and possibly even organizational incentives to carve out different turf, we must remember our interdependence. The differences may seem great from our "local" perspective within the UCD professions, but when viewed in the larger context, they are mere nuances (see Figure 1). In fundamental contrast to most other drivers of product planning and design, both subdisciplines share a deep commitment to studying actual user behavior in contrast to basing conclusion on self-report, our own technology "brain storms," or prognostications about future directions in technology.
Furthermore, both subdisciplines, when isolated from each other, truly are limited in what they provide. It is very difficult indeed to pull clear and specific product implications from rich ethnographic information without having deep knowledge both of people and of the product. When we don't incorporate this knowledge and provide clear direction for designers, we leave them to guess at what personae, usage scenarios, and thematic clusters of user needs are telling them to do vis-à-vis design. Meanwhile, usability research does indeed need to be informed by ethnographic insights so that it is evaluating the right scenarios with the right people. Usability people need to be prepared to surface fundamental conceptual design issues and feature/value issues that can best be addressed by field research outside of their labs.
We need to avoid a methodological purist stance, and instead collaborate. For example, we should combine our approaches in naturalistic field research, or even in longitudinal field research. Furthermore, both usability and ethnography professionals need to understand the product planning and design process so that they can be active participants in tradeoff decisions with other disciplines that contribute to the ultimate product design. Only by avoiding a schism between ethnography and usability can we provide an integrated approach to product utility and usability. We have to work out the integration within our profession, because our customers, the technologically oriented people to whom we provide findings in order to guide their design efforts, certainly can't be expected to do it.
As president of Dray & Associates, Inc., Susan Dray consults internationally on interface design and usability. She is a pioneer in human-computer interaction and has given over 80 talks at conferences and symposia in the U.S., Europe, and Australia. Dr. Dray received her doctorate in Psychology from UCLA and is a board-certified Human Factors Professional. firstname.lastname@example.org
David Siegel is vice president of Dray & Associates and has been involved in all aspects of usability consultation and user interface design, including project planning, interface design evaluation, formal usability testing, and user studies. Dr. Siegel has a strong background in psychology, including neuropsychology, assessment and measurement of cognitive processes, and research methodology. He received a B.A. in psychology from Princeton University and a Ph.D. in psychology from UCLA. email@example.com
©2005 ACM 1072-5220/05/0300 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2005 ACM, Inc.