Sign In

Communications of the ACM


When Is ­User-Centered Design Selfish?

View as: Print Mobile App Share:
Judy Robertson

In 2005 Don Norman wrote: "Human-Centered Design has become such a dominant theme in design that it is now accepted by interface and application designers automatically, without thought, let alone criticism. That’s a dangerous state — when things are treated as accepted wisdom." I find this statement refreshing, because I am beginning to find the juxtaposition of user centred design (UCD) with recent trends in HCI towards sensitive application areas a little worrying. The benefits of UCD are well accepted: better appreciation of user’s needs resulting in better designed products which are more suitable usable and satisfying, empowerment of users, in depth understanding of the needs of specific demographic groups built up within a research community. I have seen criticism of UCD on cost or resource grounds (often in the form of a straw man), but never on ethical grounds.

Who benefits from user-centered design according to standard wisdom? Designers and their employers benefit, because they end up with better products. End users (that amorphous generalised group) benefit, because their software-using lives are more satisfactory. Researchers benefit, because they get papers published about their thoughtful and inclusive design methodologies. What I want to know is whether particular users who contribute to the design process actually get anything out of it? And do they stand to lose anything?

Previous authors have argued that Scandinavian employees benefit as participant designers because they get to participate in work place decisions. Druin has claimed that children reap the benefits of being in intergenerational design teams. It’s quite easy to argue that contributing to redesigning the fiendish technology inflicted on us by our employers is satisfying, or that participating in a design activity has educational value. But let’s face it, for ordinary users, taking part in a design task is time consuming and may intrude into an already crowded schedule. Some people might find it tedious ("not another packet of post-it notes and a blank wall!"). Others may find it threatening ("I don’t want to show how little I know about technology"). In the end, it might be disappointing for those whose ideas don’t make it into the final product ("I wasted an hour of my life to have these jokers ignore my genius ideas?"), or those who are underwhelmed by the rough edges of a research prototype ("you realise this smart phone is made of paper?").  These are mere irritations, presented cynically and not entirely seriously. However, in cases of user-centered design work in sensitive areas involving vulnerable users there are some genuine ethical concerns which I think we need to talk about.

CHI has been expanding into new domains recently, with the laudable goal of using technology and design to improve people’s lives. But as we move from research about productivity systems in the work place to technology to improve the quality of users’ lives - in domains such as farming assistance in the developing world, health care, environmental disaster management, reminiscence, homelessness and end of life care- we must tread lightly.

We know from well established ethical principles that researchers should adhere to the general ethical principle of beneficence and nonmaleficence: by taking part in our design processes, users should benefit and should not be at risk of harm. We should not stray from our area of professional competency. We should not disrupt the lives of our participants unless we believe that the benefits outweigh the costs of the intrusion. But within the current set of user centred design practices, some of these ethical principles are compromised when working with vulnerable users.

Imagine a fictitious research project which aims to develop technology to help the bereaved share memories of those they have loved and lost. This is an academic project suffering from underfunding (as many do) and the designers can only afford to make a rough prototype which never makes it to market. The designers, realising that this is a very emotionally sensitive area, seek to consult users in order that the technology is appropriate, comforting and respectful of grief.  They interview a new widow on several occasions, and ask her to try a pilot application. They find this to be useful exercise and believe that their design is much improved because of it. They publish a paper about it, and other researchers are interested in the findings and it generates much discussion at a conference. Unfortunately, the user herself does not benefit from her contribution to the design because she finds that the software doesn’t run on her computer due to glitches she doesn’t understand. The researchers tell her that there may be a proper product developed in a few years, but she is finding it hard enough to get through each day just now, and doesn’t want to think that she will still feel this way in the future.  During the consultations, she finds it painful to talk about her loss, and tiring to keep concentrating on the design activity as she is distracted by worries about her financial circumstances. Originally she felt she ought to contribute to the project because she wanted to help other people in the future, but she is now increasingly upset by having relative strangers in her own home.

In this story we have well intentioned, respectful practitioners following best practices for design who nonetheless departed from ethical principles. They have unintentionally caused emotional distress without providing the benefits of a working system. They have intruded into the life of the user at a time when she was particularly vulnerable. They have been operating out-with their area of professional competency, as they were trained as designers and programmers rather than therapists. It is possible to imagine scenarios in other sensitive domains where similar ethical transgressions occur. What if physical rehabilitation technology trials aggravate injuries? What if involvement in a design project disrupts the participants’ social standing in a community in an unforeseen way? What if the time spent on design activities diverts vulnerable participants’ attention from addressing the key issues which are of important to them? I believe it is time to reconsider the best practices for design in sensitive areas from the point of view of the participant. Why should they help a design project? We cannot expect altruism on their part, nor a desire to further research knowledge. We must be careful that our own motivations for user-centered design activities are not selfish.




Martin Schmettow

In academic research it is common (or mandatory) practice to have planned empirical studies reviewed by an ethical committee. Up til now I thought this was not more than a formal requirement for usability studies. You proved me wrong.

Your example is compelling, however not the main stream. How about the increasing commercialisation of usability and HCI. For example, is it ethical as an HCI researcher to contribute to a funnel analysis, helping a company to persuade visitors to become consumers? How about the current trend towards UX? Some works sound more like consumer research, not longer user research [1]. Is it ethical to run such a studies on a sample of students for partial fulfillment of their courses? Is it ethical to implicitly suggest: software companies better hire more graphic designers, not usability experts?

[1] Porat, T., & Tractinsky, N. (2012). Its a Pleasure Buying Here : The Effects of Web-Store Design on Consumers ' Emotions and Attitudes. Human Computer Interaction, (September), 3741. doi:10.1080/07370024.2011.646927

Judy Robertson

Thanks for asking these broader questions. I certainly have some doubts about requiring students to participate in experiments for their studies, but I don't have much expertise about the commercial angle. I suppose there is a distinction between protecting the participants whose data was involved in funneling, and those future users would might encounter the results of this research in the new extra persuasive website.


I think these are some really interesting and through provoking points to raise. I also think that they are useful on a corporate level (as well as with our actual users). Meaning, when you give the examples above about "not another packet of post it notes and a blank wall" I associated this with our corporate customers and us pushing our user centered design methodologies onto them as well (and not always really fixing their problems). Not only to we have to keep the end user in mind (as you clearly point out), but our corporate and business users as well.

Thanks for writing this!

Displaying all 3 comments

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account