When I was studying the use of mobile phones among university students in Tanzania, I spent many days waiting for interviewees who failed to show up, without explanation. Many other interviewers, even local ones, reported having the same problem. Later I learned that the easiest way to meet users was to select a random chair anywhere in the campus, sit down, and wait. It would not be long before one of the local students would come to talk to me. Try the same at a busy Finnish university campus, and the chances are the only person who will approach you that day is a guard who wants you to leave so he can lock the doors for the night.
This is just one example of the kinds of cultural encounters that make cross-cultural interaction design both a fascinating and frustrating experience. Learning about users and context is always a part of a designer's work, but in cross-cultural design the task is more challenging because a designer often lacks the background knowledge about environment, infrastructure, politics, laws, culture, and historical and current events in the society in which users livesomething a designer can take for granted when designing "at home." Such knowledge is needed in planning a design project and for deciding which techniques to use and how to adapt them to match local practices such as communication and management stylesnot to mention that designing without this knowledge would result in isolated systems that are less likely to become successful long-term solutions.
Spotting the differences in how artifacts look or how people use them in their daily lives is not too hard, but there is more to culture than that. The most challenging cultural differences, from a design point of view, are those that cannot be seen. Based on the anecdote described earlier, for example, it would be easy to conclude that Tanzanian students are terrible at keeping meeting times, and Finnish students are so inconsiderate that they ignore other people. Yet there is something more underneath those behaviors: cultural ideas and values related to what is important in life and the best ways of dealing with daily struggles.
As ideas and values behind people's actions are not directly observable, they are easily missed by a designer who does not share them. This can lead to ethnocentric thinking: judging everything against one's own values. In cross-cultural design, it means there is a risk of misunderstanding users and making design decisions based on what designersnot usersvalue.
When culture and values are discussed in cross-cultural HCI, they are often discussed in terms of cultural dimensions, such as those of Geert Hofstede . Cultural dimensions describe cultural values using concepts such as uncertainty avoidance and individualism versus collectivism. They have been found to be relevant in both human-computer interaction and interaction between users and designers. As a design aid, however, cultural dimensions are in many ways problematic.
Cultural dimensions are based on a statistical analysis of questionnaire answers from a large number of people in different countries. Typically, researchers study a particular group of people (in the case of Hofstede, employees of IBM), but the results are generalized across a country or even beyond. In the case of Tanzania, the five scores on Hofstede's cultural dimensions are supposed to describe not only national culture in a large country with more than 100 tribes (and a variety of denominations), but also three other nearby countries: Ethiopia, Kenya, and Zambia. In the same way, even though cultural dimensions are based on questions related to a particular area of life, such as work, it is assumed that the differences found between countries are similar in other contexts, too.
Due to their abstractness, cultural dimensions cannot tell us how cultural values are put into practice . The similar scores of Finland and Tanzania in uncertainty avoidance, for example, suggest that members of these countries would feel equally uncomfortable in unknown situations. But in practice, Finnish and Tanzanian people use different techniques for dealing with the unknown: When faced with strangers, Finns observe the situation silently from a safe distance to "let other people be in peace" , while the Tanzanian strategy is perhaps geared more toward initiating contact and using talk to gather information to reduce uncertainty.
Cultural dimensions, or other similar general concepts, can never depict the richness of user groups and usage situations in a country. In Tanzania, for example, a country labeled as collectivist in Hofstede's classification, local university students do use their mobile phones collectively in many ways. But there are also many situations in which the students see their mobile phones as personal devices and are unwilling to share them. A husband or a wife may not feel comfortable with sharing the phone with his or her spouse; mobile phone thefts and the use of mobile phones for conveying threats and insults have made people careful about lending their phones to people they do not know well; and having to act as an intermediary between people who do not have a mobile phone of their own is sometimes experienced as a strain.
In a similar vein, cultural dimensions cannot provide easy answers to what design techniques should be used in cross-cultural design projects. While I was studying the use of mobile phones in Tanzania, I learned that regardless of the assumed collectivist tendencies, there were situations in which individual interviews worked well and group interviews did notit depended on many situational factors such as who the interviewees were, what kind of relationship they had with one another, how the interviewer was perceived, and in what kind of environment the interview was held.
In brief, cultural dimensions as a design aid fell short of providing the guidance I needed in the field in Tanzania. They did not give specific enough information about the wider context of use and could not predict which user-research techniques would be most useful. Being very generalized descriptions, they did not pinpoint what mattered most to Tanzanian students as users of mobile phones, and enforced rather than guarded against overgeneralizations. For these reasons, I turned to ethnographic fieldwork.
The approach I used for doing user research in Tanzania was based on a long-term, flexible research process, in which several user-research techniques were combined and the early design efforts were focused on understanding the general patterns of technology use and ownership, the user values and the context behind them, and the design context.
Long-term, flexible fieldwork. I developed the required understanding about local context and values through a relatively long period (six months) of fieldwork. The fieldwork gave me more accurate and useful information for planning the design activities and made it possible to pay attention to both cultural patterns and differences across contexts and user groups.
I found long-term fieldwork effective for many reasons. To begin with, it was my first time in Tanzania, and there was not much information available about conducting user research in the particular design setting. In such situations, learning about the wider context of use and finding the best techniques for user research takes time. In addition, lengthier time in the field provides a chance to get better data and deeper understanding, because early interactions with informants in the field are often characterized by misunderstandings and uncertainty .
Although nothing could have taken the place of the personal experiences I gained in the field, there were some opportunities to complement the fieldwork. In a mobile phone study, I could access some Tanzanian newspapers' and mobile phone operators' websites even after I had left the field; I found them to be good sources for learning about Tanzanian society, locally relevant issues, and attitudes toward ICT. And even though I'd previously had some doubts about the usefulness of long-distance techniques in cross-cultural user research, I realized that after getting some first-hand experience about the field and users, the chances of continuing successful user research through email or mobile phone would be much greater.
Another essential feature in the fieldwork was flexibility. When there was no previous information about the design setting, a flexible process in which the choice and details of techniques were decided in the field along the way turned out to be the best approach. Flexibility combined with sufficient time made it easier to deal with the unexpected events, problems, and changes of plans common in cross-cultural design.
Combining several techniques. In a situation in which it was not often clear whether the chosen design techniques would actually work, the use of several techniques was above all a precaution: Even if one of the techniques resulted in disaster, the entire study would not be at risk. There were many other advantages, too. Each of the techniques informed the subsequent ones, revealing contradicting viewpoints, concepts that were relevant to the local users, and interesting issues that needed to be studied in more detail. As such, they helped in planning and focusing the subsequent research activities and made it possible to question the first interpretations of the data.
Techniques were also used to complement one another. A diary study of students' mobile phone use, for example, did not provide data that would have been suitable for quantitative analysis, but it revealed some important usage issues, such as beeping and the use of multiple SIM cards. Questions about such practices were included in a questionnaire study, which revealed how common they were among the students and whether some user characteristics, such as age or gender, had an effect on them. After the questionnaire, I could discuss the reasons behind the observed patterns and differences in greater depth with users in a series of interviews. It was thus possible to recognize important factors that shape mobile phone ownership and use in Tanzania while also highlighting the differences between user groups.
Learning about the context of design was challenging. The country-level guidance offered by cultural dimensions was too abstract. More specific information could be found in the field, but there were sometimes controversies, both in the accounts of people who had conducted research in the same field before and in the ways in which different local stakeholders preferred the research to be conducted. Experimenting with many different techniques turned out to be the best way of doing user research and gaining information about the design context for the later phases of the project.
Studying and representing value. To elicit hidden cultural differences and question the ethnocentric views of the designer, there was a conscious focus on value  in the user research. I studied the positive and negative value the local students associated with mobile phones mainly through interviews, supported by observations about how mobile phones were presented and discussed in the local newspapers and on operator websites.
In interaction design, designers need to understand what the users value, but they also have to convey that understanding to developers and other stakeholders, many of whom may never have experienced the field. In addition, user requirements or needs alone may not be enough to convince stakeholders who are not familiar with the background behind the needs . Thus, there is a need for value representations that show what is important to users and why.
In this project, I used a value map (shown here) that illustrates both the benefits and drawbacks of mobile phones as seen by university students in Tanzania. Pros and cons were often two sides of the same coin: For example, students saw the reachability provided by mobile phones as a positive thing because it allowed them to stay informed about family matters at all times, but they also saw it as a negative thing because it distracted them from their studies. The positive and negative values in value maps were supplemented by contextual factors that explained why users found particular issues relevant. In the case of Tanzanian students, the positive value of mobile phones was highlighted by the significance of family and relatives as their main support network, the fact that students were often studying far from their families and were able to visit them only between the school years, and the way in which information at the university was spread through personal communication.
|Value map used in the project.|
Value maps such as the one described here differ from worth maps  in that they describe the value that users currently associate with existing technology, which can be of use in the early stages of design. By giving intangible values a form, value maps can provide a starting point for negotiations with stakeholders about what really matters to them.
Value maps make it easier for designers and developers to remember and appreciate user concerns, even when they are not familiar with the users' cultural context. Unlike cultural dimensions, value maps are based on fieldwork in the context of use, focus on technology, and are not limited to any predefined set of values. As such, they can provide more accurate representations of value for design purposes.
As a researcher, I found doing user research in Tanzania both a challenging and rewarding experience that provided many insights into interaction design. In a similar way, when looking at a future of HCI that appears more cross-cultural than ever, one can see many problems but also plenty of opportunities. Learning how culture affects what matters to people helps designers better understand their users. At the same time, it can challenge designers' own beliefs about what technology is for and how it should be designed.
6. Honold, P. Culture and context: An empirical study for the development of a framework for the elicitation of cultural influence in product usage. International Journal of Human-Computer Interaction 12, 34 (2000), 327345.
Minna Kamppuri is a postdoctoral researcher in the School of Computing at the University of Eastern Finland. Her research concerns cross-cultural design and inclusive design.
©2012 ACM 1072-5220/12/0300 $10.00
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 © 2012 ACM, Inc.