Data session

This is an exercise in which we use a short piece of video and guide the students in making observations. We have used video data we recorded in a call center some years ago. Using multiple video cameras we recorded the phone-call between the telerep and the customer as well as the ambient sound in the call center, the interaction with the computer system, and a wide-angled shot which captures the telerep in his cubicle environment. We provide the students with a transcript and play the video several times. We prompt them to make observations, giving hints if they have difficulty. The video has several key points. For instance, the telerep reads a wrong line, which a casual observation cannot reveal. We guide the students to notice the mistake by reviewing the video carefully and then let them state analytically the practical reasons – pushing them to think further than to attribute the mistake to the rep’s lack of capability—such as pressure not to keep the customer waiting, the design of the system and various documents, which the rep had to use quickly and sequentially.

Ethnographic interviewing

This is a lecture on how to conduct ethnographic interviews (as opposed to the more formal, highly structured – and thus more serially interrogative – kind). SEs may be familiar with doing interviews, but if so those interviews are conducted to specify what kind of functionality the customer requires, rather than focusing on the work per se. We stress that the goal of the ethnographic interview is to get the interviewee to talk – an ethnographic interview is about the interviewee not about the interviewer. Although you must prepare an interview guide with topics and perhaps even specific questions, interviewers should be flexible and it is a good practice to base next questions on what they just heard from the interviewee rather than doggedly following the prepared list of topics. In ethnographic interviews it is better to ask broad, descriptive questions, which will naturally lead to the interviewee giving longer answers that the interviewee him/herself can design. Using a transcript of an interview that one of us conducted we illustrate how to conduct interviews to get rich information by being ‘persistent but polite,’ asking for more detailed descriptions from the interviewee. Finally, we talk about some practical matters such as how to set up interviews, how to do a proper interview introduction, to bring recording devices and how to ask for permission to record, and to write field notes as soon after the interview as possible because many details will fade from memory quickly.

Fieldwork planning

This is a lecture containing some practical tips for planning a fieldwork engagement. First, we remind the trainees of the advantage of fieldwork over other methods and techniques which may be used to persuade members of an organization that fieldwork is a good idea. Since organizations are hierarchical, it is usually a good idea to start by doing interviews with higher level manager and work your way down until you come to the level of the people that are the target of observations. In getting an introduction to the organization it is very helpful to get a manager to give you a ‘Grand tour’ in which they walk you through the organization and introduce yourselves to various members. When creating a fieldwork schedule you should consider how much time you have, how many people you can ‘sit’ with, and who those people should be. It is important to allow for enough time to write field notes and do analysis and not just fill one’s time with making observations. Finally, we talk about how to report back to the organization and how it is important to protect individuals by not revealing their identities even if it means not telling your strongest stories. We remind them that fieldworkers are not there to debunk the organization they investigate, but to be neutral and take an ethnomethodologically indifferent attitude.

Co-design

This is a lecture on the approach to organizational interventions. Since our audience was SEs who develop new systems we also developed a lecture on co-design: the process of working closely with users to design new technology. In the lecture we talk about the rationale for co-design as well as its principles. We argue that a system designed by the users may well be better suited to their work than a system the designers come up with and more readily accepted and adopted. We elucidated some of the principles for co-design inspired by literature in participatory design such as the work by Schuler and Namioka (1993). As we do not normally have sufficient time to illustrate the many techniques of participatory design, we use Muller, Wildman, & White’s taxonomy (1993) to refer the students to literature where they can learn about different techniques.

Writing field notes

In this short lecture we make a distinction between writing while doing fieldwork and writing field notes afterwards. The prime purpose of writing in the field is to jot things down so that they can be remembered later; there is no need for elaborately written down observations. We urge the trainees to make detailed notes about the environment and to record conversations as literally as possible. Also, we talk about when it might be wiser to refrain from making notes (when your subject is telling you something potentially embarrassing, for instance). When returning from the field the first stage of writing field notes is to get as much down as possible and not to worry about the readability of your notes for other people, these are notes that capture your own observations, so it does not matter that not everything coheres. For more ‘finished field notes’ different styles can be used (van Maanen, 1988).

Fieldwork exercise

In order to let the students experience the value of fieldwork we designed the exercise to mimic the situation they would face if they conducted fieldwork in a customer site. We sought cooperation from different organizations within Fujitsu, focusing on organizations whose work was quite different from the work of SEs. We did fieldwork in the mailing room, the security offices, with secretaries, and the kitchen staff in the cafeteria. The students were divided into small groups of 4-5 students who had to work together during the analysis and presentation parts of the exercise. The level of access the students had to the subjects varied greatly as some organizations did not want the students to interact with the subjects at all, a situation we have tried to address with only limited success. We accompanied the students to the field and answered any questions they might have and sometimes even did some fieldwork themselves.

After the fieldwork, which lasted for about two to four hours, the students would write up their field notes individually, and then do analysis in small groups. We would sit in on these groups and work with the students and help them develop their analysis. On the third day of the course, the students would develop presentations of their observations and present them to managers of the organizations whose members they had observed. The requirement that they present to a manager created considerable pressure for the students and they invariably took the fieldwork exercise quite seriously. Invariably, these managers would be impressed with the detail of the things the students had observed and this reaction helped us considerably in convincing the students of the power of ethnography.

RECENT DEVELOPMENTS

The cadre of SEs we initially trained in our project is to carry out fieldwork at customer sites and then present to the customer how work is actually done at these workplaces, as well as what kind of problems the customers’ organization needs to solve in order to improve their current performance. While they expect their findings will eventually lead to customer orders for information systems they try not to narrow their observations to members’ use of technology.

As they accumulate their experiences, they started customizing fieldwork into a more systematic method in order to cope with constraints they have on the overall schedule. To help them we participated in these customer engagements at first, but as they gained more experience we started to review only the results of their work. The challenge was for them to go beyond merely recording their observations to deeper analysis, a skill that only comes with practice (and talent, certainly). Nevertheless, they have developed a quite effective way of doing team ethnography. Further, they have also now taken over teaching our course on ethnography to other members in the organization. As they accumulated experience and cases, and the fieldwork service continued to take shape, it became obvious that students should learn what they will be expected by customers to do, based on Fujitsu’s most recent experience.

CONCLUSION

Naturally it is expected that the professionals in one area would react to a new method, especially if the particular method comes from a different discipline. There are genuine differences in the approach to work between ethnographers and system engineers. These differences do not just have to do with introducing a novel method to their work, but have to do with what is considered competent work in the different communities; they are deeply moral issues that cannot be glossed over lightly. In our experience, the most successful way to overcome these differences is to have the system engineers experience first hand the value ethnographic observations can yield for their own work. Ultimately, the experience of teaching ethnography helped us understand system engineers work better and helped shape our message when we reported our ethnographic results; it helped us to be heard by the organization.

Inspection of projects at the maintenance and operation phase was another area where Fujitsu applied fieldwork methods (Obata et al 2007). This made it possible to look at projects operating in this phase more closely and understand why certain troublesome issues persist that may escape the standard checklist based inspection method. Thus fieldworkers were able to make managers reconsider the existent policies when they were presented with some compelling fieldwork findings. In both cases, it was SEs and Fujitsu Laboratories members who customized the fieldwork method in order to meet their own needs. It was a very positive development. After all, they know far more than we about SEs’ work and are therefore in the best position to decide how to implement fieldwork methods within their own work environment and practices. For this reason, fieldworkers in Fujitsu should be in a very good position to overcome some of the issues frequently raised in the CSCW community about making an effective bridge between design and workplace studies (Plowman et al 1995; Dourish 2006). As far as we know Fujitsu is the only software development business to train and operate a fieldwork team inside the business itself, staffed and managed by SEs (versus by researchers in a laboratory organization that must then find some way to cooperate with and effectively support the business side). We believe this is a huge step forward for ethnographic praxis in industry.

Ethnographic Praxis in Industry Conference Proceedings 2007, pp. 271–283. © American Anthropological Association, some rights reserved.