Ethnographic Persona: Section 3

In this section of this article, I will describe how ethnographic methods can be used in the development of information architecture and design. First, I will outline current research and practical experience in form of case studies, and second, I will present different types of ethnographic “products” that are widely found and could be a base for an enhancement.

Case studies
One of the most valuable and appreciated effects of ethnographic methods within information system design are their depth of data and insight (Myers, 1999, 5): “Ethnography often leads the researcher to question what we ‘take for granted”. But the downside of ethnographic analysis is that it takes much longer than many are willing to accept / to invest. As a consequence, we can find two types of ethnographic approaches to information systems: first, a research related approach, which is mainly conducted by PhD candidates (or other scientific personnel) during their studies, or second, a shorter efficiency related approach performed by consultants.

Robertson (2005) describes his holistic approach as needs analysis, which actually includes workplace observation, contextual inquiries and staff / stakeholder interviews among other methods. Interestingly, Robertson reduces workplace observation to a ‘junior role’ preparing the path to other methods: “In general, workplace observation is a good way to provide a broader context for more detailed research, such as one-on-one interviews or contextual inquiry.” As an example, observation of police officers in the office “would provide few opportunities to gain a deeper understanding of activities, issues and needs” because officers would spend the “majority of their time in Word, on email or using the phone.” Combined with contextual inquiry, Robertson sees the value of needs analysis of research “at identifying issues with currently-available information sources and tools”.

This type of method application is a good example of the efficiency related approach. It doesn’t try to analyse cultural aspects or the language, instead it takes the researchers background (Information Architect, and therefore how a system must be designed to work ‘perfectly’) and fills it up with applicable data gathered from the field.

Rose, Shneiderman and Plaisant (1995) propose a general framework based on ethnographic research to “reduce the likelihood of some common problems, such as overlooking important information and misinterpreting observations” (p. 115). The framework consists of 4 areas (preparation, field study, analysis and reporting) and serves as a guideline to projects when ethnographic methods are applied. But the case study, which was provided by the authors and explained their framework, showed a minimal use of ethnographic methods. Unfortunately, ethnography is seen by the authors as a set of new methods to gather qualitative and quantitative data, but within a typical “design approach”. The framework itself is a project pathway and is not very specific in how findings should be communicated to the design team.

A step further goes a suggestion from Hughes, O’Brien, Rodden and Rouncefield (1997). They propose a presentation framework that separates the work fields of ethnographers and designers and helps to communicate between the two groups. This said, the addressed problem is the form of communication of the ethnographer’s findings to the designer. In contrary to Robertson, we see here a split of tasks: ethnographers perform an ethnographic study, and designers do the information system design. Where design wants to reduce complexity through abstraction, ethnography wants to enrich the overall picture by details. Therefore, the suggested framework is a guideline how to present ethnographic data reduced to core concepts and key elements.

The authors of this framework suggest three dimensions of description: distributed coordination (work tasks in the context of division of labour), plans and procedures (organisational aspects of cooperation) as well as awareness of work (presenting the work of others). As I don’t want to go too much into the details here, one aspect is striking: the absence of language as important input for the design process. In my opinion, the framework could be extended to include this dimension, as the data (interviews, observations and other records from the research field) is available. The authors present extracts of interviews, emails and other documents that provide a rich semantic source for further exploitation and use in form of a domain analysis. But as the authors stress, “the framework is fundamentally motivated by the demands of design” (p. 148).

A final approach I want to highlight here is made by Simonsen and Kensing (1998). The authors present the role of ethnography in participatory design. Participatory design describes techniques that aim at “establishing a meaningful cooperation between designers and users” (p. 23). As Hughes, Robertson and others, Kensing and Simonsen use participant observation among other techniques to collect valuable data for the design process. As both authors are designers and not ethnographers, they see some value to exploit from the new method. But their approach is limited as well to the point, where it comes to face the information architects bias and the language used in the researched culture (corporation, group, …). The main idea of the authors here is to integrate users and designers (with some ethnographic know-how) into one team “to conduct an analysis and codesign an artifact” (p. 24).

Ethnographic descriptions for designers
As shown in the previous cases, many authors prefer a co-operating team of ethnographers and designers to create ‘better’ information architecture solutions. Unfortunately, only a few (Hughes, Simmonsen and others) tell how this relationship should work in detail. It is assumed that the ethnographers provide an ethnographic description to the designers as a cultural template for final realization of an information project. The question is: what is the content?

Spradley (1979) presents a spectrum of ethnographic presentations: from ethnocentric descriptions to ethnographic novels and many in between. The difference among them is their use of language, or better: the use of language of a specific culture. Ethnocentric descriptions on the one side of the spectrum neglect native language and even ignore what things mean. On the other side, ethnographic novels are written by natives themselves, composed in a “foreign” language and from an insider’s point of view including subtle meanings to things.

So, what does this mean at all? The point is, that the product of an ethnographic study is always a verbal description of something, mostly a culture or society or a group of people. Verbal description includes language, and most of time, the description is composed in the author’s language or in the language of his / her readers (here: designer). As the author’s language is different from the observed cultural scene, how can the ethnographer provide a genuine description? In fact, the author must translate meaning from one culture to another, and this is the case with ethnographic studies for information projects too.

The spectrum of ethnographical description as shown in the illustration is the result of different approaches how to describe an unfamiliar cultural scene. The ethnographic description is vague and offers easy understandable stereotypes, but doesn’t touch the heart and soul of an ethnographic description. In fact, it resembles a description of a focus group or a market segment.

Social science descriptions also use stereotypes to describe groups and do not reflect any cultural knowledge, whereas standard ethnographies employ sometimes-native languages to describe a cultural fact. The base of the description is still the author’s own cultural background and concepts.

Monolingual ethnographies and life histories are almost descriptions of a culture in its own terms, giving a way to the cultural scene members to describe ‘them’. The ethnographer will then translate the original description into a language he/she understands.

At the far end of the spectrum, ethnographic novels mark the most intense form of incorporating alien culture norms, objects and language into an ethnographic description. They are composed by natives and very rich in descriptions. And: they use the language spoken in the community.

Illustration: Types of ethnographic descriptions (Spradley, 1979, 22)

It seems obvious that there are many possibilities in what form a team of designers will receive an ethnographic description. My argument here is that a) designers need a compact translation of a culture due to time and resource restrictions instead of a lengthy dense description, and b) the translation needs to take the small meanings and details of a cultural scene into account to be effective.

This demands are quite ambivalent first, but important. Designers may learn about ethnographic methods, but they are designers first. It is good to know about the restrictions and the advantages of a method, and therefore useful for their cooperation with ethnographer. But as the ethnographers don’t design information objects, and it’s a benefit when they know about the possibilities there, they focus on their speciality. So should designers. Ethnographers should provide a description that is useful to and used by the designers when constructing their information project. But what designers need from ethnographers, is a new sort of ethnographic description.

Links:

One Response to “Ethnographic Persona: Section 3”

  1. [...] Next section (section 3) [...]

Leave a Reply

You can use these XHTML tags: <a href="" title=""> <abbr title=""> <acronym title=""> <blockquote cite=""> <code> <em> <strong>