I wrote a short memo to the Ushahidi team about what exactly an ethnographer does and how ethnography as a discipline could be useful to Ushahidi (and Crowdmap in particular). I’m thinking of actually writing more about this and interviewing ethnographers working at technology companies to shed some light on this growing field.
What is ethnography?
Ethnography is a research method, with roots in anthropology, that aims to gain a rich perspective of user communities. Ethnographic research projects require the researcher to be deeply immersed in a specific research context (also called “participant observation”) and to develop an understanding that would not be achievable with other, more limited research approaches (Lazar, Feng, Hochheiser: 2010). Ethnography emerged from the practice of early anthropologists who studied “new” cultures (ethnographers were traditionally from western Europe and were studying civilizations very different from their own). As such, the ethnographer is required to step outside of the role of a dispassionate observer and directly engage with the participants in a community.
Three key principles of ethnographic studies include:
Deep engagement: unlike other research methods where the researcher is distant from research subjects (such as in quantitative studies using surveys), ethnography requires deep engagement with members of a community, using techniques like participant observation and interviews.
A science of the contextual: ethnography requires the researcher to understand the different contextual forces at play within a particular community. It is based on the notion that individuals may describe what they do in ways that are inaccurate and that in order to get a true understanding of what is happening in a community, the researcher must encounter it firsthand.
Ethnography is inductive: rather than starting with a hypothesis and finding data to back that hypothesis up (as in the case of most quantitative studies), ethnography is about using real examples (and other types of data) to build a rich perspective on a community
What ethnography is not
Ethnographers will not start with a broad survey or questionnaire but will instead start with a microcosmic view and try to get a more detailed understanding of a small part rather than a broader understanding of multiple communities. Instead of doing a survey of all Crowdmap users, the ethnographer will rather start by interviewing users from a particular deployment and trying to understand the context (motivations, socio-economic conditions etc) of a particular Crowdmap story.
Ethnography is not the same as user experience research. Whereas UX research might involve researchers watching a user perform a specific task, ethnography would involve the researcher in participant observation of the daily life of the user in their community (not just the times that they are using technology).
Ethnography is not about facts, it is rather about stories. Rather than talking about the number of map users and the effectiveness of mapping in disasters, ethnographers will try to understand what the map (as an artifact or concept) means to members of a particular community and tell the story of how the community engages with maps.
Ethnographers hate to make generalisations. Because the research is focused in on very specific contexts of use, it means that it becomes difficult (in theory, at least) to make broad generalisations that the conclusions of one study can be applied globally. To me this is one of the values of ethnography: it enables us to gain more insight into the conditions necessary to make software more successful, rather than making claims that “because it works in one place, it can work anywhere” when we know there are varying levels of success in different places.
How could ethnography be useful to Ushahidi?
Ethnography is human centered and works well with human centered design projects for understanding the needs and contexts of users. Ethnography is useful in understanding how new products should be designed. Because of the focus on what users actually do/need, rather than what they say they do/need, ethnography can be useful for understanding how technology can aid existing workflow/practices. For example, instead of building a system for providing complex algorithms to assist Ushahidi users in verifying information, ethnography would enable us to understand how Ushahidi users currently verify information, and provide ways to enhance existing practices.
Ethnography is about enhancing the ways that communities are already working with one another, rather than trying to supplant communal interaction. This is especially useful for Ushahidi when one considers the tension between technology providers and humanitarian organisations. Ethnographic studies would focus on all the stakeholders in a particular context and understand the tensions and dynamics between them.
Ethnography can help us understand why some deployments fail/succeed. What are the specific conditions of a particular context that led to failure/success? Ethnographic studies can reveal the features of a particular context that were detrimental or that enabled it to better succeed and to document that story so that others might learn from it.
What would a Crowdmap study entail?
(working on this section with Brian)
A Crowdmap study would involve looking at two or three deployments, participating in online community discussions, interviewing participants, and doing some content analysis of reports.
Some interesting questions a study might seek to understand:
Who are the Crowdmappers here? Is there a big gap between the work of administrators and users? Who does what?
How do the Crowdmappers perceive a map? What is their experience with maps? What do maps represent for them? Did they learn maps in school? What were they required to know about maps? How do they perceive “check-ins”? Are there any local analogies?
How do Crowdmappers tell others about their project? What challenges do they face that are particular to the context in which they are operating? What is the “reach” (social/cultural/political/gender) of the project? What are the factors defining this reach?
How do Crowdmappers talk to one another? How do they define their community?
How do Crowdmappers understand and deal with misinformation, inaccuracies, spam and verification problems? How much do they trust the information that is being produced? Does “truth” matter to them? Are users “learning” how to write more structured/standardized reports over time? How are they learning?
Why do Crowdmappers do what they do? How do they understand the effectiveness of what they’re doing? What do they want to happen? How do they understand and experience agency?