Ethnography
Ethnography [noun] – A branch of anthropology dealing with the scientific description of individual cultures. (Anon, 1995)
As a UX Designer, my mantra is to always consider the users’ wants and needs before my own to ensure the end design is as easy as possible to use and navigate. This involves studying different types of users to determine the best way to cater to everyone’s needs. For this reason, ethnographic research can produce interesting results in a user testing environment.
Ethnographic research is research conducted in the participants’ natural environment or an environment of which relates to the product. This type of research is not conducted in a conventional UX testing environment (e.g. in a laboratory) unless of course, this is the environment in question, but rather in a setting where the user can be naturally affected by surrounding factors like sounds and sensations. As well as collecting data from the structured interview, the interviewer also gains a sense of how they typically perform with natural stimuli around them.
Contextual Inquiry
This user research method feeds into Beyer and Holtzblatt’s contextual inquiry model, using a ‘master-apprentice model’ (Cooper et al., 2014). This outlines the user as the master of their craft and the interviewer as the apprentice looking to gain more knowledge. They outline four fundamentals for conducting ethnographic interviews:
- Context – The user interview setting is to be their natural work environment or an environment which directly relates to the subject of the product.
- Partnership – Research conducted with the user should be collaborative to explore their thought and opinions at the same time as observing their actions.
- Interpretation – Without relying on an assumption, the interviewer must extract implied information from the user to feed into the design implications.
- Focus – The interviewer must steer the interview appropriately to not stick to a regimented scheme, not randomly drift through questions, but to assess events as they occur to ensure optimum, comfortable compliance from the user.
Preparing For and Conducting Interviews
There are four main parts to ethnographic user interviews. These are:
- Creating a persona hypothesis
- Planning the interview
- Conducting the interview
- Analysing the results
These tasks, if executed well, make for a great research session and yield results otherwise impossible to uncover if tested under different conditions. I will go through each part to further explain it’s relevance in the user-centered design process.
Create a Persona Hypothesis
To begin with, the designer must come up with a hypothesis of a potential user to base their first set of interview questions on. This persona is to be based on assumed, likely behaviour patterns rather than demographics. If the designer begins assumptions based on demographics, they run the risk of assuming that all users in the same demographic behave similarly. It’s a lot more effective to analyse if sequences of demographics emerge from recorded behavioural patterns.
The designer also needs to consider the difference between business vs. consumer users’ goals along with a difference in domain and technical experience. Firstly, a business user will most likely have a solid job description, therefore making the hypothetical usability of the product a lot easier to map out. On the other hand, consumers will have a broader range of goals and expectations making them trickier to predict. Designers may find it difficult to curate consumer ideologies because they can lack accuracy. In regard to domain vs. technical experience, it is common for a set of users to have a varied knowledge of their particular industry and the technology involved with the product in question.
One important type of behavioural distinction is the difference between technical expertise (knowledge of digital technology) and domain expertise (knowledge of a specialized subject area pertaining to a product). (Cooper et al., 21014)
It is important to remember that this stage of persona creation will not be perfect and is encouraged to be an iterative process.
Planning the Interview
At this point, the designer can now start to consider demographics. This can help to narrow the user pool if demographic research will play a substantial role in the development of the product. About Face 3: The Essentials of Interaction Design states that for an averagely sized domain, in 4-6 initial interviews, covering identified roles, behavioural variables, demographic variables, and environmental variables, the designer should have enough data for a first iteration.
Conducting Interviews
Interviews should roughly take about an hour and contain two designers; one to take brief notes and another to cover more detailed aspects. This UCD research process can be broken down into three main stages:
- Early interviews – These preliminary interviews are non-specific in nature and aim to cover a lot of ground knowledge to provide a wide platform for further interviews.
- Middle interviews – The designer will apply trends from data gathered in the early interviews to form a follow-on session. They will have applied ethnographic research methods such as picking up on subtle clues from the users’ surrounding environment to form a fresh set of more specific, data-driven questions.
- Later interviews – This last set of interviews is used to consolidate all previously collected data using close-ended questions to confirm any assumptions that may be left.
Some key points to consider throughout are:
- None of these tests should follow a uniform structure of type or topic. The designer should vary this as they feel fit to gain the best knowledge of the industry.
- The designer should be picking up on subtle clues all along the way of how the user is reacting to their natural environment’s sounds, sensations, and distractions. Applying this is the golden ticket of ethnographic research. For example, if there are a lot of sticky notes with information on lying around the work environment, could the product benefit from incorporating an organisational tool?
- The designer should be formulating their questions around asking ‘how’ and ‘why’ they will benefit from the product. Only when this data is collected can they ask ‘what’ in regard to product performance.
- Do not let the user become the designer. They are merely there to give you insiders tips on how to improve the service.
- The designer should be encouraging real-life stories and experiences with the product. This technique is excellent for defining users’ pain points. Notice how their tone of voice and body language change when they mention things that please or anger them. This can sometimes lead to a show and tell scenario.
Analysing Results
Post early interview, data can be analysed to see similarities, differences and if any patterns are emerging. Designers can now begin to put thought into a user persona (as opposed to a hypothetical one), create mood-boards for the team to contribute to if any new information comes to light and strategically come up with new approaches for upcoming interviews.
Inductive vs. Deductive
Inductive research, otherwise known as ‘top-down’, is the type of research carried out when a theory is to be proved or disproved. When a researcher has a theory about behaviour patterns of a user, they can then create a hypothesis to test. The relevant experiments are carried out and observations are recorded for analysis. Once all the data has been evaluated, confirmations can be made that will either line up with the original theory or possibly let the researcher devise a new one.
A visual representation of a deductive research method
Deductive research is carried out when a researcher or designer notices some interesting goings-on and can correlate these into notable patterns. From this, they can envision a hypothesis to test. They may go through multiple hypotheses’ before deciding on a legible theory. This theory, if the designer so wishes, can then be researched deductively. These methods can be used interchangeably.
A visual representation of an inductive research method
References
www.dictionary.com. (2018). the definition of ethnography. [online] Available at: https://www.dictionary.com/browse/ethnography [Accessed 12 Nov. 2018].
Experience UX. (2018). What is Ethnography research? | Experience UX. [online] Available at: https://www.experienceux.co.uk/faqs/what-is-ethnography-research/ [Accessed 11 Nov. 2018].
Cooper, A., Reimann, R., Cronin, D. and Noessel, C. (2014). About Face 3: The Essentials of Interaction Design. 4th ed.
Socialresearchmethods.net. (n.d.). Social Research Methods – Knowledge Base – Deduction & Induction. [online] Available at: http://www.socialresearchmethods.net/kb/dedind.htm [Accessed 11 Nov. 2018].