Anthropological methods and theory can be used to study the changing digital landscape in a variety of ways. We can study how technology is used, or how it could be used, how it is perceived, or how it could be designed. Whatever the scope and purpose of the study, anthropologists have a set of methods for crafting good ways to understand people’s behavior and their attitudes.
I should add that design research – when we study how tech could be designed – is a bit different, however. This type of research is not purely scientific, so to speak. Anthropologist doing design research analyze the raw data they collect in order to inform and inspire the design process. They try to reconfigure what they see and record in completely new ways, creating new worlds of design possibilities. For those who aren’t familiar with this specific field: user research is done before or during the design phase, and user testing takes place after the design phase.
The list of tools and approaches used in both “classic anthropology” and user research is ever growing, but here’s a very quick overview of the main concepts and methods.
Key concepts & methods
How to select people in our studies? Participants in both user and social research studies can be selected by probability sampling (random or stratified) or non-probability sampling.
Probability sampling is commonly used when we investigate so-called attributes of individuals in a population. If we want to understand processes rather than attributes we’d use non-probability sampling. Examples of non-probability sampling methods used in anthropology are quota sampling, snowball sampling (i.e. in studies of communities and networks), judgment sampling (i.e. used in life history research) and convenience sampling (especially in exploratory research). In user research non-probability sampling is often done with selected groups of volunteers or convenience sampling.
Triangulation allows the social scientist and the design researcher to explore a research issue from at least two different perspectives. This can be achieved by using different research methods, different sources of data or type of data to address the same research question. Triangulation is particularly important in phase of validation of research results, and it offers a more complete overview of the research issue.
Interviews can be done in person, by email, Skype or other VoIP solutions. In structured interviews the interviewer poses questions similar to those in a questionnaire; semi-structured interviews feature both open and closed questions; informal or open interviews cover several topics and follow a sequence of open questions; group interviews or focus groups are conducted with several participants at the same time.
All interviews, no matter the format, should follow a script – more or less specific and structured. All researchers performing interviews should be trained in order to reduce bias, challenge expectations, follow ethical guidelines and make the respondent/s feel comfortable. An anthropologist is also normally trained to understand cultural differences adjusting the interview style according to the respondent.
Probing is a way to stimulate the respondent to produce more information. The silent probe is often an effective way to achieve this, but there are other good probing techniques.
Design games use gaming dynamics to generate discussions on current challenges and to collaboratively envision future solutions.They are often used in participatory design (PD) sessions in order to involve users in the concept development and design process, and they take place either in the users’ own social environment or in a ‘lab’ setting. This article provides a very useful overview of the main tools and techniques of participatory design. Read more about design games here.
This technique was established at the end of the nineties with The Presence Project, where groups of elderly people in three European countries were studied using a set of probes in a package created by the researchers. The package included pre-addressed postcards, seven maps (i.e.. a local map, a world map etc), a disposable camera, a photo album and a media diary. Participants were asked to answer questions related to the items in the package and were instructed to take specific photos, compile a diary and tell their life story through the photo album.
The aim of the project was to find new ways to understand how technology can be used to increased the presence of older people in their local communities. More resources about cultural probes here.
Questionnaires are a classic way to gather data, and they can feature both closed and open questions. Survey research is an established way to gain insight about “target populations” (it has been used for over 200 years), and it is effective as long as the questionnaire is well designed and worded in a clear way. Questionnaires can be self-administered (i.e. internet surveys), can take the form of face-to-face interviews or can be used in telephone interviews.
In survey research complex variables (i.e. the amount of stress experienced while performing a task) are usually measured by scales. The most commonly used scales in questionnaires are the semantic differential scale, developed in the 50s at the University of Illinois and very easy to use, and the Likert scale – probably the most used (see examples of both in the image below). Other interesting examples are the Cantril Ladder of Life and the Faces Scale.
Researcher using computer based surveys have several new digital tools available to give respondents a better user experience. My favorite is Typeform. Another options is Qualtrics. A good visually-based survey tool specifically designed for user experience and usability testing is Usabilla.
DIRECT OBSERVATION & ETHNOGRAPHIC FIELDWORK
Observing and recording human behavior is essential to complement the raw data collected with interviews and questionnaires.
Anthropologists use ethnographic fieldwork to learn more about what people do. During fieldwork the researcher can be a complete participant, a participant observer or a complete observer, but most ethnographic research is based on participant observation. Participant observation is a great tool to reduce reactivity (people changing their behavior when they know they are observed and studied), especially when carried out for long stretches of time, like in a classic anthropological fieldwork of several months. As Bernard puts it: “as you become less and less of a curiosity, people take less and less interest in your comings and going” (Russel Bernard, Research Methods in Anthropology, an excellent manual covering all the anthropological methods in detail).
Participant observation also extends the validity of data collected with interviews and questionnaires.
While classic anthropological fieldwork normally takes months, applied ethnographic research is usually carried out over few weeks or even few days, using rapid assessment procedures (RAPs). In this case there is generally not that much time to build rapport.
Design researchers can use a variety of useful frameworks for observation in the field in order to optimize their scarce time and keep the study in focus. One of them is suggested by Robson in Real World Research. A Resource for Social Scientists and Practitioner-Researchers. Robson suggests to pay attention to several dimensions while doing direct observation in the field – namely the activities, goals, actors, space, objects, acts (specific actions), events (i.e. special events), time and feelings. Video and audio recording are often used in direct observation to capture as much data as possible.
Direct observation in controlled settings is often used in usability testing – both in labs and in the participants’ normal environments with portable laboratories. This type of observation is also accompanied by audio and video recordings. To learn more about the pro and cons of observation in labs and with portable laboratories I recommend this article by usability.gov.
The think-aloud-protocol is a tool used in order to understand the users’ thought process during usability testing: participants are encouraged to think out loud and explain what they are doing while they interact with a specific product or service.
The behavior of people can also be monitored using diary or camera studies, web analytics and logs.
This method is part of the contextual design approach toolkit. Contextual inquiry was inspired by anthropology, psychology and design and seeks to understand the wider context in which the technological solution that is designed will be applied. Fieldwork is essential to contextual inquiry. The researcher enters a master-apprentice relation – the user is the expert, while the fieldworker learns as much as she can in order to become a master.
BEHAVIOR TRACE STUDIES
In anthropology behavior trace studies are a sort of “behavioral archaeology”, as Russell Bernard pointed out. In these studies researchers collect evidence about human behavior (the ‘traces’ people leave behind, like garbage or graffiti) without telling the subjects that they are monitored. Could this method be effectively used in service design, perhaps in an attempt to understand practices related to e-waste?