The term participant observation may be confusing to those of us in user experience. We think of participants as the people who we study, and we think of observation as the way we study them. So to us, participant observation sounds like what we do already—observing participants. But in this case, participant means that the researcher is an active participant in an activity while observing it.
Does Participant Observation Make Sense for Design Research?
Participant observation studies have provided some of the most valuable insights in anthropology and sociology. So why haven’t we adopted participant observation in design research? If we look at its advantages and disadvantages, perhaps we can determine whether we can adapt this method to the needs of design research projects.
What Are Its Advantages?
In comparison to pure observation studies that don’t involve any participation by the researcher, participant observation provides the following advantages:
- Participating in a group as a new member requires the researcher to learn about the group and its activities in much greater detail than when simply observing and taking notes.
- Performing activities with the group gives the researcher greater empathy, as well as a much more in-depth understanding of the group members and their activities.
- As a new group member, the researcher often notices things that group members take for granted, such as group rules and norms.
- Group members often feel more comfortable and act more naturally when a researcher participates in their activities rather than just sitting back and observing silently.
- As a group member, the researcher spends more time with the participants and gets to observe them in more varied situations.
- Participating eliminates the formality of scheduled research sessions, in which participants expect to answer questions or demonstrate specific tasks. The researcher is simply with the group all day, observing and participating in whatever happens.
What Are Its Disadvantages?
The following disadvantages of participant observation have probably prevented its wider adoption in design research:
- It can be very time consuming.
- It generates a vast amount of data.
- By participating in activities, the researcher can inadvertently influence the other participants’ behavior.
- Active involvement in the group can cause the researcher to lose objectivity and may lead to bias.
- It can take a long time for the group to accept the researcher as a member and become comfortable with him or her.
- It’s difficult to participate, observe, and take notes at the same time.
- The researcher has to be careful not to cross the you-are-not-the-user line and start designing for his or her own wants and needs.
- The need for specialized domain knowledge makes it difficult to participate in some groups. For example, it might be easy to participate in a group of fast-food workers, but it wouldn’t be easy for most researchers to join a group of molecular biologists.
For What Types of Projects Is Participant Observation Appropriate?
Participant observation is useful whenever the goal is to study a user group and how they use a product, system, or service whose use consumes a large portion of their time. Therefore, it’s ideal for service design, process redesign, and business application design projects. It’s especially useful in learning about groups of employees, their activities, the systems they use, and the services they perform. For example, participant observation would be a great method for learning about
- customer-service personnel working in a call center
- nurses in an urgent-care center
- fast-food workers and the customer experience of a fast-food restaurant
- real-estate agents
Because participant observation is group focused, it wouldn’t be the right research method to use for activities that people primarily do alone or that they do infrequently—for example, consumers making purchases on an ecommerce site. Since people shop online alone and shopping takes up only a small percentage of their time, it would make more sense to do a series of contextual inquiries in this case. However, participant observation would make sense in studying a corporate procurement group making purchases for their company on B2B commerce sites.
How to Conduct Participant Observation
By this point, you may agree with me that participant observation sounds like a good way to learn about a group of people and their tasks, but you may be wondering how you could realistically apply this method within the tight timelines and budgets of typical UX design projects. So let’s look at how to do participant observation within the context of UX design research.
Do Some Initial Research
First, to help you plan your study, do some initial research to learn about the domain, the users, and their activities. Look at any previous research that others have done on the topic. This will help you to determine where to focus your research. Later on, your being well informed will help you to make sense of what you observe.
Plan Your Study
As with any study, you should first determine what your research goals are and the questions that you want to answer. However, doing participant observation requires that you address a few additional considerations.
Plan What to Observe
Unlike an anthropologist studying a tribe in New Guinea, you typically won’t have much time to conduct your design research, so limit your focus. Determine the groups, the people, and the activities that you want to observe. Since you can’t observe everything, choose a representative set of activities to observe and determine when would be the best times to be present to witness those activities.
Determine How Much Time to Spend
Consider how much time you can spend on participant observation within the practical constraints of your project schedule. The time that you would need depends on how many different groups you must observe, the number of people in each group, the number and types of activities you need to watch, when those activities happen, and the complexity of the domain.
Decide on the Number of Observers
You can conduct participant observation either alone or with two or more other observers. If each observer participates in a different group, your data will have greater reliability. Observing only one group doesn’t reveal whether characteristics, behaviors, actions, and situations are unique to that group or common across groups. One observer is also more susceptible to bias and the subjective interpretation of data. Combining data from multiple observers helps you to overcome such difficulties. Plus, it gives you other people with whom you can discuss the findings.
Determine Whether Your Participation Will Be Overt or Covert
You can conduct participant observations either overtly, informing the group members about your study and getting their consent to participate, or covertly, joining the group without letting them know that you’re a researcher conducting a study. Each approach has its own advantages and disadvantages, but for design research, overt participant observation is more appropriate.
On the other hand, the primary advantage of being covert is that you can see the most natural behavior in participants. Since you’re pretending to be a regular group member, your presence won’t make the other group members uncomfortable. And you’ll avoid the Hawthorne effect—the tendency of people to act differently when they know that they’re being observed as part of a study.
However, covert participant observation has more disadvantages than advantages. Being covert requires deception and deprives the participants of the opportunity to provide their informed consent. Maintaining your cover is a burden on you as a researcher, and it makes it almost impossible for you to take notes. Also, in some situations, it’s impossible for a researcher to pretend to be a group member because he or she is so obviously different from the group—either in physical characteristics such as age, gender, or ethnicity or in knowledge or experience.
For design research, it’s much better to conduct overt participatory observation. Being honest makes it easier for you to join and become an accepted member of the group, openly take notes, and ask questions. While you may not fit the profile of the typical group member, that won’t matter if they know you’ve joined the group because you’re doing a study. The problem of participants’ being uncomfortable or acting differently because they know they’re part of a study fades over time, as the group becomes more comfortable with you. In fact, people often feel more comfortable with a participant researcher than with a researcher who just sits back, observes, and takes notes.