Participant Observation

Tacit knowledge-deep hanging out

• What do people do now?
• What values and goals do people have?
• How are these particular activities embedded in a larger ecology?
• What are the similarities and differences across people?
  pay attention to all the artifacts
  look for workarounds and hacks
  "Errors" are a goldmine

What we’re going to talk about in this lecture is observing people to discover their needs, goals, and values.

One effective starting point for designing new technology is to clearly identify an existing problem or need. And that’s because finding a big problem and need often yield[s] important untapped opportunities for design.

Observing people can also help you build empathy and think from their point of view, to stand in someone elses shoes — or maybe to wear someone else’s gloves.

Mike Kuniavsky is a colleague of mine and a design consultant and author. When we were talking about the role of fieldwork and design, he gave me the example of an electronics firm that makes devices for truckers. Apparently, the systems were underused and sometimes, when responses were provided, they were pretty minimal. These mobile devices had small physical keyboards. When the designers went into the field and spent time with the truckers that used them, they found that many truckers had big hands and wore bulky gloves, making it near impossible to use the tiny keys. The resulting redesign featured a large touch screen. This interface provided common responses with one click, as opposed to lots of typing. And the dynamic display made it possible to have big buttons and a stylus was introduced for precision input when that was necessary. From this example you can see the wisdom of something attributed to Yogi Berra, that “you can observe a lot just by watching.” What we’re going to talk about today is participant observation techniques, for standing in someone else’s shoes. Of course, observing people isn’t the only way to begin designing an effective user interface. Great designs emerge through all sorts of approaches. That said, it's often a good strategy to begin your design process by attuning yourself to your users. The techniques in this lecture are inspired by the fieldwork strategies that anthropologists use to learn about and document culture. In 1914, Bronis?aw Malinowski travelled to Papua New Guinea, where he conducted fieldwork at Mailu in the Trobriand Islands. While he is there, World War I breaks out. He has two options: either hang out in the Trobriand Islands, or face internment. I think you can guess which option he picked. During this period, he developed the practices of participant observation which remain a hallmark of ethnographic research to this day. In this photograph, Malinowski is being taught to play a stringed instrument. The picture of Malinowski is a wonderful illustration of what Genevieve Bell calls “Deep Hanging Out”: By spending time with people doing their work and living their lives, you can get beyond the surface things that people say, to learn about what they actually do. If you’ve had lived in another country, I bet you’ve found out all sorts of things that are normal to the people that lived there and completely unfamiliar to you. But you don’t to go that far, or even go anywhere, to begin to recognize the beautiful complexity of culture. It’s all around us. For example, being a student requires an enormous amount of practical knowledge and constitutes a large number of practices that you enact every day and are rarely conscious of precisely because it’s such an everyday behaviour. But if someone were to be suddenly dropped into being a student, with no knowledge of what student life was like, they’d have all sorts of trouble. Furthermore, much like the intuitions that we’d build up when we learn to play musical instruments, there’s all sorts of stuff that we do as part of our everyday behaviour that’s really tough to articulate. So, what we’re going to hopefully learn by participant observation in this class is five key things: First, what do people do now? What’s the baseline that we’re starting from? Second, what values and goals do people have? Most often we’ll want to build technologies that align with what people care about and what they hope to accomplish. And by that I don’t mean literally building what people asked for because people often don’t know, especially for disruptive technologies. Rather, what I’m talking about is designing technologies that will weave themselves into the fabric of people’s everyday lives. Even if they introduce new concepts and functionality. Third, we’re going to look at how these particular activities are embedded in a larger ecology of behaviours. For example, we might be tasked with designing a better technology for a bus or subway. Of course, for any individual bus or subway user, the bus or subway segment is just one piece of a larger activity — like getting to a friend’s house, or commuting to work, or going to the grocery store — and, by understanding the constraints and goals of that larger activity, you may come up with ideas that you wouldn’t have if you were just thinking about the bus ride more narrowly, like, “What leads somebody to select or not select the bus?” Even if your original design brief was about improving the literal bus, what you might end up with, as a designer, is something more broad, like creating a mobile application that helps people figure out when a bus is coming or nearby, or the best way to get from one destination to another. Taking this broader view can help you be more effective as a designer, by helping [you] design from the larger activity that people are engaged in. And designing for that activity can often take you away from the narrower brief that you originally began with. Fourth, what are similarities and differences that you can find across people? In our bus example, a low mobility user might care enormously about the accessibility of the bus. Somebody else may be primarily concerned with costs. And somebody else still may be primarily concerned with efficiency in getting there. One of my favourite examples of participant observation comes from Jack Whalen and colleagues at Xerox PARC. They were studying a call centre for photocopier repair, so these were people who field questions for technicians, and, over the telephone, help them work through troubleshooting broken photocopiers. Doing this over the phone can be extremely difficult. What Whalen and colleagues found was that, as you might expect, the most proficient person in this copier repair centre was the person who’d been the longest. It was the skill that they’d built up over a period of time. What’s interesting is that the second most effective person at this repair centre was not the person who’d been the second longest, but rather the person who’d sat next to the person who had been the longest.

What they realized was that, by sitting next to an expert, these repair technicians were able to pick up all of the informal skills of doing repair work that aren’t written down in manuals anywhere. And it’s this apprenticeship model that helped somebody really excel in their job. One effective strategy for learning about the work practices of your users is to apprentice yourself. For example, my former PhD student Ron Yeh did his dissertation on software tools for field biologists, and as part his dissertation research, Ron in essence became a deputy field biologist. He trained as a docent up at the Jasper Ridge Preserve near Stanford, and he accompanied a field biology group with them on their field site trips to places like the Mexican rainforest. So you can see another advantage of doing technology design is you get to go to a lot of fun places too, sometimes.

Even, say, you’re interested in making a better supermarket checkout system, you might apprentice as a clerk, and one of the things you’d learn from that is the different workarounds that clerks do as part of their daily activities. Being a good apprentice is an interactive process. To get going, you’ll want to set up a partnership with the people that you’re working with. And, of course, you’ll probably get some formal training — that’s the process part of the job. Often, that formal training includes a bunch of asides, if you’re getting trained by somebody who currently does that job. And those asides are the extemporising that happens as part of the job. Definitely, once you actually get involved, you’ll realize all of the everyday hacks and workarounds. If you see something that catches your eye, it can be important to validate that with those that you’re working with, and so you can reflect things that you back to people, and that can help you understand why those things are being done.

Pay attention to all of the artefacts(人工制品) that compose part of people’s work, and in particular look at ways that people have hacked their equipment to make their work more effective. One of my favourite examples of this is Post-It notes. Think about a fax machine or some other piece of equipment that has Post-It notes all over it, reminding people how to use stuff. That’s a great example of an opportunity for innovation. Or, for example, in my recording setup here, I have a note that I have made myself, to turn off the DropBox files syncing software when we record. DropBox is a great way to keep all of the machines that we use up-to-date, but it sometimes seems to interfere with the video recording, so I’ve left myself a note. You could see that as an opportunity for redesign, to, for example, turn off file syncing automatically while video recording is under way.

In particular, errors can be a real goldmine. And what’s interesting is that often I think that when somebody can’t figure out how to use technology, it can be easy to assume that it’s just because they’re stupid or they don’t know what they’re doing, and that attitude is, I think, less pervasive than it used to be, but still quite common, that the development team often looks down the nose of the users. And one of the most effective ways I have ever seen to combat seen is in the 1980’s: an anthropologist named Lucy Suchman worked with Xerox PARC to study photocopier usage. And one of the things that she did was she produced a video of two people trying to produce a double-sided copy of 50 pages of paper. You’ll see this video here.

(There’s a bunch of buttons here. So, “Output paper tray”, “Copy mode”, “1-sided”. We want “2-sided copies” “Unload top-paper tray” [fast-forward gibberish] That’s only the half of it, right? I only got the… [inaudible] Ok, 49 must be on the other side. Hahahaha!) This video has become a legend here in the human-computer interaction field. We’ve just a short clip out of what’s a much longer video, and, legend has it that when this video was shown, the executives said, well, you know, “Those users are just dumb. They don’t know what they’re talking about. They don’t know what they’re doing.” And, at that moment, Lucy Suchman and John Seely Brown who collaborated with her on this work, gave their review, which is that those two users were Allen Newell and Ron Kaplan, two of Xerox’s premier research scientists. They were giants of computer science research, and, in this way, you couldn’t say it was because they didn’t know how to use technology.

All of a sudden, it made you realize that if you are unfamiliar with a particular piece of technology, that stuff could be a lot more difficult, and that led Xerox to help focus more strongly on the user interface.

What we’ve talked throughout this lecture is the difference between what people say and what people do, and I’d like to close by giving you an example from Walmart. A couple of years ago, in a survey, Walmart asked its customers whether they would like the aisles(走道) to be less cluttered（雜亂的）. I bet you can guess what the response is: “Yes!” I think anybody, if asked that way, would say “I’d love the aisles to be less cluttered!” So, very diligently（勤勉地）, in response to these survey results that Walmart got, they decluttered their aisles. As this news article reports, Walmart spent hundreds of dollars decluttering its aisles, removing inventory, cleaning things up. And what happened is that they lost a billion dollars in sales. This seems surprising — they did exactly what people asked. Right? But Walmart made two mistakes that, after taking this class you’ll be able to avoid.

The first is that they paid attention to what people said, rather than what they did. The second is that their survey asked a leading question, “Would you like the aisles less cluttered?” Put together, this gave them a direction that was exactly the opposite of what turned out to be the most effective. I hope these techniques that we’ve covered today in these examples will help you as you do your needfinding in your project assignment. And, if you’re curious and would like to learn more, here’s a few resources that you can use.

Interviewing

Choosing Participants

• Representative of target users
• May be current users of a similar system
• Might be non-users

Say you were designing...

• A lecture support system
• Who would you interview?

Recruiting Participants

• Get a diverse set of stakeholders
• Use incentives(刺激丑蛤，優(yōu)惠) and motivation（動機(jī)）
• Approximate better than nothing(may not be ideaL, but better than nothing)

The importance of being curious

The trick to finding ideas is to convince yourself that everyone and everything has a story to tell. -Malcolm Gradwell

What are good questions

“Is the daily update an important feature to you?”

If you wanted to learn about the daily update, participant observation might be a lot more effective because you could see whether people actually use the daily update or you might even use log files as a way of driving your questions.

“I see from the log that you’ve never used the daily update. Why is that? Tell me more.”

**the more open-ended your questions are, the more interesting the answers that you’ll get. **

Other Types of Questions to Aviod

• What they would do/like/what in hypothetical(假設(shè)的) scenarios- too hard to know.

• Asking us how often we do things — We often lie to ourselves.

• how are things similar.
  This works better if you can make things concrete. “How much do I exercise in a typical week?” — Not very reliable. “How much did I exercise this week?” Well, I’m likely to give more truthful answer to that one. It’s more concrete, and more recent in my memory.

• How much they like things
  avoiding asking people how much they like things on an absolute scale: What does “seven” mean? ask people to compare the product to something else.

• binary questions, things that have a yes or no answer.
  “Do you like grapefruit?” – “Yes.” Not a very interesting interview.

Good Questions

So what are good questions? Especially at the beginning of an interview, I recommend open-ended questions.

After asking the question, give somebody a chance to respond. A little bit of silence is golden.

Additional Needfinding Strategies

Longitudinal(縱向的)or Sporadic（零星的） Behavior

Diary Studies

• give people a diary that they complete at a specific time or interval
• Structured task
• Can use journals, cameras, voice, video
• Tailor(裁剪) the recording to the context
• Can scale better than direct observation
• Easier tools -> better results
• May require some practice, training, reminding

Experience Sampling

lead users
extreme users

Keeping Users in Mind

Personas

• A model of a person, an example.
  • include demographic information, but should also capture a person's motivation, beliefs, intentions, behavior and goals
• Draw a picture of your persona or use a photo
  • give him or her a name, an occupation, a background, social situation, some hopes, dreams, and goals etc. Give the persona a story to tell
• Knowing what our persona thinks, does, and feel help build empathy(移情)
  • so that you can understand the state of mind, emotion, philosophy, beliefs, or point of view of the user
• Empathy leads to insights which leads to design opportunities

Ultimately, it's the design