Is the data pure?

ew days back (at AC4D), there was a discussion about field studies for design research and synthesis. We talked for a bit aboutĀ contextual inquiry. In a nutshell, this is a process of going to “the research field” and observing the subject and environment first hand to gain insight about the problem that is being investigated. We were talking about an example of doing contextual inquiry for a tooth brush company who want to understand the benefits of using an electric brush and in turn, use that data to design a near ideal tooth brush. While formulating the steps for doing the contextual inquiry, the following steps were outlined/discussed as a guideline –

  1. Introduce yourself
  2. Get familiar with the field (in this case, home of some participant)
  3. Ask to see the place where action takes place (in this case, bathroom)
  4. Observe the action (brushing)
  5. Observe the surroundings to make meaningful inferences
  6. Talk, rinse and understand.
In general, I understand the value of doing field work. I think it is the best way to observe somebody in action and understand the hows and whys of any problem. For instance, if I am designing an information portal for some farmers in India, I need to be there to understand the needs so that I can solve the problemĀ for them. However, the process of doing contextual inquiry is something that I don’t entirely agree upon. Apparently, the process involves shortlisting candidates for participatory research through some screening and then schedule an appointment with them for a field study (contextual inquiry). Then, at that fixed time, you go and observe. I think this procedure works only well in some cases like
  1. It does not make people conscious
  2. It does not induce any disruption in their activities
Doesn’t me observing somebody brush his/her teeth make them conscious? If I am using this as a datapoint to derive information to provide to my client/use in my research, this data point has a level of uncertainty and no longer pure. For this study, I would rather go to an airport or hotel and engage into conversations in the restrooms with people who are brushing their teeth by connecting with them in the angle of a fellow traveler. I think there can be more information from that exercise. In general, I think this “text book” definition of contextual inquiry ticks me off. It cannot be this procedural. For instance, if my trying to understand the life of a potter, I should become potter and spend time with him/her making pottery rather than asking that person to make pottery in front of me. I brought this up in the discussion and the counter statement was, “well, if you just watch people, your understanding of the problem is based on only your view points”. Why? I ask. If I observe enough people, then can’t I find patterns and generalize? Well, isn’t eliminating biases by talking to more people a method in contextual inquiry and participant research?


4 thoughts on “Is the data pure?”

This article makes some good points. Contextual inquiry is not guaranteed to be accurate. The data is only based upon when the observer(s) is present and what the observers deems as important to document as “data”. We have noticed that what the observer deems important is often not the same as what the participant deems important. Multiple observers conducting the same research in different cities may not have the same criteria for data. This leads to inconsistent data and makes patterns difficult to find and unreliable.

In many cases the observer makes the mistake of looking for singular moments of inspiration as opposed to patterns across all respondents. Singular moments of inspiration can be invalid and are always difficult to align project teams to the perceived opportunity.

The real issues here that contextual inquiry is just one method. It is not the super method it is cracked up to be. The approach is about what people do TODAY. There is typically no emotional content. There is no data about what people desire in the future. There is only the misguided belief that “pain points” are the same as what people desire. They are not! The absence of a negative is not the same as the addition of a positive. For getting accurate information about what people do today there are other methods such as self documentation.

The real issue here is getting at what people desire. This is accomplished through participatory or co-creative techniques. It has been going on for two decades and is now becoming more prominent. This approach produces actionable data about what people want, is repeatable so the qualitative becomes quantitative, and generates alignment at the highest levels of organizations such as P&G and Motorola.

The myth that people can’t tell you what want is only somewhat true. People can express what they want given the right tools. So ultimately those who continue to cling to this outdated opinion have been unable to figure out how to enable people to express themselves.

In conclusion I am not saying that contextual inquiry is bad. It is just one of many tools a good design researcher should have in his or her toolbox. You would not use a hammer to cut a board into two pieces!

Hi Marty – Thanks for sharing your thoughts.

In your experience, what other tools are valuable for understanding/collecting meaningful data? Are there any references/anthropological work I can follow to understand more about the problem?


Comments are closed.