Let's suppose that the general manager of a GPS device product comes to you and says, "Can you help us design the new version?" Well, they might say that they have a lot of sales data and they've also done a survey with users, and in some of their products they've actually been able to collect usage statistics because these products are online. Well, they might tell you that this is the data that they have and maybe you could use it for your research. Now, of course this kind of data is useful as a way to understand overall trends about how users of the product use the product. Also it tells you something about the existing product in particular. In particular, all of the things that the designers of the system and the data that's collected could anticipate all of that will have been collected. On the other hand, this kind of quantitative data is not particularly useful as a way of really understanding in-depth how users actually use a product or what the new possibilities are for the product. So, as somebody who does user needs assessment, you would actually say actually I prefer to try a different kind of methodology for devising new features for the product. Some of the questions that you'd be interested in answering are things like; how do users actually input a destination into the GPS device? Do they make multiple detours through different methodologies because they get stuck? Are they really frustrated at any phase? What are their emotional responses to the different user interfaces? Another question might be; are there users who are interested in learning how to navigate a particular route from point A to point B without the GPS device, and would it help to have some kind of quiz that the device actually administers? Then other users might tell you, well, we like having features for the shortest path or the quickest path to a destination. But really when we go on these trips where we need a GPS device, what we want is in some sense the most fun trip. We want to be able to make a stop every hour and let the kids out and have some fun. So, it's some path that actually finds parks along the way would be most useful. Now, these are the kinds of things where you couldn't necessarily anticipate the issues in advance and therefore it's difficult to design a survey, let's say, or to collect data in advance that would tell you whether these are useful features. This is where qualitative research comes in. So, qualitative research as I mentioned previously is characterized by non-numerical data. This is exactly the opposite of quantitative research in which you're dealing primarily with measurement. In qualitative research, you're interested in the rich details that users and other stakeholders of a particular product or service would be interested in. You're also interested in the larger context in which people use the service. You might be interested in the rooms in which they use it, in the car if it's a GPS device, and other contexts like that. What qualitative research is really good for is in extracting the richness and detail of experiences that users have. Some of the strengths of qualitative research are that they give you very in-depth information. You can find out things that are very difficult to measure. They also cover a broad range, so you might be able to hear things that you wouldn't otherwise get if you were just administering a predetermine survey. Qualitative research also generates hypotheses about why people do things and they might also point to causal explanations. Now, qualitative research is actually a very broad term for a large class of methodologies. They sometimes involve interview's, different kinds of observations. You might have heard of focus groups and so forth. Then there's also different kinds of data analysis methodologies that involve coding, interview transcripts, doing various kinds of analysis on the content, and so on. In this particular course, we will be focusing on interviews, semi-structured interviews in particular, and a certain kind of observation where we'll be watching users actually use the product and understanding how they use the product. Finally, we will be doing something with something called affinity walls, where we will take a combination of what's called thematic analysis and recursive abstraction as a way to understand the overall sense of the detailed data that we'll collect. Since I've mentioned some of the strengths of qualitative research, I should also tell you about some of the caveats. So, qualitative research tends to be weak at exactly the things that quantitative research is good at. So, for example, if you're interested in what proportion of users do a particular thing or what percentage of users take certain kinds of actions, that wouldn't be a good kind of question to ask with qualitative research. Similarly, if you were interested in whether a particular feature would increase usage of that feature, again, that's something that qualitative research wouldn't necessarily be good at although you might get hints about it. Then finally, if you're interested in very specific questions like does feature V cause effect W, that's again a question that you'd like to reserve for other kinds of research not necessarily those that are qualitative. Although again qualitative research might provide you hints. Overall, the weaknesses of qualitative research is that they don't give you a very good sense for numbers, for proportions, for relative measurements between different kinds of phenomena. So, what you're really interested in with qualitative research is getting this rich detail and the rich kind of possibilities that a product or service provides. Not in understanding the numerical quantities are associated with it. One thing I will mention is that the qualitative research methodologies that we'll be teaching in this course are actually extremely useful in a wide range of research contexts. Anthropologists use methodologies like this. Sociologists use it as a way to understand how people interact. At schools of information, they use it as a way to understand let's say how people use social media or how people use their smartphones. So, it's a sort of methodologies that are applicable very widely, and if you learn this it's really foundational for all kinds of research.
