Once you've conducted the interview and observation, you'll now have a whole bunch of notes and, hopefully, some audio recordings. Now, in some sense, you can think of the audio recordings and your notes as your raw data. But from that raw data, we'll need to make some sense of it for analysis, and that's a process that I call data extraction. Basically, the idea is to take your raw notes and the audio and then turn it into little bits of bite-sized information that you can use for analysis. This process is something that people call generating affinity notes, and that's partly because the affinity notes will eventually become part of an affinity wall. Basically, what you want to do is as soon as possible after the interview, hopefully within a couple of days, you'll want to review all of your notes in detail and actually listen through the audio recording so that you can write down these affinity notes. An affinity note is basically just a sticky note, something that looks like this, on which you write down a code for the user that you were interviewing as well as a statement or a question. And these statements should come from your notes or from the audio recording in a way that they're basically extracting different bits of information that are relevant for your overarching question. You want to include on a sticky note, again, just one idea at a time. Either factual statements, these should be complete sentences, they could be participant quotations where one of the interview participants said something interesting or relevant. They could be observations that you've made, again, as full sentences, or they could be interpretations that you've thought about of any of the above. Another thing to think about sometimes is to add questions. Where you may have a question that emerged from your interview that you haven't yet answered, that can also go on a sticky note. The tips with respect to writing these affinity notes are that you should make sure that every note is understandable on its own. It should be a stand-alone sentence or a question. You want to be as concrete as possible but also concise so that it's not a whole lot of text on any given sticky note, and then you want to have about one affinity note per minute of interview. So for a one-hour interview, you'd have about 60 affinity notes. Now in theory, that's the case, but in practice, you'll often find that some cases, some interviews are not as revelatory as others. You might only have, say, 40 sticky notes or you might have as many as 80. But I would say, anything less than 30, for example, for one hour of interview, is too little. And anything over 100 suggests that you might be making them too granular and more detailed than you need. I mentioned earlier that you'll want to add interpretations to your affinity notes. And these are things that you might have realized as you were taking notes, as you were conducting the interview, or even as you were generating other sticky notes. So if you notice that there are repeating themes, you might want to write that down and say, here's a repeating theme. People seem to get very frustrated when they enter the final stage of a certain process on a website or contradictions or conflicts that you've noticed. You can also note key findings that you may have found that may not be visible in any one of the given affinity notes that are from your interview participants or from your observations but which summarize something overall. You'll also want to note things like root issues that you might have noticed as you were observing and anything else that might ultimately be relevant but which isn't necessarily in one of your audio recordings or in the notes that you've taken. The other thing to think about is what kind of quotes to include from your interview participants. Good quotes are those which are unexpected because they say something that you really didn't expect, or which in some sense represent many other users' ideas of a particular situation. So going back to the example of a checkout process at Amazon, you might have noticed that everybody seems to get stuck on the promo code section. If that happens, then a user who's made a quote about that particular issue might be something that you want to include. Now, this module is probably the most intensive in terms of the assignments because you'll have to conduct interviews as well as extract these individual affinity notes. The interviews and observations, you want to remember to take at least two sets of interviews, each which will take about an hour. And if you take notes and do the audio recordings, you should find that you can easily generate about 100 affinity notes from those two interviews. What I'd like you to do for the assignment is to, first of all, take 20 of the notes that you think are most interesting or most representative of your interviews. Where you choose out of those 20, 5 should be factual statements, 5 should be observations or interpretations that you've made, 5 should be quotations from one of the users, and the rest of the 5 can be anything that you like. And these will be the ones that will be submitted for peer-review because, otherwise, it's impractical to assume that somebody else is going to look over all the 100 or so of the notes that you've created. More of the details of the assignment will be in the module handout, and I suggest you take a look at that. And let me again emphasize that this module probably has the most intensive workload within this course. Again, you're going to be conducting interviews that might take between two or three hours total. And then the generation of the affinity notes can be quite time consuming. And so they'll often take about double the time that you took to do the interview itself, so possible amount of work for this module itself will be about eight hours. In the next section, we'll talk about how to analyze this data that you've created as affinity notes.
