Understanding Deductive Approaches in Qualitative Data Analysis (Full Transcript)

Speaker 1: This is one of two videos that describe different approaches to tackling qualitative data. In this video, we'll focus on the deductive approach. Deductive methods approach qualitative data with a predetermined framework in mind. This means that before you ever really get into analyzing the data, you've already thought about what you're going to look for, so you can focus on what's relevant and ignore what's not. This is in direct contrast to inductive methods, where you avoid predetermined frameworks and focus on any naturally occurring patterns. So, when do you use the deductive approach? First, if you have specific questions from a stakeholder or client that you're looking to answer. In a case like this, the only insights you may care about would be the ones that shed light on those questions, so most everything else could get ignored. A second scenario that makes the deductive approach useful is if you have a lot of data and you just need to get things organized so you can make sense of it all. You might begin by grouping things according to the types of tasks given in a usability test, or questions asked in interview data, or even important research participant characteristics. A third reason to analyze data deductively is if you're seeking additional evidence of previous research findings. Perhaps some exploratory research, such as a survey, revealed that a lot of users are experiencing a couple of main problems. You could analyze your follow-up interviews with these problems in mind, and begin by only really focusing on additional insights related to those initial findings. I think about the deductive approach like a metal detector. When you use a metal detector, you're only really looking for things that are metal, regardless of what other cool non-metal stuff might be out there. Any strategy you use to narrow your focus during analysis is some form of a deductive approach. To conclude, I have two words of caution about the deductive approach. One, don't miss interesting new information. And two, watch out for confirmation bias. If you begin your analysis assuming you'll find a certain thing, it can be all too easy to force the data to validate your hunch. If you really aren't seeing what you expected to see, then that's important to recognize. A deductive approach can be very efficient and help you quickly make sense of a lot of data. You just need to be cautious that you aren't defaulting to this approach because it can be easier.

Speaker 2: Thanks for watching. If you want to see more of our UX videos, take a look at these over here and consider subscribing to our channel. On our website, nngroup.com, you can access our free library of over 2,000 articles. You can also register for one of our UX courses that offer live, hands-on UX training.