How AI Shifts Agency in Qualitative Data Analysis (Full Transcript)

[00:00:00] Speaker 1: Hello everyone. I just want to share with you a dashboard that I've created using Cloud AI. So this is a dashboard that I created. This is how it came about. I'm preparing for a presentation, a one-hour presentation next month with Maskudie and the topic I'm going to work on, why mastering qualitative analysis matters more than ever in the AI era. So this is a topic I was preparing, and in the presentation I want to show the kind of skills that we need to have in this AI era, and I want to use a dashboard to show. So what I did is I went to Cloud AI and described what I want, and within a few minutes the system was able to create this wonderful dashboard for me. And I just want to give you a gist of it, and I also recommend that you come to this presentation. I'm going to put the link in the comment section or description section for you to register. You're going to learn a lot from this presentation. And also the good thing is that in March, the second edition of my book, Step-by-Step Guide to Qualitative Coding, will be out, and you can get access to it. And the interesting thing about this book is that in Chapter 15, I talk about how to use AI to analyze your qualitative data. So that's going to be really helpful for you if you want to learn more about AI, and also get foundational knowledge about qualitative analysis. So back to the dashboard. So let me make it bigger. So and also I'll put a link of this dashboard in the comment section or description section so that you can get access to it and explore and learn about what kind of skill do you need in this era as a qualitative researcher, especially when you are analyzing your qualitative data. During my presentation, the two concepts I will be talking about is researcher agency and cognitive offloading. So in simple terms, when it comes to researcher agency, just think about this as the control that you have in doing your research. So when you're doing qualitative research, there are actions and decisions that you have to make, right? So having control in the action and decision that you're going to make is very important. Nowadays, the control that we have in doing our qualitative research is somewhat threatened by AI, and I think we have to be aware of that. Another concept that I'll be talking about is cognitive offloading. Cognitive offloading is thinking about giving part of your mental tasks, things that you have to do mentally, to a tool so that you'll be able to focus more on issues that are more important. Or think about it as using a tool to help you to reduce specific mental activity that you have to do when you are doing your research. One example is that using qualitative analysis software to help you to organize the cause and the theme that you have. Instead of doing it manually, which may be a little bit challenging, right? You are using a software to help you to organize. So you have offload some of your mental activities to this tool so that you get time and focus on making sense of your data, interpreting and addressing your research question that you have. So these two are the concepts I'm going to talk about. So in this AI era, this is what is happening. I also created another dashboard, right? So as we are introduced with new technology, our cognitive offload is increased. You may also lose control in terms of decisions that you have to make. One example is qualitative analysis, right? At the beginning, as you can see here, we use pencil, paper and highlighters to help us to manually code our data. We offload a little bit, right? Because using the pencil and paper help us to not have the task of memorizing or putting all the information that comes to mind in memory. It's going to be too much for you. So you use pencil to underline the areas that you are interested in on the data so that you'll be able to go back again and see where you highlight the significant information that you highlighted. So it saves you time. It saves you the stress of looking for the significant information again from the data because you have already highlighted the place that you are interested in, right? So you offload a little bit, right? So the cognitive offload is a little bit low because using the mental capacity to make sense of your data, identify information that are significant and develop codes and themes, you do it by yourself. So the cognitive offload is low. And then agency is high because you have control. You make decisions. You identify information that are significant. You develop codes. You categorize the code to develop themes, right? You have direct interaction with the data. And that's why your agency level is good. As you move on to the next one, using Word document and Excel, your agency level is also very high. The cognitive offload is low because you are just using Word document to maybe manually code in your data. But as you are using a qualitative data analysis software, see how you have offloaded a little bit, almost half of your mental tasks, right? Maybe helping you to organize, to search for information, visually establishing relationship between concepts, right? But as you can also see here, yes, the agency is high, but you have lost your agency a little, but it's not too bad, right? Include AI. See, it's moderately high. You are allowing the AI to suggest code for you, right? So the control over the data analysis process has decreased a little bit compared to when you were using paper and pencil and you are using Word document or Excel spreadsheets. The agency here, yes, you can see it's moderate, right? It's a little bit of concern because you are losing your control. But when you use AI model, see what happens here. You are offloading a lot of what you have to do for the AI to do it for you. Maybe you are asking the AI to suggest code for you, to develop things for you, to write the interpretation, right? To produce a report, right? All these are causing you to depend on the AI system in making sense of your data. And your agency is very low. It's low to moderate. You don't have much control over the data analysis process and you are losing your skill. That's the concern. The ability to code your data, analyze your data is going down because now AI is doing that for you. We are in a stage of AI agent where you can ask AI to do a task for you and it will independently make decisions and then bring you your outcome. In this situation, let's say you want to analyze your qualitative data and you have an AI agent. What you could do is to give all the transfer to the AI and there are three things that you can give to the AI. The background information about a study, the objective about a study and also the outcome that you want. And you leave everything to AI too and AI can go make decisions and then within minutes you get your result. So you have offloaded a lot of important decisions to AI to make. So the cognitive offload is very high when you use an agent and your agency is low now. You don't have control over the decisions that the system makes for you and as you are losing control you're also losing the skill and knowledge. So that is the problem that we're going to face in this kind of situation. I think you have to be aware and in my presentation I'm going to talk about what actions that you can take in order to not lose your agency. To be able to assess the quality of the information the AI too produces. And that's what I'm going to really talk about. I hope this introduction helps you. This is just a brief information. These are the skills that you need. The good thing about this dashboard is that there are some scenarios that you can learn from. You can click on it and get all the information that you need. And also the sales assessment is also here just to help you to know the kind of skill your level of skill that you have in this kind of situation. So when you click on it you click on complete and then to give you the results and then you read more about the kind of skill that you are lacking and what you can do to improve those skills. So this is what I have for you. I'm hoping that you attend this webinar and you're going to learn a lot from my presentation. I'm continuing to improve the presentation and I'm hoping to see you there. Thank you so much for your time.