Using AI to Analyze Qualitative Data in Evidano (Full Transcript)

[00:00:00] Speaker 1: Evidano is a qualitative analysis software that is powered by AI to help you to make sense of your data. There are three main steps that you need to follow to be able to finish analyzing your data with the help of AI. You have the option to upload any kind of data that you have, and you select the analysis that you want to use to help you to make sense of your data, and then you move to the next where you explore the theme that has been generated and also the virtual representation. So these are the three steps that you have to follow in order to arrive at your findings. I'm going to also show you other functions that will be helpful. So basically what you can see here is that you can easily upload your files into the system and then you'll be able to select the analysis that you want. Do you want to do semantic analysis, evidence synthesis? Do you want to use granite theory? So this one gives you the chance to choose the type of analysis that you want to run, and then the system will be able to run the analysis for you and provide you an output. It provides you a theme and the sub themes and also a visual representation as you can see here. It gives you the themes, the sub themes, and also codes. So it displays also the quotations that are linked to the theme, and then it can give you the visual representation. As you can see here, we have the hierarchical graph here. So it shows you all the themes and the sub themes. Here it shows you the connection between the theme that has been developed. You can explore and see what is going on. You can also download if you want to. It gives you also the frequency, the number of participants, or the number of quotations that are linked to a specific theme. And then you can also explore the connection between demographic information and also the codes that you have or the themes. One interesting function that you can also explore is generating and setting up interviews using AI avatar. So this means that you can upload all your interview questions here, select the avatar, and then what happens is that you send links out to potential participants. They click on it and then the avatar asks the question. It has an option for you to explore and see the function. They have three options. You can use the avatar. You can also use only voice. So the functions and the capabilities are limitless. You can do many things if you have qualitative data and explore with the help of AI. So let us explore. I have my data so we can start the process. I'm going to click on dashboard. This is what it's going to show. If you want to see your recent project, you can click on that and it will show you all the projects that you have. So you see that I have all my projects here. The ones that I've done using this software. So I have a project for burnout. And if you want to create a new project, you go here, create a new project, and then you will have the chance to use existing files that you already uploaded. Or you can bring a new file. In my case, I want to bring new files for my new project. So what do you have to click? Upload files. And then I can just drag and drop my files in there. So I have five participants. So I'm going to drag and drop them here. So it's just showing to you that all the five transcripts are here. And then, so this is what you have. You see that everything, the one that you uploaded has been checked now. You click on continue. So if we click on continue, you have to scroll down and then this is what you have to see. You see here, you see all the transcripts you have uploaded. If you have a research question, research framework, any information, anything that you want to add to give the system a context, you can bring them here. You can drop the document here. Maybe you have a research proposal and you want to put it here or in your interview questions. Anything that will help the system to get background information about your study, you can upload them here. You can also manually type that information. So the background of the study. It's very important to do that because you want a system to know what your study is about so that you're not giving you generic information. You want it to give you customized information that is based on your study, the background of the study, your theoretical framework, and any other information that you think it will help the system to know more about your study. In my case, what do I have? Let me try to see what I have here. So I have the purpose of my study and two research questions. So I'm going to just copy that and put that information here. So I have my purpose of the study. The purpose of the study is to explore the causes and solutions of burnout from the perspective of primary health care physicians in the U.S. And I have my two research questions here. One is for the causes of burnout and one is for the solutions of burnout. So when you are done, you click here. You can click here to help AI to extract information that will be helpful. So I'll go ahead and click here and the AI will extract information about, you know, research framework and other things. So you have uploaded a file. Here you are choosing the report structure. How do you want a report to be? You want a theme and then the code under. If you want that, you choose this one. Do you want to see the research question and the codes are under them? If you want that, you can choose that. And a good thing is that they have samples for you to look at to see how the structure is going to be. So you see how this one is. You want to make sure that the information it generates is in line with what you really want. So that's why it's asking you about all these things. So you look at the structure. If it works for you, you can select that. So which one should I select? I guess I want to select the first one. So this one will indicate the theme and then the codes under that and an indicator of which research question is the theme addressing. So I like the first option. What else should I choose? So in terms of the segment comparison, how are you going to compare that? Let me click on the view sample and see. Do you want to see the insides or the output like this one? If yes, you choose that. So I'll choose this one and then I go to next configuration. Here I'm in academia, so I'll choose this one. If you are in evaluation field, you can choose this one. It gives you all the options related to evaluation. And if you want to do market research and choose a third option to have all the options here. So I'm in academia, so I will choose this one thematic analysis for now. You can go further and choose whether you want the reflexive thematic. Let me choose maybe this one, right? So academia and reflexive thematic analysis. So the next step is to go to the next configuration. So the next one, it's asking about the next one is the development of AI themes in a codebook and whether you want to also insert your own codes for the system to use it to run the analysis. So as you can see here, I want AI to generate the themes for me. I will just go ahead and click on get themes and sub themes. Okay, so AI has generated some of the theme and sub themes for us. So you can see here that one of the theme is career stage shapes burnout experience. And then this one is the definition of the theme. And we have the second one here, which is a sub theme and also the definition. So what you have to do is to have to reveal them. And you can always make an adjustment to this one. You can change the name. You can change the definition if you want to. You can also add a code if you want to add that. So you always have to review making sure that everything is correct before you go ahead to the next stage. So let me see. It doesn't indicate whether it's addressing research question one or two. Maybe when we go to the next stage it will indicate that. I don't see the place that shows whether it's addressing. But as you can see here, you can see that some are obvious in terms of which research question is addressing. Yes, this one is addressing research question two about the solutions of burnout. And this one too, you can always delete. You can remove the theme or you can add to the theme. You can add another sub theme. You can remove it if you want to. So it's time for you to like review making sure that everything is right before you click on next configuration. So this is the time for them to indicate the research question. So and this is your research question and you can make an adjustment to it if you want to. So you review and if everything is right for you go to next configuration. So you can see here that now we have the theme, the definition of the theme, and the research question that is addressing. This one is addressing the research question number one. And this is one of the causes of burnout administrative and clinical workload as systemic stressor. So it's addressing the research question and here too if you think it's addressing different research question you can click on it and then choose that research question it's addressing. So that's very good. So this is my initial impression here and I don't know what you are thinking. Let me know your thought in a comment. So I can see that the system is walking through you through the process. It's not just taking your information and then providing you the findings. It's walking you through the process and you are leading the data analysis process. The system is suggesting things to you and then your role is to review and make the necessary changes. The ones that are not addressing the research question you can leave them alone. You don't have to connect any question to that. So you go to next configuration. Here is where you add the demographic information. Right. So if you have the demographic from this a good time to do that. If you don't have you can click here and say start analysis without comparison. So I have the demographic information with me. So I click here to import it. It's an excel spreadsheet. So let me go for it. Let me show you first before I put it there. So this is the demographic information. So you can see all the participants with their IDs and their age, gender, years of experience and ethnicity. So what I'm going to do is I'm going to drag and drop it here and then I click on. So which one should I click on. Guess which one I should click on. So let's see if you want to use AI. Right. Slow but more accurate. And here faster but root base. So which one should I choose. Maybe I should choose the second option. So yes you see here that because I have all the demographic information I chose the second option because all the information is already there. So you just have to review making sure that everything is correct. If you are OK we go to confirm and tag document. So when you do that all the demographic information will be here. The next step is to click on create project and start analysis. And then it's now doing the analysis. You can see the progress by clicking on it. 21.2 percent done. So we have about 79 percent left. I think it's finished extracting all the themes. It's now doing the team analysis and to give you word cloud and chat for you to review. It also can give you summaries for each of the themes and cross comparison analysis. It's going to provide you a summary analysis for each of the files and also overall summary. And then it finalize everything for you. So these are the things that you're going to get. And then you review everything making sure that they are correct for you before you use it for your study. So this is done. So the next step is to view the project. So let's click on view project and see what we have here. OK. So this one is the code book. We have the theme frequencies. OK. Interesting. So this overall. OK. Let me click on this one. So when I click on team frequency this is what you're going to see. So these are all the themes and the number of documents that are connected to. So this one five documents. So five out of five were connected. This one five four documents are connected to this theme and three and two is here for the last one. So let's see. OK. So you can arrange it in alphabetical order if you want to. You can indicate percentages. Here you can also look into the age. And here let's look at gender. So OK. So the gender is females are green males are blue. Let me change the percentage to the number. So two participants from the male group were connected to this theme and one was connected to this theme. OK. So interesting. So let's go to overview. So the overview gives you all the documents and then the demographic information. So you can see here that P1 is 37 female and also African-American. Here you can see that this is the theme. So these are all the themes and then who is connected to what. I can call it like a summary table because it had all the participant information here and also how they are linked to specific theme. So these are the seven themes that was developed. If you click here it gives you the visualization and also the sub-themes. So you can see that the sub-themes that are connected to the main theme. And when you click on visualization you can see hierarchical graph is here. This just shows the connection between the themes, sub-themes and the codes. So this one to you can export it if you want to. Let me go to co-occurrence network. So this one gives you the connection between the themes. Right. So this is a theme another theme here. So this theme is connected to that theme. So it shows how many documents you have in common. The document that they have in common is P4 and then this one is P2. They have in common is heavy work week of hospital rounds and office work and it's linked to clinical. It doesn't show everything but maybe what you have to do you can download it so that you can see everything there. So let's move to the next one. Then what is this one? Sub-themes and their frequencies. So it's quite similar to the main themes and your frequency. This one is the sub-themes. So you can look at each of the sub-themes and then explore the frequency. So there are a lot of things that you can explore. There's another sub-theme. So let me bring this one back a little bit. So this is the second theme. It has visualization. It has the sub-themes. When you click on it you see the code that is linked to the sub-themes and also the definition and the quotation that is linked to the sub-theme. So you can see that we have two participants P1 and P2 are linked to this. So you see how that nicely everything is arranged. When you go to executive summary it gives you introduction, provide you the main themes and the observations. So you just have to read and see how whether it makes sense to you. Remember all these are generated by AI so you have to also review making sure that everything is correct. So if you are an organization and you want me to come and show you how to use a software like this to analyze your qualitative data and also learn about how to evaluate AI to make sure that they are correct, then you can contact me and I'll be happy to organize a training program for your colleagues about how to use a software and also verifying the AI output. I also provide one-on-one consultation in terms of doing qualitative research so you can always contact me and I have my contact information in the description area. It also provides a word frequency. So you see here word frequency, different types. You just have to explore and choose the one that you want. It's not required for you to use the word frequency when you are presenting your findings. You have to make sure that it makes sense to you. So the word frequency is all about showing the kinds of words participants use. So the bigger the font, the higher the number of times they use it. So when you see that and some of the fonts are bigger, this means that they use it more and ask yourself does this illustration make sense to you in terms of addressing helping you to communicate your findings. If it doesn't make sense, you don't have to use it. So just use an illustration that makes sense to you. We have the bar chart here. You can explore and see whether this one will be useful for you. You can download it if you want to. So this is what I have for you. You can also chat with your document. There's an option for you to chat. You can chat with your document. You see that. Let me go back again. You click here and you can chat with the document that is here. Ask questions. You can even use only this chart to help you to analyze the data. Imagine you have four steps that you can follow. You come here and what it's going to do is first you can say that I'm doing this study. I'm focusing on this area. Can you go through the data, identify information that are significant, and then develop codes? And the system will do that. And then you can say that can you arrange the codes based on this research question? And the system can do that. You can also chat that. Can you categorize the codes, develop themes to address my research question? And then the system will be able to do that. This only you can even use only the charts to help you to make sense of your data. So explore and let me know what you think. If you have any questions you can put in the comment section. I will be happy to address them for you. I hope this one was helpful. In terms of the plan, I have a pro plan. But you can also start with a free plan and explore the functions and see whether it's going to be helpful for you to do your research. But if you want to get the pro plan, you can use my code to get 30% off. So let me show you my code. When you use this code, I do 30, you get 30% off. Let me make it bigger. Yes. So this is the code. If you use I do 30, you get 30% off. And then don't forget to subscribe to this channel and share. Thank you so much for your time.