Grounded Theory Explained: Memoing, Coding, and Models (Full Transcript)

[00:00:00] Speaker 1: Hello everyone. Today we're going to have a conversation about granite theory. This is what I'm going to do. I'm going to give you an overview about the granite theory approach. If you want to learn more about granite theory, this book will be the best. Give you all the information that you need about constructivist granite theory, right? So if you want more information, this book will be the best for you. Back to the topic. Granite theory approach. So think about granite theory approach as a way of collecting and analyzing your data with the main goal of developing a theory to explain something. So let's say you want to focus on burnout and want to develop a theory to explain how people experience burnout. You can use granite theory. So whenever you want to develop a model or a theory or an explanation to help you to understand or help you to explain something that you are focusing on to your audience, granite theory approach will be the best. There are two important issues that you have to think about when you are using granite theory, right? So technically you are developing a theory, right? And you ask yourself, what is a theory? In qualitative study, it's a statement that shows relationship between things or relationship between concepts that is used to explain a phenomenon or to describe a process, right? So in granite theory, you are developing a theory. The label granite means that the theory should be based on the data. It's not your own understanding or your own expertise. It's based on the data. You are developing a theory based on what you collected from participants or based on the data that you have. So it's data driven. So everything that you're going to do is data driven and it should be systematic. So these are the two concepts that you have to focus on when you are developing a theory. So let's say you have a request question. How does burnout occur? The burnout is a phenomenon, or you can see burnout as a process, right? And then you are collecting data. Let's say you focus on healthcare practitioners, and then you collect data from them about burnout. And then you analyze, develop themes, and find out the connection between the themes. And based on that, you'll be able to come up with an explanation that is a statement that shows the connection between explaining the phenomenon, which is the burnout, right? So the next thing that you have to really think about is memory. It's very critical when you are using granite theory approach. You have to have a way of documenting your action decisions and your reflections, right? So documenting your action decisions and your reflections is very, very important when it comes to utilizing granite theory approach. So let's start about memory. Memory is more about taking note of all the information that comes into mind as you are collecting and analyzing your data. And why do you have to do that? First, it helps you to be focused because you need full attention in collecting and analyzing data. Any information that comes to mind as you are doing something, you don't want to forget about it. So you have to put them aside by documenting it. And then after your activity, you can go back to the data, or you can go back to what you have recorded and review and learn more about it. So it helps you to be focused, right? Because you need a full attention of interacting with the data, collecting data, so that you'll be able to develop a theory to explain a phenomenon. And also documenting can help you to take note of your ideas and reflections that may be useful as you are going through the process. So when you are analyzing the data, many issues come to mind. Many great information comes to mind, but you don't want to stop and focus on that great information. So you want to document that. It also helps you to document your preconceived ideas and perceptions. Because when you are analyzing the data, as a qualitative researcher, you are an instrument. This means that all the information pass through you so that you can make sense of that data. And then you document the process and put every information down. But very important here is that if you don't take care, all your preconceived ideas, your biases, your perceptions, your background may impact how you make sense of the data. So in order to prevent them from unduly influencing what the sense you are making, you have to document those. You have to set them aside. So one example is that, let's say you are working on a burnout, right? A topic about burnout among primary health care physicians. And then you have experienced burnout before. And that's why maybe you are interested in that topic. So it's very important for you to document all your preconceived ideas. Maybe one of your preconceived ideas is that if you are working in the health field, there's a high probability that you have burnout. Maybe that's your perception. You have to put them aside. Because if you don't do that, it will cloud how you make sense of what they are presenting to you. If you have experienced burnout before, you have to put them aside. If you have some expectation about what you're going to get from the data, you have to put them aside. And so we call them bracketing, reflecting, documenting your preconceived ideas and putting them aside so that they will not have undue influence of what you are interpreting or making sense of, right? It doesn't mean that your background may not influence. It may, but you are making sure that you are managing those influence, right? Being aware and setting them aside. And also, for memory, you can also record your decision and actions. Because at the end of the day, you have to show your work. If you want people to believe what you found after analyzing the data, you want to show your work. How did you identify information that are significant? How did you develop themes? How did you connect the theme to develop a theory to help you to explain a phenomenon? You have to show your work. How do you show your work? Start documenting every action and decision that you are making in your study so that you'll be able to clearly demonstrate how you come up to your conclusion so that people will believe what you found. It improves credibility if you do that. So these are the things that you have to think about when you are using a grounded theory. There are about, you know, three main stages that you have to pass through when you are analyzing your data. And I'm going to briefly talk about them. And also, if you have any questions, you can put in the comment section. I'll be happy to address them for you. And also share my videos and also subscribe to my channel so that when I post any video that is useful, you'll be able to get access to that video as soon as possible. So these are the three main stages in terms of using grounded theory approach. First, initial coding. Initial coding, the beginning way of making sense of your data. So imagine that you have your data, you have collected your data, you have reviewed, you have familiarized yourself with the data. So the next step is to conduct initial coding, right? And initial coding is where you go through each sentence, you go through what we call a line-by-line coding, go through each of the participants' information, and then extract information that are significant based on your research question or based on your phenomenon that you want to explain. And then you develop a label that represents that, right? So let's say if one of the participants talks about the fact that the main reason why I'm having a lot of stress when I'm working is that I have so much to do and I don't know what to do to manage that. So that is the significant information because it talks about burnout. So what do you have to do? You develop a label, maybe having higher workload can be a label representing the significant information. So that's initial coding. You go through the data, any information that are significant, you develop a code, right? And then you can use one code more than one time, right? You can use it more than once. Let's say participants are talking about an issue, the same issue all the time, you can use a specific code to label that. So see codes as containers. If you see any information that are significant, you drop it into a specific code or a specific container. So at the end, you're going to have a lot of codes, right? Then you move on to focus coding. Focus coding is very important. So focus coding, first step is to identify all the codes that are dominant. And you may ask what makes a code dominant, right? Or what makes a code important? Can you put it in the comment section? If somebody say that I have identified a dominant code, what is the person telling you? Or how do you define a dominant code? There are two main criteria that you use to determine a dominant code. You can determine it first based on numbers. How many participants are connected to a specific code, right? So let's say you have 10 codes and one of the codes, maybe 8 out of 10 participants are connected to that code, right? And the rest are maybe 7 out of 10 and 2 out of 10, right? So the dominant code is the one that has 8 out of 10 participants. But not only that, you have to consider the number of statements that are connected to that code, a number of quotations, right? So all the significant information that are connected to the code, how many of them, right? So based on a number, you can determine whether the code is dominant. Not only that, a dominant code is a code that is really explaining or connected or close to the phenomenon that you are focusing on. A dominant code is a code that has so much rich information in terms of the context. A dominant code is a code that most participants have confirmed the existence of the code, right? So you have to think about all these factors to help you to come to a conclusion about the dominant code. Sometimes it's a little bit subjective because sometimes I may see this code as dominant but other person may see it as not dominant. This is where you can involve other researchers or involve a second person to help you to decide which of the codes are dominant, right? So as you can see there are a lot of factors so you can choose the factors that can help you to determine the dominant code. But I think that the lesson here is that don't focus on only numbers. Don't focus on the number of participants connected to the codes and also number of statements or quotations connected to the codes. You also have to consider the richness of that code in terms of explaining the phenomenon. Is it very close to the phenomenon that a participant gave very good rich detailed information, deep detailed information? This is a little bit of subjective process that you have to go through. And the second, the other way is to invite another researcher to help you to decide which one is the dominant code. So after identifying dominant codes, what do you have to do? Then you, let's say you identify maybe three or four or five dominant codes. The rest of the codes that are less dominant, you connect them to the dominant code, right? Another strategy is that you can do a sorting strategy, right? A sorting strategy is that like you don't have to identify the dominant code. What you're going to do is to explore the characteristics of each of the codes and then see whether you can group them into clusters or groups. So at the end, maybe you have 20 codes and at the end of the day, you group them into maybe five groups, right? And then you label those groups and then that label will be a theme. So at the end of the day, you should come up with themes that are helping you to address your research question, also helping you to explain the phenomenon. So that's all about focus code. And if you want to know more, as I said, you can read Shama's book about Granite Theory. Then now that we have about five or six or seven themes, you move to the stage, we call it theoretical coding. This is where you explore the connection between the themes. What is the connection between the themes, right? Are they happening at the same time, which is the concurrent relationship? Is one theme happening before the other? Is it sequential? Is one theme under the other? Is it embedded relationship? You explore the relationship between the themes and at the end of the day, you come up with a proposed model or a theory. Then you move on, you compare what you have developed to the data that you have, right? Because Granite Theory, you are not just developing the theory. It should be based on the data. It should be data driven. So you have to compare what you have developed to the data. Is the data confirming or supporting the theory that you have developed? So that's the process. And this one is it's a continuous process. And I'm going to talk a little bit about that, but these are the main three stages that you have to pass through. And every decision should be based on the data, right? So if you are in a focused coding stage and trying to connect codes and all group codes, you always have to think about, does the data support that connection, right? So it's a back and forth process. So also I just want to emphasize, I was talking about the last stage where it's called theoretical coding. You explore relationship between themes, right? There are different types of relationships that you can explore. One of them is called concurrent relationship. Are the themes happening at the same time? We have chronological relationship or sequential relationship. That's one theme comes and then followed by another theme. We also have divergent relationship. Are the opposites contradicting each other? We also have a better relationship or hierarchical relationship. It's one theme under the other, right? We also have a causal relationship. It's one thing leading to the other. So these are the types of relationships that you can explore based on the themes that you have and also the data, right? So would the data support these kinds of relationships? As you are doing that, you are thinking about coming up with a theory, right? As I said, a theory is just a statement that shows relationship between themes or concepts explaining the phenomenon that you are focusing on, right? So as I say, it's an iterative process, right? And these are the concepts that you also have to think about when you are thinking about the three stages that we talk about in terms of the coding. One strategy, which is called constant comparative analysis. It's constant comparative because you are comparing what you are developing with all to the data that you have. It's continuously comparing what the cause and the categories and the themes that you are developing with the data that you have, making sure that everything that you do is doing supported by the data that you have. So let's say you have grouped them into categories, right? The cause into categories. Does the data support that, right? So, you know, always using the existing data or new data to help you to establish a connection is very important. You can also do the same thing when you reach the theoretical coding stage where you explore relationship between themes, right? So the constant comparative analysis is so important. Constantly comparing what you are doing with the data, right? To make sure that the data reflects what you are doing. Sometimes you may have to go back and interview some participant to get more information, especially when you have limited information in terms of coming up with a connection between the codes or the themes, or you want to develop a theory, but the existing data that you have, the initial data is not enough. Then you have to also, we call it theoretical sampling. Going back to participants that can give you rich information and collecting that information and then using that information to compare the new data that you have with the outcome that you have developed. Another concept that you have to think about is called adaptive resonant. It's a combination of inductive and deductive resonant, right? So inductive means that you are moving from having a data to developing a theory. So normally in qualitative study, we follow the inductive resonant. We collect data, analyze and develop themes, and sometimes go beyond developing themes to develop theory. Deductive means that you already have a theory or hypothesis, and then you want to collect data to test the theory and see whether the theory is true or not. This is done mostly when you are doing a quantitative study. But in granite theory, you have both approach. So combination of inductive and deductive is called adaptive. It happens because at the beginning, you don't have any theory. You are going through the data, identifying information that are significant during the initial coding stage. You go to focus coding, you categorize the codes and then have some themes or categories. You go to theoretical coding, you explore the connection between the themes, and then you develop a theory. But when you develop a theory, you have to go back and apply constant comparative analysis, comparing what you have proposed in terms of the theory to what is in the data. So you see that at the middle, you suggest a theory based on the initial data that you have. You go back to the data again, or you collect more data and see whether the data is confirming the theory that you have developed. If it's not confirming, can you use the data to adjust the existing theory? Can you use the data to reject the theory? Maybe the theory that you've already gotten is not reflective of the data that you have. So adaptive reasoning is first developing a theory and then using new data to help you with the help of the constant comparison analysis to help you to adjust or support the theory that you have developed. Another concept is saturation. So there are two main types of saturation. I think maybe you may know the first one, which is common. The first saturation is when at the data collection stage. When you're interviewing participants, as you continue to interview a lot of participants, you reach a stage where no new information is being collected, right? There's a repetitive information. You hear it several times. So because of that, you stop the interview and focus on analyzing your data. There's also another type of saturation when you are analyzing your data. As you are making sense of your data, you are comparing the theory with the new data that you have. As you are comparing, doing constant comparative analysis, right? You reach a stage where introduction of the new insight doesn't change the existing theory. All the new data is supporting the existing theory is not changing anymore. Then you can think about the fact that you have reached saturation. So these concepts are very important as you go through the data analysis process. So the most important information here is that it's not like a straight step-by-step process when you are using a granular theory approach. It's a continuous refinement of what you have developed, right? You may have to refine the initial code. You may have to refine the focus coding outcome. You may have to refine the initial themes based on the data, and sometimes have to go back and collect data and analyze again, right? It's a continuous process, and that is very important, and that's unique about granular theory approach. And the next one that you have to think about is that there's a lot of myths about granular theory approach, but the two main ones that I want to talk about is that the first one is that some people think that when you're using granular theory, you have to develop a theory. And that is really wrong, because what happens is that if you focus on that, I have to develop a theory, this means that you may be tempted to develop a theory, even if you don't have substantial information from the data that supports the theory, because your end goal is to develop a theory, although you don't have enough data, right? Sometimes what happens is that you may not have enough data to develop a theory. What do you do? You can stop at the focus coding stage, right? You can say that, okay, after doing the focus coding, I don't have enough data to go further and do theoretical coding, right? And that's fine. The most important is you are transparent. But if you focus on the aim of developing a theory, no matter what, this is where you can come up with a theory that has nothing to do with the data, or you don't have enough information to support. So you can use granular theory without developing a theory, but the most important thing is to be transparent and say that maybe I reached the focus coding process and I was able to develop themes, but I couldn't go further because I don't have enough data, or I don't have the resources to collect data from participants, more data from participants, to go to the next stage, which is the theoretical coding. Or sometime you reach the theoretical coding at the initial stage, but you don't have, you didn't go further to collect new data to test the theory. So you can see, you can conclude that I have a proposed or suggested model or a suggested theory that future researchers can work on and, you know, confirm the theory. So the conclusion here is that you don't have to develop a theory if you don't have enough data to develop a theory, right? The second myth is that you have to develop one theory. You can develop multiple theories. It all depends on what the theory is explaining, right? Let's say you are focusing on two main aspects of a concept, right? One theory can explain one aspect, another theory can explain another aspect, right? So these are the two main myths, but as I said, there are a lot, but these are the two main that I want to really share with you. Also, you can also, in the comment section, you can also put some of the myth there. And if you have any questions, let me know, and I'll be happy to provide you my answer. And also, I provide methodological consultation. If you are interested, you can contact me. I put my email address in the description section there. My email address is info at drphilipadu.com. You can contact me so that we can have a discussion about your study. So this is what I have for you. Let me know what your thoughts are, and then I'll be happy to address all your questions for you. And also, thank you for your time. I really appreciate it. Bye for now.