Second blog post - Theme 6: Qualitative and case study research
During the seminar we discussed case studies. With a case study you can obtain a deeper understanding of a phenomenon. You focus on one or a small amount of cases, though it is important to be aware of that the knowledge that you gain can be limited to be representative to that specific case. A case study is something that you use when you know very little about a phenomenon. After you build a theory upon that case. It can be applied when you don’t know or expect anything and there is a possibility that anything can be found in the research.
A case study is not a research method in itself. You can apply either a quantitative or a qualitative method or a combination. In theory-building research it is important to not have a theory under consideration and no hypotheses to test. Otherwise this might be a factor that would contribute to subjectivity and limit what we are trying to reach. Basically we should try to avoid to think about relationships between variables and theories.
Overlapping data analysis with collecting data gives you a good overview of the analysis and makes it possible to be more flexible. The possibility to make adjustments during the data collection process is apparently what characterize theory-building case research.
Within case analysis processes permits the unique patterns of each case to be revealed before researchers generalize patterns across cases. When you reach closure you stop adding cases and stop iterating between theory and data. The first means that the incremental learning is minimal. The second is that the iteration process stops when the incremental improvement to theory is minimal. The final product is hopefully a conceptual framework or a concept or it could just be a disappointment and replacement of an existing theory.
We also discussed semi-structured interviews. In order to be able to conduct valid semi-structured interviews it is important that the participants are at the right level of knowledge in the subject that is to be examined. If you want to examine a certain question that requires an expertise in that area you must target the participants otherwise you won't get accurate answers. Wrong answers will generate misleading results which in the end will not contribute to new knowledge. For instance in social science it can be an investigation of couples. Then it is crucial that the participant is in a relationship otherwise they do not have the knowledge in the area.
With semi-structured interviews you are using open ended questions which gives you the possibility to ask follow up questions which depends on the answers that the participants gives you. This type of method can generate a large amount of data which is time consuming to analyse. The benefit is that you can obtain knowledge that would not be possible to extract from closed interview questions.
Hi! Thanks for an interesting post!
SvaraRaderaYour comments on how a case study can be applied on cases of which researchers do not have prior knowledge, and hence it enables any kind of result to show itself. I think that is a very important aspect of case studies, and I am glad to see you bring it up.
Case studies do not origin with set out hypotheses, but instead offer researchers the ability to pose them when sufficient data is gathered to pose relevant questions. Just like you write, the possibility to adjust the data collection process (and hence let analysis and data interact) is essential to the concept of case studies. This overlap offers researchers a flexibility that I have found hard to identify in other methods, and I am happy to see you bring it up in your post. It's a sign of understanding!
Your notes on semi-structures interviews were also interesting, even though I would have enjoyed reading a little bit elaborated passage about the validity of study samples. I think you illustrated it well with the example of a couple, but should I not have done a great deal of qualitative studies in my past I would probably have needed some more info. However, your comments on how invalid samples lead to misguiding results probably would have made me understand the importance of making sure that your are interviewing the right people!
Good job!
Hi! Thanks for your interesting explanation of the case study it seems that you have a good understanding of the theme!
SvaraRaderaI would like to add some to the semi-structured interview. During the seminar my group also discussed this, as it was one of the methods used in the chosen paper. And I also came to the understanding that the amount of flexibility also means that the questions that are asked should be well thought of as this can influence the result. We discussed how certain words as ‘never’ and ‘ever’ in a question influences the answer that the participants give, as they often end up replying in the same manner ‘no, never’. While if the interview was held what strict and good formulated questions this phenomena could have been avoided.
Den här kommentaren har tagits bort av skribenten.
SvaraRaderaI like how you raise the question of generalizability of case studies. One way of increasing the generalizability is to compare the concepts, theories and hypothesis with extant literature. It would have been interesting if you had expanded on the subject by including the concepts of grounded theory and triangulation. I would like to add to your description about iteration and the overlapping of data within a case study setting that this process should not exceed ten cases according to the article for this theme. I would also have liked some explanation on why you were discussing semi-structured interviews in relation to this theme.
SvaraRaderaHey! In our seminar we did not come across semi-structured interview so it was indeed interesting to read your thoughts about it. I think gathering valid data is always a hard case when it comes to interviews and questionnaires, however, semi-structured interviews benefit in a way that a participant can ask follow up questions which allow to understand question completely. These kind of interviews, in my opinion, allow informants the freedom to express their views/ideas in their own terms thus, as you say - it can provide reliable, comparable qualitative data. Thanks for your reflection!
SvaraRaderaI think what you wrote in the beginning, that case study can be applied when the researchers have little or no previous knowledge, is very interesting. And that after you build a theory upon that case, that case study can be applied when you don't expect or know anything, and that opens up for the possibility of anything being found, is very interesting and important. It's a very important aspect of case studies.
SvaraRaderaI think what you wrote about semi-structured interviews is interesting as well, and something I've worked a lot with before. I agree with you that it's very important to choose the right participants, but also to ask the right questions. To really know what you want to get out of the interviews.
A very well written reflection on this weeks topic. You cover all the major points. For example that case studies are mainly used to come up with research questions in an previously fairly unexplored research field. You also mention, that by not having a specific hypothesis before hand, researches can keep a more open mind and look at what is happening instead of trying to see something that relates to their hypothesis. You also mention that by using qualitative methods, researches might become aware of problems that they did not know and would not have come up with themselves. However you also mention, that case studies are not limited to qualitative methods, but can use any method as long as they provide to finding a research question.
SvaraRaderaNonetheless I think the level of knowledge necessary by interviewees depends on what one is trying to investigate. And especially in theory building case studies I think the level of knowledge about the topic does not need to be very high. To stick with your example: if there would not have been prior research on couples and one would start investigating in that direction the perception of people not in a relationship could still be of interest. That being said after the initial researches and after one has a more in-depth hypothesis it will become very important, that the interviewees are actually in a relationship.
To go into a case study with a clean plate is well emphasized by you. What you mention about the quality and background of a sample is also something I find very important. It's a good example to mention a certain experience such as being part of a couple. You also would have to consider the fact that people (might, sometimes) lie. Also, the reason why they want to participate in a study. There's a film wherein a couple pretends to get married so that one of the partners can get a green card in the US. Just saying that no matter the study, these are things to also consider. How can you ensure the authenticity of a sample? Maybe it's also harder within social sciences, when you want to verify the nature of a certain relationship which, in essence, consists of a social contract between (in this example) two individuals.
SvaraRaderaOy,
SvaraRaderaI totally on the point that you are making that we need the participants in the study on their right mind. If they have a definition or a wrong perceptions of something, their result would be altered and useless to the original research, in some case the misperceptions could be totally off track.
Thanks for the reflection
Hello! Thanks for interesting blog posts both before and after the theme. I got the feeling you even deepened your understanding compared to the first post. :) In addition to what you self wrote little about the problematics of generalizability, I myself also started thinking more thoroughly about the validity and verification especially in the context of case studies that are conducted by using qualitative methods, such as semi-structured interviews. When examining a phenomenon in a specific time and space by interviewing different people related to the same phenomenon, there's often a conflict between the narratives. This sets the researchers into an interesting position, trying to solve the reasons behind these contradictions. One reasonable solution is cross verification, meaning the narratives are strengthened when ever multiple sources have verified them. Another commonly used term in social sciences, triangulation, is also a good approach for case studies because of its complex nature: combining two or more methods when measuring the same subject!
SvaraRaderaAnyway, thank you again for these and well done!
Hi,
SvaraRaderaI found you’re reflection quite interesting.
I think I might disagree when you’re stating that a case study research is not a method, because I believe it shows the same features as other concept of methods.
It might not have specific, commonly used structures or processes, but it is a method in means of always approaching a specific phenomenon and thereafter letting the process evolve as various data are collected. I’d say that it is a method that uses sub-methods, combining qualitative with quantitative in order to reach further understanding instead of a specific result. Also, although it lacks a hypothesis, case studies are founded on a research question in order to establish some sort of direction.
However, as it has no specific process, I can see why you chose not to call it a method.