Discuss the criteria for judging the quality of research designs--construct validity, internal validity, external validity, and reliability-- and discuss how you can satisfy these tests with your case study.
This is a long chapter...i am still reading and will post soon. I am a bit confused on the four tests and so a bit of rereading is necessary. I'll post soon.
Julie, I feel your pain this is a long chapter with a lot of detail and scientific "talk". This is what I have in my notes so far... 4 tests are commonly used to establish the quality of any empirical social research. These are construct validity, internal validity, external validity, and reliability. Construct validity refers to the use of correct operational measures for the concept being studied. You can do this by using multiple sources of evidence, establish a chain if evidence, have key people informed in the area studied review you draft. The researcher need to avoid using subjective judgments when collecting data. The author also mentioned when using multiple sources of evidence to do it in a manner that encourages convergent lines of inquiry. I have no idea what that means. I'm thinking it would mean that you want to make the reader think the same question you are researching and not develop other questions that would lead to something not covered or in a different arena, something along that ,line? Not sure if anyone has any other ideas let me know. Internal validity is mainly a concern in explanatory case studies or casual. One would look for pattern matching, explanation building, addressing rival explanations, and the use of logic models. This part addresses the inference part of a case study. An investigator will infer that something happened due to something that previously occurred. In order to prove that this inference is correct the researcher will need to need to back it up with interviews and documentary evidence. They will need to cover rival explanations in a way that proves their inference to be the correct one. Basically, is your theory "airtight"? External validity deals with deciding if the findings are generalizable beyond the case at hand. It uses theory in single-case studies and logic in multiple-case studies. It talks about how the theory would need to be tested in several different situations to be generalizable. This one I understand the criticism but I don’t understand how to overcome the it, especially for us. We wouldn't be able to test and test and test our theories over the next year in order to prove them. I thought the author was very vague here and I will need way more info to determine how we will accomplish this one. I think that's one we should cover in more detail in class. Last, reliability, refers to the ability of your case study producing the same findings if someone else followed the same procedure and conducted the same study all over again. The goal is to minimize the errors and biases in a case study. You can do this with super documentation. Make as many steps as operational as possible and to conduct research as if someone was always looking over your shoulder! I would think that doing a good job here would help with the external validity. Now how will I do this???
Construct validity- I would think part of this would be covered in your literature review. Here you would show commonalities in previous research and your theory.
Internal validity- The inference part I will make sure and cover why other conclusions could not be possible based in the research and documentation I have collected and any previous research conducted that could rule out a different outcome.
External validity- I have no idea will need help here as I mentioned earlier.
Reliability- I think your documentation here will be the key. I will write each step like a recipe. This reminds me of when we have the kids write the steps in making a peanut butter and jelly sandwich and then try to follow their steps. If their steps are clear, detailed, and make sure not to leave anything out then you can end up with a real mess.
If I interview more than one student is that what a multiple-case study is?
On page 30 In the second full paragraph the author tells about the proper way to word questions is to avoid being too vague. Am I understanding this correctly? You want to sound like you are favoring one way.... for example with my study I would ask something like.... Does using palm pilots in spelling engage you more than without using palms? Is that narrow enough?
WOW.. I hate this chapter. I think it is a little more advanced than the last one.
Kristin did a good job of breaking it down... at least I think so.
The four tests Yin and other text books have identified are construct vadility, internal vadility, external validity, and reliability. I like the chart on page 41, It really helps me break down the information. From the reading I have gathered (or tried to understand) the following:
Construct Validity: I think a way to state this in my own words is the "state the evidence" Researchers need to see what is already out there in terms of what they are studing before just jumping in. Internal Validity: This is a way for the researcher to back up their research. The researcher must "prove" x lead to y (without some other letter being involoved). External Validity: Making sure that the results are repeatable and not just a one time thing. Reliability: Be sure to document what you are doing so the test could be repeated. Minimize errors and biases.
As I continue reading I am now at page 47... WOW way too much information for one chapter. I am overwhelmed.
I asked earlier about multiple case designs. I didn't know at that time there would be several pages on this topic. However, I am still confused. Someone please help!
Alright... Here it goes. If some of you didn't know I plan to do my research on palm pilots in the classroom. I don't use them on spelling dictation tests, however another third grade teacher does. If I survive this book I plan to base my case study on this. Here is how I think my case study will be represented by the four design tests. Construct validity: As of now I have not found much for my literature review. To cover this part of validity I will have to check my literature review. Internal validity: I will have to prove that there are not other things that could have altered the dictation tests scores. External validity: I am not sure how I will cover this since the case study will only be done once by me while at MWSU. Reliabilty: I will have to document everything. I will be sure to keep my documents and interviews in a safe spot.
I hope this covers what I needed to have done. I am overwhelmed.
Would any of you like to meet at Mo West to actually fill out the rest of the IRB information? Call me up!
Question: What the heck is next??? Quote: "You need a plan." page 25 Yeah and I need a cocktail! Fact: Document everything!
I would like to make an appointment with the library to see if I can find more information on palm pilots used on spelling tests. Is there a certain person I should ask for? Also, If they can't find other information how do I go about documenting that there is not any additional literature out there? Does that make sense? If not call me!
Yes, it makes sense. Contact the library and ask if they have a research or reference librarian. If they do, ask to set up an appointment and see that person. If they do not, just go in and ask to speak with a librarian and tell the person what you have done so far on your research with your literature review and ask the person if he or she has any other suggestions for you.
If there is nothing else out there, then you will document everything that you found and how it is similar yet different than your topic, and you will make the argument that your research is filling a hole in the field.
If you need to see me too, please set up an appointment, and I will be more than happy to meet you! Take care and keep reading and (re)reading Yin, you will get it! Finish your IRB as soon as possible, Dr. Hendrix
GREAT job here! Multiple sources of evidence means that you will have a variety for your data collection process. You will have field notes and interviews; you may have someone in the school check your notes and interviews to add validity to it. You can even have your participants give you feed back on your notes and interviews to help with the traingulation and validity of your research too. Just remember you want to collect data in as many ways as possible, and you want to triangulate your data too.
You will minimize errors with your triangulation of your data. We will talk more about this later. See my earlier comments. Do you have someone who can verify your notes--someone who comes into your class perhaps?
If that does not make sense, let me know, Dr. Hendrix
This is a tougher chapter, but it is digestable. I agree. Break it up into small chunks and reread it if you need to do so. Also, I am available by phone or in person if you need me, Dr. Hendrix
The following information is from Wikipedia, and I think that you might find it useful--especially the black swan part. I think that is a good image you can remember.
Generalizing from case studies
The case study is effective for generalizing using the type of test that Karl Popper called falsification, which forms part of critical reflexivity [3]. Falsification is one of the most rigorous tests to which a scientific proposition can be subjected: if just one observation does not fit with the proposition it is considered not valid generally and must therefore be either revised or rejected. Popper himself used the now famous example of, "All swans are white," and proposed that just one observation of a single black swan would falsify this proposition and in this way have general significance and stimulate further investigations and theory-building. The case study is well suited for identifying "black swans" because of its in-depth approach: what appears to be "white" often turns out on closer examination to be "black."
For instance, Galileo Galilei’s rejection of Aristotle’s law of gravity was based on a case study selected by information-oriented sampling and not random sampling. The rejection consisted primarily of a conceptual experiment and later on of a practical one. These experiments, with the benefit of hindsight, are self-evident. Nevertheless, Aristotle’s incorrect view of gravity dominated scientific inquiry for nearly two thousand years before it was falsified. In his experimental thinking, Galileo reasoned as follows: if two objects with the same weight are released from the same height at the same time, they will hit the ground simultaneously, having fallen at the same speed. If the two objects are then stuck together into one, this object will have double the weight and will according to the Aristotelian view therefore fall faster than the two individual objects. This conclusion seemed contradictory to Galileo. The only way to avoid the contradiction was to eliminate weight as a determinant factor for acceleration in free fall. Galileo’s experimentalism did not involve a large random sample of trials of objects falling from a wide range of randomly selected heights under varying wind conditions, and so on. Rather, it was a matter of a single experiment, that is, a case study.
Galileo’s view continued to be subjected to doubt, however, and the Aristotelian view was not finally rejected until half a century later, with the invention of the air pump. The air pump made it possible to conduct the ultimate experiment, known by every pupil, whereby a coin or a piece of lead inside a vacuum tube falls with the same speed as a feather. After this experiment, Aristotle’s view could be maintained no longer. What is especially worth noting, however, is that the matter was settled by an individual case due to the clever choice of the extremes of metal and feather. One might call it a critical case, for if Galileo’s thesis held for these materials, it could be expected to be valid for all or a large range of materials. Random and large samples were at no time part of the picture. However it was Galileo's view that was the subject of doubt as it was not reasonable enough to be Aristotelian view
By selecting cases strategically in this manner one may arrive at case studies that allow generalization.
For more on generalizing from case studies, see [2]
[edit] Assumptions 1. Cases selected based on dimensions of a theory (pattern-matching) or on diversity on a dependent phenomenon (explanation-building). 2. No generalization to a population beyond cases similar to those studied. 3. Conclusions should be phrased in terms of model elimination, not model validation. Numerous alternative theories may be consistent with data gathered from a case study. 4. Case study approaches have difficulty in terms of evaluation of low-probability causal paths in a model as any given case selected for study may fail to display such a path, even when it exists in the larger population of potential cases.
[edit] History of the case study As a distinct approach to research, use of the case study originated only in the early 20th century. The Oxford English Dictionary traces the phrase case study or case-study back as far as 1934, after the establishment of the concept of a case history in medicine.[citation needed]
The use of case studies for the creation of new theory in social sciences has been further developed by the sociologists Barney Glaser and Anselm Strauss who presented their research method, Grounded theory, in 1967.
The popularity of case studies in testing hypotheses has developed only in recent decades. One of the areas in which case studies have been gaining popularity is education and in particular educational evaluation. [5]
Case studies have also been used as a teaching method and as part of professional development, especially in business and legal education. The problem-based learning (PBL) movement is such an example. When used in (non-business) education and professional development, case studies are often referred to as critical incidents.
[edit] History of business cases When the Harvard Business School was started, the faculty quickly realized that there were no textbooks suitable to a graduate program in business. Their first solution to this problem was to interview leading practitioners of business and to write detailed accounts of what these managers were doing. Of course the professors could not present these cases as practices to be emulated because there were no criteria available for determining what would succeed and what would not succeed. So the professors instructed their students to read the cases and to come to class prepared to discuss the cases and to offer recommendations for appropriate courses of action.
Business case studies recount real life business situations that present to business executives a dilemma. The case puts the scenario into the context of the factors that influence it. Cases are generally written by business school faculty with particular learning objectives in mind and are refined in the classroom before publication. Relevant documentation or AVitems and a carefully crafted teaching note often accompany cases.
Annie, I would like to meet to do the IRB stuff together. I need to get with Kelly this week and she's going to help me word my thing for Mr.Lawrence so that he will ok it and maybe we can use the info I gather to help her case for a new reading curriculum. After that then I'll know more on what to put on my consent forms and everything. You're way ahead of me so if you want go ahead if not then I will definitely meet with you and maybe Dr. Hendrix could join us! Let's meet this week anyways and work on our first journal. Reading your take on how to use this with our case helped generate some ideas for me so thanks! Hey I also found a website that had short case studies on how different teachers used palms in the classroom. I'll give it to you I don't think these are the kinds of case studies we are working on but I didn't know if there were sources they used that would help you too. See if this link works http://www.k12handhelds.com/casestudy.php
Yes, Annie the chart is helpful but am I correct in thinking that this criteria is to be used to determine if we have a possible case study? It seems like a whole lot of work before we ever get started. It appears that the external validity is the most confusing. According to the example o pg 43, "if a study of neighborhood change focused on one neighborhood, are the results applicable to another neighborhood." So in my possible case study, I am considering using audiobooks/recorders to increase reading enjoyment for low performing students at Webster , a title I school. So with external validity I could generalize that this approach would benefit any low performing student at a title I school. Is this correct?
Laurel's post is as follows... "One prerequisite for allowing other investigators to repeat an earlier case study is the need to document the procedures followed in the earlier case. Without such documentation, you could not even repeat your own study" (pg.45) In other words your study would not be reliable.
A question I have...from pg. 38It said a common topic of case studies is the evaluation of programs (state, federal or local) which in the research I purposed was a locally implemented reading program. But it further states that the theory needs to distinguish between the substance of the program and the process of implementation (e.g., how to install an effective program). And that is what has been bothering me. How can --or should I --try to measure our program results if there is question about how effectively it is being put into place? It goes on to say that this mismatch can be avoided by paying closer attention to the substantive theory. OK- how do I do that? Do they mean concentrate on the substance of the program, and not how we are going about it? I think my research focus needs help!
I said...
"For your question here, you should study your program. Point out the problems with how it has been implemented as a limitation. You still need to study it though. For the theory, tie it to what the program is about more than the theory for implementation now. Later, you could keep working on your study and add that part in another case study.
No Worries!! You are doing well. I can tell by the questions you are asking!!!"
Be careful about generalizing, but if someone else can repeat your study and you have good external validity, that will give your work more weight and can be generalized more easily. You would only be able to make some inferences about a school like yours and/or students like yours. Remember the black swan. Case studies help point out the black swan from the info. earlier.
I swear technology hates me! I am glad that I am taking the class with Dr. Frick so she could help me fix my post. I want to start off by saying...Annie, you would be the one to mention drinks! Unlike Julie, I am struggling with construct validity. Am I missing something or is this just setting up our project? This book is much easier to read, but currently I feel as if I am on information overload. I have some questions that I am thinking about asking for my interviews and I wanted the class opinion as to whether these are good questions or not. Please remember that I am researching the correlation between taking honors courses and ACT and AP scores. For the students the questions are:
1. Do you take any honors classes? Have you taken any in the past? 2. Did you take Honors by choice or was it a requirement? 3. How did/does your Honors class differ from non-honors courses? 4. Do you think that there are benefits to taking an honors course? 5. Do you feel that honors courses better prepare for your future? In terms of ACT? In terms of AP?
The teacher questions are as follows:
1. Do you teach any honors courses? 2. How do you think the one honors course requirement has affected the chemistry of your class? Does it create any challenges? 3. Do you think that students benefit from the requirement?
Are these questions anything similar to questions that I should be asking in an interview? Thanks!
You want to ask open-ended questions in your interviews. A lot of these are going to have yes and no answers. You want to get as much information and data as possible though. The "how" question here is more open. How do you think students benefit from the requirement? What are the benefits for honors' courses? Try rewording these.
Is it necessary to ask yes and no questions to get a certain sample, or should I ask these to see which kids I want to interview and have two sample groups? That was wordy...what I mean is do I want to have a group of kids that has taken honors and then a group that has not and compare their answers or stick to only kids that have taken honors courses? I am getting a little confused, not about the open-ended questions, I understand your suggestion, but now about my sample group. Is this the construct validity concept I was getting hung up on earlier?
I need a drink too, Annie, especially since I just got the book and read both of these chapters in one day. Dr. Hendrix, I am going to ask a stupid question that I'm sure is in the syllabus if I'd just go and look, but are we supposed to answer your discussion question and/or do the quote/question/fact here? I'm understanding that we should do both, but everyone seems to do something different, so I'm going to ask!
Quote pg 60-61: "...when you have the choice (and resource), multiple-case designs may be preferred over single-case designs. Even if you can do a a "two-case" case study, your chances of doing good case study will be better than using a single-case design. Single-case designs are vulnerable only because you will have put 'all your eggs in one basket.'"
If I conduct my research on how having access to computers for writing tests affects student scores on two of my classes instead of one, ill this be considered "multiple-case design?"
Question:
Fact pg 56-58: In figure 2.5, it is important to note the dotted line looping back to "develop theory" after conducting case studies that did not fit the original design. You may need to reconsider one or more of the study's original theoretical propositions, and "redesign" should take place before proceeding further.
Question: What situations would make a single-case design appropriate? (pgs 47-50)
I would do the two groups. You will get more data that way--especially since you are adding the test scores as well for a MANOVA. You do want to select your participants and talk about that in your work. You also want to have as much validity and triangulation as possible too. If you have test scores of both groups of students in your data, you should have interviews with some of each group too. You want to get them to really talk, so I suggested the open-ended questions.
Keep working, reading, and asking questions, Dr. Hendrix
It depends how you write up your findings and themes. Look at Dr. Mitchell's dissertation that won the award. Her study is a multiple case study. Part of it is about her own identity like autoethnography and then she had a few (dis)abled participants. When she did her data analysis, you can see how that (multiple cases) adds more weight than just focusing singly on her own case.
I have seen more single case designs in psychology. What if you only have one person with a certain mental illness--a rare illness? At times such as those, single cases are informative. Look at Sybil. I think she had more documented personalities than any other person with her illness, so her case was unique and useful in that field.
Ok I am completely overwhelmed by this chapter. As I read this I am figuring out that I will need to do a multiple case study design. I am unsure how case study fits in with the plan that I have for my research project. I was planning on using my math benchmark scores from the past, previous, and future school year to try to document the effects that teaching math using CGI theory has on student achievement. At this point I am unsure about what types of questions I should ask my students or better yet exactly what I am looking for. I guess I am struggling with how this all fits together. Should I try another topic? HELP!!
No! Think about the research questions you are using for your quantitative part. Now, think about what you could glean from adding field notes and interviews about your topic. Could you add depth from student and/or teacher responses about their student achievement? Could you add more to your data if you interviewed district officials? If that will help, then create or play with your research questions, so they are for a case study too or so they go with a case study. Keep your topic! Keep thinking and questioning!
Ch. 2 Discussion Post Discuss the criteria for judging the quality of research designs--construct validity, internal validity, external validity, and reliability-- and discuss how you can satisfy these tests with your case study.
First off, one reason for making sure that you start out using a good quality research design is to ensure that others will take your case study seriously. A research design, according to Yin, can be thought of as a "blueprint" for your research and you must look at at least four things; what questions to study, what data are relevant, what data to collect, and how to analyze the results. For my case study; Construct validity: identifying correct operational measures for the concepts being studied. I will have to make sure my plan is precise and that I follow that plan throughout, without deviating from the original design purpose (meaning I had better make sure it isn't flawed from the beginning) Internal validity: seeking to establish a causal relationship, whereby certain conditions are believed to lead to other conditions. I will have to make sure that I don't make a lot of detrimental "inferences" with my conclusions. External Validity: defining the domain to which a study's findings can be generalized. I will have to be sure I don't follow a pattern that will lead me to make generalizations about grammar study, and that my study will conclude a more specified conclusion. Reliability: demonstrating that the operations of a study-such as the data collection procedures-can be repeated, with the same results. Basically, I need to make sure that I follow precise, recorded methods, so that anyone wanting to can follow my results and find the (basically) same conclusions.
I was reading Amanda's post and I saw that she had some questions in mind already...I need to get busy! I've got some in mind but I'd like to hone in and get them more specific. I want to make sure that I am not asking "leading" questions...
Good points here! Yes, make sure that your questions are not "leading" your participants. Try to get them to talk open with you. Focus on having open-ended questions.
First of all, I would like to say this book is much easier to read and understand than last semester's. It must be all that knowledge we gained!
ReplyDeleteJulie,
ReplyDeleteIt does get easier!!!!! Case study research is not as dense as some of the other ones too.
You are researchers now!
Take care,
Dr. Hendrix
This is a long chapter...i am still reading and will post soon. I am a bit confused on the four tests and so a bit of rereading is necessary. I'll post soon.
ReplyDeleteJulie,
ReplyDeleteI feel your pain this is a long chapter with a lot of detail and scientific "talk". This is what I have in my notes so far...
4 tests are commonly used to establish the quality of any empirical social research. These are construct validity, internal validity, external validity, and reliability.
Construct validity refers to the use of correct operational measures for the concept being studied. You can do this by using multiple sources of evidence, establish a chain if evidence, have key people informed in the area studied review you draft. The researcher need to avoid using subjective judgments when collecting data. The author also mentioned when using multiple sources of evidence to do it in a manner that encourages convergent lines of inquiry. I have no idea what that means. I'm thinking it would mean that you want to make the reader think the same question you are researching and not develop other questions that would lead to something not covered or in a different arena, something along that ,line? Not sure if anyone has any other ideas let me know.
Internal validity is mainly a concern in explanatory case studies or casual. One would look for pattern matching, explanation building, addressing rival explanations, and the use of logic models. This part addresses the inference part of a case study. An investigator will infer that something happened due to something that previously occurred. In order to prove that this inference is correct the researcher will need to need to back it up with interviews and documentary evidence. They will need to cover rival explanations in a way that proves their inference to be the correct one. Basically, is your theory "airtight"?
External validity deals with deciding if the findings are generalizable beyond the case at hand. It uses theory in single-case studies and logic in multiple-case studies. It talks about how the theory would need to be tested in several different situations to be generalizable. This one I understand the criticism but I don’t understand how to overcome the it, especially for us. We wouldn't be able to test and test and test our theories over the next year in order to prove them. I thought the author was very vague here and I will need way more info to determine how we will accomplish this one. I think that's one we should cover in more detail in class.
Last, reliability, refers to the ability of your case study producing the same findings if someone else followed the same procedure and conducted the same study all over again. The goal is to minimize the errors and biases in a case study. You can do this with super documentation. Make as many steps as operational as possible and to conduct research as if someone was always looking over your shoulder! I would think that doing a good job here would help with the external validity.
Now how will I do this???
Construct validity- I would think part of this would be covered in your literature review. Here you would show commonalities in previous research and your theory.
Internal validity- The inference part I will make sure and cover why other conclusions could not be possible based in the research and documentation I have collected and any previous research conducted that could rule out a different outcome.
External validity- I have no idea will need help here as I mentioned earlier.
Reliability- I think your documentation here will be the key. I will write each step like a recipe. This reminds me of when we have the kids write the steps in making a peanut butter and jelly sandwich and then try to follow their steps. If their steps are clear, detailed, and make sure not to leave anything out then you can end up with a real mess.
If I interview more than one student is that what a multiple-case study is?
ReplyDeleteOn page 30 In the second full paragraph the author tells about the proper way to word questions is to avoid being too vague. Am I understanding this correctly? You want to sound like you are favoring one way.... for example with my study I would ask something like....
Does using palm pilots in spelling engage you more than without using palms?
Is that narrow enough?
By the way. I agree girls.... I am also still reading. More to come soon.
ReplyDeleteWOW.. I hate this chapter. I think it is a little more advanced than the last one.
ReplyDeleteKristin did a good job of breaking it down... at least I think so.
The four tests Yin and other text books have identified are construct vadility, internal vadility, external validity, and reliability.
I like the chart on page 41, It really helps me break down the information. From the reading I have gathered (or tried to understand) the following:
Construct Validity: I think a way to state this in my own words is the "state the evidence" Researchers need to see what is already out there in terms of what they are studing before just jumping in.
Internal Validity: This is a way for the researcher to back up their research. The researcher must "prove" x lead to y (without some other letter being involoved).
External Validity: Making sure that the results are repeatable and not just a one time thing.
Reliability: Be sure to document what you are doing so the test could be repeated. Minimize errors and biases.
As I continue reading I am now at page 47... WOW way too much information for one chapter. I am overwhelmed.
I asked earlier about multiple case designs. I didn't know at that time there would be several pages on this topic. However, I am still confused. Someone please help!
Alright... Here it goes.
If some of you didn't know I plan to do my research on palm pilots in the classroom. I don't use them on spelling dictation tests, however another third grade teacher does. If I survive this book I plan to base my case study on this. Here is how I think my case study will be represented by the four design tests.
Construct validity: As of now I have not found much for my literature review. To cover this part of validity I will have to check my literature review.
Internal validity: I will have to prove that there are not other things that could have altered the dictation tests scores.
External validity: I am not sure how I will cover this since the case study will only be done once by me while at MWSU.
Reliabilty: I will have to document everything. I will be sure to keep my documents and interviews in a safe spot.
I hope this covers what I needed to have done. I am overwhelmed.
Would any of you like to meet at Mo West to actually fill out the rest of the IRB information? Call me up!
Question: What the heck is next???
Quote: "You need a plan." page 25 Yeah and I need a cocktail!
Fact: Document everything!
Dr. Hendrix,
ReplyDeleteI would like to make an appointment with the library to see if I can find more information on palm pilots used on spelling tests. Is there a certain person I should ask for? Also, If they can't find other information how do I go about documenting that there is not any additional literature out there? Does that make sense? If not call me!
Annie
Annie,
ReplyDeleteYes, it makes sense. Contact the library and ask if they have a research or reference librarian. If they do, ask to set up an appointment and see that person. If they do not, just go in and ask to speak with a librarian and tell the person what you have done so far on your research with your literature review and ask the person if he or she has any other suggestions for you.
If there is nothing else out there, then you will document everything that you found and how it is similar yet different than your topic, and you will make the argument that your research is filling a hole in the field.
If you need to see me too, please set up an appointment, and I will be more than happy to meet you!
Take care and keep reading and (re)reading Yin,
you will get it!
Finish your IRB as soon as possible,
Dr. Hendrix
Kristin,
ReplyDeleteGREAT job here! Multiple sources of evidence means that you will have a variety for your data collection process. You will have field notes and interviews; you may have someone in the school check your notes and interviews to add validity to it. You can even have your participants give you feed back on your notes and interviews to help with the traingulation and validity of your research too. Just remember you want to collect data in as many ways as possible, and you want to triangulate your data too.
Keep up the good work,
Dr. Hendrix
Kristin,
ReplyDeleteYou will minimize errors with your triangulation of your data. We will talk more about this later. See my earlier comments. Do you have someone who can verify your notes--someone who comes into your class perhaps?
If that does not make sense, let me know,
Dr. Hendrix
Everyone,
ReplyDeleteThis is a tougher chapter, but it is digestable. I agree. Break it up into small chunks and reread it if you need to do so.
Also, I am available by phone or in person if you need me,
Dr. Hendrix
Class,
ReplyDeleteThe following information is from Wikipedia, and I think that you might find it useful--especially the black swan part. I think that is a good image you can remember.
Generalizing from case studies
The case study is effective for generalizing using the type of test that Karl Popper called falsification, which forms part of critical reflexivity [3]. Falsification is one of the most rigorous tests to which a scientific proposition can be subjected: if just one observation does not fit with the proposition it is considered not valid generally and must therefore be either revised or rejected. Popper himself used the now famous example of, "All swans are white," and proposed that just one observation of a single black swan would falsify this proposition and in this way have general significance and stimulate further investigations and theory-building. The case study is well suited for identifying "black swans" because of its in-depth approach: what appears to be "white" often turns out on closer examination to be "black."
For instance, Galileo Galilei’s rejection of Aristotle’s law of gravity was based on a case study selected by information-oriented sampling and not random sampling. The rejection consisted primarily of a conceptual experiment and later on of a practical one. These experiments, with the benefit of hindsight, are self-evident. Nevertheless, Aristotle’s incorrect view of gravity dominated scientific inquiry for nearly two thousand years before it was falsified. In his experimental thinking, Galileo reasoned as follows: if two objects with the same weight are released from the same height at the same time, they will hit the ground simultaneously, having fallen at the same speed. If the two objects are then stuck together into one, this object will have double the weight and will according to the Aristotelian view therefore fall faster than the two individual objects. This conclusion seemed contradictory to Galileo. The only way to avoid the contradiction was to eliminate weight as a determinant factor for acceleration in free fall. Galileo’s experimentalism did not involve a large random sample of trials of objects falling from a wide range of randomly selected heights under varying wind conditions, and so on. Rather, it was a matter of a single experiment, that is, a case study.
Galileo’s view continued to be subjected to doubt, however, and the Aristotelian view was not finally rejected until half a century later, with the invention of the air pump. The air pump made it possible to conduct the ultimate experiment, known by every pupil, whereby a coin or a piece of lead inside a vacuum tube falls with the same speed as a feather. After this experiment, Aristotle’s view could be maintained no longer. What is especially worth noting, however, is that the matter was settled by an individual case due to the clever choice of the extremes of metal and feather. One might call it a critical case, for if Galileo’s thesis held for these materials, it could be expected to be valid for all or a large range of materials. Random and large samples were at no time part of the picture. However it was Galileo's view that was the subject of doubt as it was not reasonable enough to be Aristotelian view
By selecting cases strategically in this manner one may arrive at case studies that allow generalization.
For more on generalizing from case studies, see [2]
[edit] Assumptions
1. Cases selected based on dimensions of a theory (pattern-matching) or on diversity on a dependent phenomenon (explanation-building). 2. No generalization to a population beyond cases similar to those studied. 3. Conclusions should be phrased in terms of model elimination, not model validation. Numerous alternative theories may be consistent with data gathered from a case study. 4. Case study approaches have difficulty in terms of evaluation of low-probability causal paths in a model as any given case selected for study may fail to display such a path, even when it exists in the larger population of potential cases.
[edit] History of the case study
As a distinct approach to research, use of the case study originated only in the early 20th century. The Oxford English Dictionary traces the phrase case study or case-study back as far as 1934, after the establishment of the concept of a case history in medicine.[citation needed]
The use of case studies for the creation of new theory in social sciences has been further developed by the sociologists Barney Glaser and Anselm Strauss who presented their research method, Grounded theory, in 1967.
The popularity of case studies in testing hypotheses has developed only in recent decades. One of the areas in which case studies have been gaining popularity is education and in particular educational evaluation. [5]
Case studies have also been used as a teaching method and as part of professional development, especially in business and legal education. The problem-based learning (PBL) movement is such an example. When used in (non-business) education and professional development, case studies are often referred to as critical incidents.
[edit] History of business cases
When the Harvard Business School was started, the faculty quickly realized that there were no textbooks suitable to a graduate program in business. Their first solution to this problem was to interview leading practitioners of business and to write detailed accounts of what these managers were doing. Of course the professors could not present these cases as practices to be emulated because there were no criteria available for determining what would succeed and what would not succeed. So the professors instructed their students to read the cases and to come to class prepared to discuss the cases and to offer recommendations for appropriate courses of action.
Business case studies recount real life business situations that present to business executives a dilemma. The case puts the scenario into the context of the factors that influence it. Cases are generally written by business school faculty with particular learning objectives in mind and are refined in the classroom before publication. Relevant documentation or AVitems and a carefully crafted teaching note often accompany cases.
Annie,
ReplyDeleteI would like to meet to do the IRB stuff together. I need to get with Kelly this week and she's going to help me word my thing for Mr.Lawrence so that he will ok it and maybe we can use the info I gather to help her case for a new reading curriculum. After that then I'll know more on what to put on my consent forms and everything. You're way ahead of me so if you want go ahead if not then I will definitely meet with you and maybe Dr. Hendrix could join us! Let's meet this week anyways and work on our first journal.
Reading your take on how to use this with our case helped generate some ideas for me so thanks! Hey I also found a website that had short case studies on how different teachers used palms in the classroom. I'll give it to you I don't think these are the kinds of case studies we are working on but I didn't know if there were sources they used that would help you too. See if this link works
http://www.k12handhelds.com/casestudy.php
Whoever had the multiple case study question--please look back through Yin's pages 53-55.I think that will help you with understanding those more.
ReplyDeleteIf not, please let me know,
Dr. Hendrix
This is a test!
ReplyDeleteYes, Annie the chart is helpful but am I correct in thinking that this criteria is to be used to determine if we have a possible case study? It seems like a whole lot of work before we ever get started.
ReplyDeleteIt appears that the external validity is the most confusing. According to the example o pg 43, "if a study of neighborhood change focused on one neighborhood, are the results applicable to another neighborhood." So in my possible case study, I am considering using audiobooks/recorders to increase reading enjoyment for low performing students at Webster , a title I school. So with external validity I could generalize that this approach would benefit any low performing student at a title I school. Is this correct?
Laurel's post is as follows...
ReplyDelete"One prerequisite for allowing other investigators to repeat an earlier case study is the need to document the procedures followed in the earlier case. Without such documentation, you could not even repeat your own study" (pg.45)
In other words your study would not be reliable.
A question I have...from pg. 38It said a common topic of case studies is the evaluation of programs (state, federal or local) which in the research I purposed was a locally implemented reading program. But it further states that the theory needs to distinguish between the substance of the program and the process of implementation (e.g., how to install an effective program). And that is what has been bothering me. How can --or should I --try to measure our program results if there is question about how effectively it is being put into place?
It goes on to say that this mismatch can be avoided by paying closer attention to the substantive theory. OK- how do I do that? Do they mean concentrate on the substance of the program, and not how we are going about it? I think my research focus needs help!
I said...
"For your question here, you should study your program. Point out the problems with how it has been implemented as a limitation. You still need to study it though. For the theory, tie it to what the program is about more than the theory for implementation now. Later, you could keep working on your study and add that part in another case study.
No Worries!! You are doing well. I can tell by the questions you are asking!!!"
Julie,
ReplyDeleteBe careful about generalizing, but if someone else can repeat your study and you have good external validity, that will give your work more weight and can be generalized more easily. You would only be able to make some inferences about a school like yours and/or students like yours. Remember the black swan. Case studies help point out the black swan from the info. earlier.
Take care! Good points!!!
Dr. Hendrix
I swear technology hates me! I am glad that I am taking the class with Dr. Frick so she could help me fix my post. I want to start off by saying...Annie, you would be the one to mention drinks! Unlike Julie, I am struggling with construct validity. Am I missing something or is this just setting up our project? This book is much easier to read, but currently I feel as if I am on information overload.
ReplyDeleteI have some questions that I am thinking about asking for my interviews and I wanted the class opinion as to whether these are good questions or not. Please remember that I am researching the correlation between taking honors courses and ACT and AP scores. For the students the questions are:
1. Do you take any honors classes? Have you taken any in the past?
2. Did you take Honors by choice or was it a requirement?
3. How did/does your Honors class differ from non-honors courses?
4. Do you think that there are benefits to taking an honors course?
5. Do you feel that honors courses better prepare for your future? In terms of ACT? In terms of AP?
The teacher questions are as follows:
1. Do you teach any honors courses?
2. How do you think the one honors course requirement has affected the chemistry of your class? Does it create any challenges?
3. Do you think that students benefit from the requirement?
Are these questions anything similar to questions that I should be asking in an interview? Thanks!
Amanda,
ReplyDeleteYou want to ask open-ended questions in your interviews. A lot of these are going to have yes and no answers. You want to get as much information and data as possible though. The "how" question here is more open. How do you think students benefit from the requirement? What are the benefits for honors' courses? Try rewording these.
Good start,
Dr. Hendrix
Is it necessary to ask yes and no questions to get a certain sample, or should I ask these to see which kids I want to interview and have two sample groups? That was wordy...what I mean is do I want to have a group of kids that has taken honors and then a group that has not and compare their answers or stick to only kids that have taken honors courses? I am getting a little confused, not about the open-ended questions, I understand your suggestion, but now about my sample group. Is this the construct validity concept I was getting hung up on earlier?
ReplyDeleteI need a drink too, Annie, especially since I just got the book and read both of these chapters in one day. Dr. Hendrix, I am going to ask a stupid question that I'm sure is in the syllabus if I'd just go and look, but are we supposed to answer your discussion question and/or do the quote/question/fact here? I'm understanding that we should do both, but everyone seems to do something different, so I'm going to ask!
ReplyDeleteHere is my quote, fact, and question.
ReplyDeleteQuote pg 60-61: "...when you have the choice (and resource), multiple-case designs may be preferred over single-case designs. Even if you can do a a "two-case" case study, your chances of doing good case study will be better than using a single-case design. Single-case designs are vulnerable only because you will have put 'all your eggs in one basket.'"
If I conduct my research on how having access to computers for writing tests affects student scores on two of my classes instead of one, ill this be considered "multiple-case design?"
Question:
Fact pg 56-58: In figure 2.5, it is important to note the dotted line looping back to "develop theory" after conducting case studies that did not fit the original design. You may need to reconsider one or more of the study's original theoretical propositions, and "redesign" should take place before proceeding further.
Question: What situations would make a single-case design appropriate? (pgs 47-50)
Amanda,
ReplyDeleteI would do the two groups. You will get more data that way--especially since you are adding the test scores as well for a MANOVA. You do want to select your participants and talk about that in your work. You also want to have as much validity and triangulation as possible too. If you have test scores of both groups of students in your data, you should have interviews with some of each group too. You want to get them to really talk, so I suggested the open-ended questions.
Keep working, reading, and asking questions,
Dr. Hendrix
Jill,
ReplyDeleteIt depends how you write up your findings and themes. Look at Dr. Mitchell's dissertation that won the award. Her study is a multiple case study. Part of it is about her own identity like autoethnography and then she had a few (dis)abled participants. When she did her data analysis, you can see how that (multiple cases) adds more weight than just focusing singly on her own case.
I have seen more single case designs in psychology. What if you only have one person with a certain mental illness--a rare illness? At times such as those, single cases are informative. Look at Sybil. I think she had more documented personalities than any other person with her illness, so her case was unique and useful in that field.
Take care,
Dr. Hendrix
So I am going to adventure into the library this weekend and check for more info on my topic. If anyone wants to go as my date call me up!
ReplyDeleteOk I am completely overwhelmed by this chapter. As I read this I am figuring out that I will need to do a multiple case study design. I am unsure how case study fits in with the plan that I have for my research project. I was planning on using my math benchmark scores from the past, previous, and future school year to try to document the effects that teaching math using CGI theory has on student achievement. At this point I am unsure about what types of questions I should ask my students or better yet exactly what I am looking for. I guess I am struggling with how this all fits together. Should I try another topic? HELP!!
ReplyDeleteChristie,
ReplyDeleteNo! Think about the research questions you are using for your quantitative part. Now, think about what you could glean from adding field notes and interviews about your topic. Could you add depth from student and/or teacher responses about their student achievement? Could you add more to your data if you interviewed district officials? If that will help, then create or play with your research questions, so they are for a case study too or so they go with a case study. Keep your topic! Keep thinking and questioning!
Ch. 2 Discussion Post
ReplyDeleteDiscuss the criteria for judging the quality of research designs--construct validity, internal validity, external validity, and reliability-- and discuss how you can satisfy these tests with your case study.
First off, one reason for making sure that you start out using a good quality research design is to ensure that others will take your case study seriously. A research design, according to Yin, can be thought of as a "blueprint" for your research and you must look at at least four things; what questions to study, what data are relevant, what data to collect, and how to analyze the results.
For my case study; Construct validity: identifying correct operational measures for the concepts being studied. I will have to make sure my plan is precise and that I follow that plan throughout, without deviating from the original design purpose (meaning I had better make sure it isn't flawed from the beginning) Internal validity: seeking to establish a causal relationship, whereby certain conditions are believed to lead to other conditions. I will have to make sure that I don't make a lot of detrimental "inferences" with my conclusions. External Validity: defining the domain to which a study's findings can be generalized. I will have to be sure I don't follow a pattern that will lead me to make generalizations about grammar study, and that my study will conclude a more specified conclusion. Reliability: demonstrating that the operations of a study-such as the data collection procedures-can be repeated, with the same results. Basically, I need to make sure that I follow precise, recorded methods, so that anyone wanting to can follow my results and find the (basically) same conclusions.
I was reading Amanda's post and I saw that she had some questions in mind already...I need to get busy! I've got some in mind but I'd like to hone in and get them more specific. I want to make sure that I am not asking "leading" questions...
ReplyDeleteErin,
ReplyDeleteGood points here! Yes, make sure that your questions are not "leading" your participants. Try to get them to talk open with you. Focus on having open-ended questions.
Keep reading!
Your contents are more then sufficient for me.
ReplyDeleteppi claims calculator
The abysmal you dig into the content and communicate us the precise aggregation is appreciable.
ReplyDeleteppi claims calculator
thank you. visit my site, https://www.weddingphotoz.co.uk
ReplyDeleteNorth Indian Wedding Traditions
ReplyDelete