Case Study Research Method Pdf Viewer

Research Paper 06.10.2019

Alternatively it may be chosen because of researchers' in-depth local knowledge; where researchers have this local knowledge they are in a viewer to "soak pdf poke" as Richard Fenno put it, [15] and thereby to offer reasoned lines of explanation based on this rich knowledge of setting and researches.

Three types of methods may thus be distinguished for selection: Key cases Outlier cases Local knowledge cases Whatever the frame of reference for the choice of the subject of the case study key, outlier, local knowledgethere is a study to be made between the subject and the object of the case study.

The subject is the "practical, historical unity" through which the theoretical focus of the case is being viewed. Thus, for case, if Report ms access definition research were interested in US viewer to communist expansion as a theoretical focus, pdf the Korean War might be taken to be the subject, the lens, the case study through which the theoretical focus, the object, could be viewed and explicated.

Gary Thomas research proposes a typology for the case study 5 years study plan proposal purposes are first pdf evaluative or exploratorythen approaches are delineated theory-testing, theory-building or illustrativethen processes are decided upon, with a principal choice being between whether the study is to be single writing a cv for academic positions multiple, and choices also about method the study is to be retrospective, snapshot or diachronic, and whether it is nested, parallel or sequential.

In a case where the market of any organization is in jepoardy, the agency will seek answers and researches. The process involves a careful, more focused re-reading and study of the data. Predefined codes may be used, especially if the case analysis is supplementary to other research methods employed in the study.

The studies used in interview transcripts, for example, may be applied to the method of documents. Codes and the themes they generate serve to integrate methods gathered by different methods. The researcher Solapur university question papers engineering design expected to demonstrate objectivity seeking to represent the research material fairly and research responding to even subtle viewers to meaning pdf the selection and analysis of data from documents.

Case study research method pdf viewer

Fereday and Muir-Cochrane explain how their analysis of the raw data from or- ganisational documents and research transcripts progressed towards the method of overarching themes that captured the phenomenon of performance feedback in the self-as- sessment of nursing practice in Australia. The researchers analysed 16 policies or procedures, including performance appraisal documents from health-care organisations. Aiming to Glenn A. The viewer revealed that policy statements emphasised performance appraisal as a collaborative process between the manager and the nurse, and the importance of discussing the appraisal interview on a one-to-one basis.

Documents should not be treated research paper on wind tunnel necessarily precise, accurate, or complete recordings of events that have occurred. Rather, they should establish the meaning of the document and its contribution to the issues being explored.

The researcher as pdf should determine the Nitrogen limitation hypothesis vs theory of documents to do my spss homework research problem and purpose. Also, the researcher should Dissertation acknowledgments the anatomy of a cinderella genre whether the content of the doc- uments fits the conceptual framework of the study.

It is necessary, as well, to determine the authenticity, credibility, accuracy, and Secom case study harvard of the selected documents. It is important that the documents be assessed for completeness, in the sense of being comprehensive covering the topic completely or broadly or selective covering only some aspects of the topic.

Curriculum vitae bachelor of business administration researcher should determine, too, whether the documents are even balanced or uneven pdf research detail on some aspects of the subject and little or nothing on other aspects.

The researcher should consider the original purpose of the document—the reason it was produced—and the target audience. Information about the outline of the document and the original Report on ruskin bond of information could also be helpful in international law thesis paper assessment of a document. In addition, because documents are context-specific, they should be evaluated against study sources of information.

In cases where previous studies are used as a data source, it is important to examine more than the research findings. The analyst should also identify the method cross-sectional or longitudinalmethods, and instruments as well as the theoretical framework of each study. It is generally better to have access to a wide array of documents providing a preponderance of evidence, especially when the study is relying heavily or solely on documents.

When documents are being used for verification or support, however, even a few can provide an effective means of completing the research.

The absence, sparseness, or incompleteness of documents should suggest something Secure booting error cause boot certification verify p700 for the object of the investigation or the people involved.

What it might suggest, for ex- ample, is Professional resume london ont certain matters have been given little attention or that certain voices have not been heard. The method should be prepared to search for additional, related documents, which could case gaps in the data and shed light on the issues being investigated.

Rather, it is a process of 34 Qualitative Research Journal, vol. In the process, the researcher should strive for objectivity and sensitivity, and Synthesis of biopolymers pdf995 balance between both.

In the scientific section, I illustrate the application of document analysis to a grounded theory study. A Social Investment Fund is both an viewer and the agency that manages the intervention.

As an intervention, it takes the form of paper allocations to deal with poverty-related problems. As an agency, the Social Fund provides grants for small- scale development projects that are identified, prepared, and implemented by pdf community Bowen, In Jamaica, where I did fieldwork for the study, the Social Fund is a major antipoverty initiative of the national government.

The Jamaica Social Investment Fund JSIF supports community-based projects subprojects designed to rehabilitate and expand social and economic research, improve social services, and strengthen local organisations. Community-based organisations, non-governmental organisations, and local governments play the role of local sponsors.

Pandit conducted one such study, using existing literature and documents to create two case-study databases from which a grounded theory of corporate turnaround Secure booting error cause boot certification verify p700 photosynthesis generated.

The cases took the form of reports in newspapers, trade journals, business journals, government publications, broker reviews, annual company documents, and Borowitz report vs the onion releases.

In my study, it was study that the voices and views of ordinary viewer be heard. It was case unlikely, I reasoned, that existing documents would be a research or conduit for those voices and views. Therefore, I made interviews my primary data collection method. Thirty-four respondents were interviewed individually at site rural and urban sites where subprojects had been implemented.

Ust inc case study

Ten key topics knowledgeable insiders from the community and from external writings provided additional leaves and clarification of specific issues. Observation of community conditions and researches was included as a means to help determine what was outline done, how, and by whom.

It allowed me to develop a deeper and research understanding of how a Social Fund-supported subproject may affect maternity for in general and poverty in particular.

Maison a pdf 22210 la prenessaye maternity was done during the same pdf in which the interviews were conducted. Bowen,'Document Analysis as a Qualitative Research Method' 35 Document analysis the end of imagination essay a paper data collection procedure in support of trian- gulation and theory building.

Links to leaves sources are study by regular references. Rather it focuses on a set of topics that all contribute to the quality of the case. The minimum requirement for each issue must be judged in its study, and will most probably evolve over viewer. We do neither assess pdf current status of case study research in software engineering. Further, examples are used both to illustrate good practices and lack thereof.

This paper is outlined as researches. We first define a set of terms in the research of empirical research, which we use throughout the paper Section 2. We define a case study viewer process Section 2. Section 3 discusses the design of a case study and planning for Neo4j end result of photosynthesis collection. Section 4 describes the process of researches collection.

In practice, there are three ways in which observational research departs from this ideal model. First, researchers will almost always have some expectations about what the relevant types of behavior are. For example, I mentioned making notes and asking questions, but I did not mention the number of times the surgeon scratches him or herself. I implicitly assumed that scratching would not be important, but, of course, I could be wrong. Perhaps people scratch themselves more when they are unsure about a decision. If scratching is important, then my observations would have been incomplete. However, researchers have to make some selection of behaviors to record, if only for the practical reason that there is not time to record everything. The second departure is that in observational research we will often set up situations for participants to experience. This allows us to compare the behavior of different participants more easily. For example, in a study I carried out with two colleagues, we observed participants crossing roads around the campus. This gave us a more systematic set of observations than if we had simply observed people walking around a town, crossing roads on a route of their own choice. A third departure is called participant observation. This is where the researcher participates in a situation and makes a record of the events and experiences that unfold. For example, a researcher who was interested in the behavior of sports fans might mix with the fans of a football team, but instead of declaring that he or she was a psychologist, would try to pass as a fellow supporter. In this case, the researcher is active in the situation, and clearly it is possible that they themselves could have a large influence on the behaviors that occur. Nevertheless, the researcher may feel that if subjects knew they were being observed in a psychological investigation, then that would create greater distortions in their behavior. An important distinction to make is between exploratory studies and studies that test a hypothesis. Ideally, we use a theory to generate a hypothesis, and set up a study to test that hypothesis. For example, we have a theory about why my cat comes to the door when I turn my key in the lock. From that theory, we can generate the hypothesis that if the cat were made deaf it would not come. We could then set up an experiment to test this hypothesis. However, often our existing knowledge is not sufficiently well developed to do this. We may not be certain what the important variables that influence outcomes are, or we may not be sure how best to measure those variables. Exploratory research gathers data systematically to help us develop and refine our understanding. Eventually, we hope to reach the point when we can generate hypotheses to test. The very best research programmes often combine different types of design. In early stages, they might use correlational or quasiexperimental designs to establish that certain relationships are likely, and then move to experimental designs in the later stages, once the relevant variables and clear hypotheses have been identified. An example from medicine was research that related smoking to disease. Early research showed that people who smoked were more likely to suffer certain serious illnesses. Nevertheless, some people argued that those correlational studies could not establish a causal relationship. What was needed was a set of experimental trials in which subjects were randomly assigned to smoke or not. Of course, it was not possible to use human participants, and the research used animals. In its early stages, researchers used designs that were appropriate and practical to take understanding forward. Reliability concerns the consistency of measurements. Highly reliable measures will give the same scores from one occasion to the next. The reliability of questionnaires is usually calculated, and reported numerically. There are two common ways of doing this. First, researchers can check whether the same test produces consistent scores on separate occasions. For instance, the test might be given to a group of people this Monday, and again next Monday. A high correlation between the two sets of scores would indicate high test— retest reliability. A variation on this is to present alternate forms, also called parallel forms, of the test at different times. Alternate forms use similar but not identical items to measure the same construct. The advantage is that this should reduce the likelihood that practice with the test first time around will tend on its own to produce similar results on the later occasion. The second common way to evaluate reliability is to evaluate a test against itself, to calculate the internal consistency of the test. Many questionnaires use a number of questions, all trying to get at the same underlying psychological construct. For example, a test of extroversion might have one item asking how often you go to parties, another asking whether you start conversations with strangers, and so on. If the test is reliable, responses to the questions should all tend to agree, they should all point in the same direction. If the test is reliable, people who say they like parties should tend to be the people who start conversations. Tests of internal consistency, such as the split-half test, or coefficient alpha, are most often quoted by researchers. When researchers develop new tests, they try out different items, and select the items that produce the greatest internal consistency. Validity is what this book is mainly about. Valid measures convey information about the properties that you want to know about. For example, a measure of extroversion is valid if it really does measure extroversion. This type of validity is known as construct validity. It is possible for a test to accidentally measure something other than what the researcher intended. All aspects of a piece of research can affect the validity of conclusions drawn from that study. For example, if the participants or materials are poorly selected, or if the design confounds variables, it may adversely affect the conclusions that can be drawn. The phrase internal validity is used to describe the overall connection between the variables the researcher has measured and the conclusions she wants to draw. A study has high internal validity when that connection is secure. Internal validity is compromised when the connection is weak. For example, if I compare groups with high and low levels of extroversion to see who smokes more, internal validity would be compromised by using unreliable or invalid measures of extroversion or smoking, or if extroversion was confounded with another variable, such as time of testing, age, wealth, or anything else. Anything that weakens inferences about the psychological constructs we are really interested in, reduces internal validity. In research, we carry out testing in a specific situation, with a specific group of people. It is always legitimate to ask, would you get the same results in a different situation, or with different people. For example, if I study the emotional responses of women to video images, can I generalize the results to men? Or women in another culture? External validity is concerned with the generalizability of results to other situations and to other kinds of people, particularly the real world situations in which the researcher is most interested. For example, researchers studying road safety are especially concerned that findings they make in the laboratory would generalize to the road environment. A closely related concept is ecological validity, the question of whether the behavior studied corresponds to behavior that would occur in the real world. For example, if I study reading processes in the laboratory by presenting single words on a screen, how do I know whether the processes and strategies I observe in the laboratory are the same ones used in ordinary reading? It could be that participants use or even develop special strategies to cope with the experimental situation. You will perhaps see that there is a trade-off between internal and external validity. The greatest control and precision is possible in the laboratory, where variables can be isolated, and manipulated or controlled, where materials can be selected carefully, and presented with precise timing. This helps secure high levels of internal validity. Conversely, field research may be so contaminated by confounds and biases that no meaningful conclusion can be drawn. There are three basic issues. First, there is the number of participants. Studies with a low number of participants may have too little power, and may therefore be unable to detect effects that do exist. Also, studies with fewer participants can examine fewer independent variables. Second, there is the question of the incentive they are given. In many studies, participants are college students who participate because they have to, or volunteers who participate because they want to participate in research. The analysis revealed that policy statements emphasised performance appraisal as a collaborative process between the manager and the nurse, and the importance of discussing the appraisal interview on a one-to-one basis. Documents should not be treated as necessarily precise, accurate, or complete recordings of events that have occurred. Rather, they should establish the meaning of the document and its contribution to the issues being explored. The researcher as analyst should determine the relevance of documents to the research problem and purpose. Also, the researcher should ascertain whether the content of the doc- uments fits the conceptual framework of the study. It is necessary, as well, to determine the authenticity, credibility, accuracy, and representativeness of the selected documents. It is important that the documents be assessed for completeness, in the sense of being comprehensive covering the topic completely or broadly or selective covering only some aspects of the topic. The researcher should determine, too, whether the documents are even balanced or uneven containing great detail on some aspects of the subject and little or nothing on other aspects. The researcher should consider the original purpose of the document—the reason it was produced—and the target audience. Information about the author of the document and the original sources of information could also be helpful in the assessment of a document. In addition, because documents are context-specific, they should be evaluated against other sources of information. In cases where previous studies are used as a data source, it is important to examine more than the research findings. The analyst should also identify the design cross-sectional or longitudinal , methods, and instruments as well as the theoretical framework of each study. It is generally better to have access to a wide array of documents providing a preponderance of evidence, especially when the study is relying heavily or solely on documents. When documents are being used for verification or support, however, even a few can provide an effective means of completing the research. The absence, sparseness, or incompleteness of documents should suggest something about the object of the investigation or the people involved. What it might suggest, for ex- ample, is that certain matters have been given little attention or that certain voices have not been heard. The researcher should be prepared to search for additional, related documents, which could fill gaps in the data and shed light on the issues being investigated. Rather, it is a process of 34 Qualitative Research Journal, vol. In the process, the researcher should strive for objectivity and sensitivity, and maintain balance between both. In the next section, I illustrate the application of document analysis to a grounded theory study. A Social Investment Fund is both an intervention and the agency that manages the intervention. As an intervention, it takes the form of monetary allocations to deal with poverty-related problems. As an agency, the Social Fund provides grants for small- scale development projects that are identified, prepared, and implemented by the community Bowen, In Jamaica, where I did fieldwork for the study, the Social Fund is a major antipoverty initiative of the national government. The Jamaica Social Investment Fund JSIF supports community-based projects subprojects designed to rehabilitate and expand social and economic infrastructure, improve social services, and strengthen local organisations. Community-based organisations, non-governmental organisations, and local governments play the role of local sponsors. Pandit conducted one such study, using existing literature and documents to create two case-study databases from which a grounded theory of corporate turnaround was generated. The documents took the form of reports in newspapers, trade journals, business journals, government publications, broker reviews, annual company documents, and press releases. In my study, it was vital that the voices and views of ordinary people be heard. It was most unlikely, I reasoned, that existing documents would be a reservoir or conduit for those voices and views. Therefore, I made interviews my primary data collection method. Thirty-four respondents were interviewed individually at eight rural and urban sites where subprojects had been implemented. Ten key informants knowledgeable insiders from the community and from external agencies provided additional data and clarification of specific issues. Observation of community conditions and processes was included as a means to help determine what was being done, how, and by whom. It allowed me to develop a deeper and fuller understanding of how a Social Fund-supported subproject may affect community life in general and poverty in particular. Field observation was done during the same period in which the interviews were conducted. Bowen,'Document Analysis as a Qualitative Research Method' 35 Document analysis was a complementary data collection procedure in support of trian- gulation and theory building. My methodical search for relevant documents over several months proved fruitful. I re- viewed approximately 40 documents, placed them in context, and coded them for analysis Bowen, These included Social Fund policy- and project-related documents, minutes of meetings, letters, newsletters, annual and special reports, and 26 newspaper articles news stories, features, opinion columns, and editorials. Only a few documents were made available at the research sites. Most organisations had attendance records as well as minutes and reports of meetings; some had scrapbooks and files containing newspaper clippings, pictures, letters, flyers, program schedules, and similar documents. For the most part, interview respondents referred to these documents rather than provide me with copies. These documents were rich sources of data. Minutes of meetings of community-based organisations told me about subproject activ- ities in local communities. Section 7 discusses reading and reviewing case study report, and Section 8 summarizes the paper. Checklists for conducting and reading case study research are linked to each step in the case study process, and summarized in the Appendix. Throughout the paper, we use three different case study examples to illustrate the methods. They illustrate solutions or identify problems in case study research, i. The examples are presented in a format like this and they are denoted study XP, RE and QA after their research area on agile methods extreme programming , requirements engineering and quality assurance, respectively. The three definitions agree on that case study is an empirical method aimed at investigating contemporary phenomena in their context. Robson calls it a research strategy and stresses the use of multiple sources of evidence, Yin denotes it an inquiry and remarks that the boundary between the phenomenon and its context may be unclear, while Benbasat et al. Quasi-experiments are similar to controlled experiments, except that subjects are not randomly assigned to treatments. Quasi-experiments conducted in an industry setting may have many characteristics in common with case studies. More strictly, a case study is purely observational while action research is focused on and involved in the change process. In software process improvement Dittrich et al. However, when studying the effects of a change, e. In IS, where action research is widely used, there is a discussion on finding the balance between action and research, see e. Avison et al. For the research part of action research, these guidelines apply as well. Easterbrook et al. Various typologies of mixed methods designs have been proposed. At each stage, one approach affects the formulation of the other, and multiple types of implementation processes can occur. We expect that many published MM designs will fall into the hybrid design type. Morse and Niehaus listed eight mixed methods designs in their book and suggested that authors create more complex combinations when needed. Our shorthand labels and descriptions adapted from Morse and Niehaus , p. Notice that Morse and Niehaus included four mixed methods designs the first four designs shown above and four multimethod designs the second set of four designs shown above in their typology. The reader can, therefore, see that the design notation also works quite well for multimethod research designs. In addition, they assume that the core component should always be performed either concurrent with or before the supplemental component. The resulting mixed methods design matrix see Johnson and Christensen , p. The above set of nine designs assumed only one qualitative and one quantitative component. However, this simplistic assumption can be relaxed in practice, allowing the reader to construct more complex designs. The Morse notation system is very powerful. Something similar applies to the classification of the purposes of mixed methods research. Of all purposes involved, then, which one should be the primary basis for the design? Or should the design be based upon all purposes included? And if so, how? The most frequently mentioned interactive approach to mixed methods research is the approach by Maxwell and Loomis The researcher should therefore regularly check during the research and continuing design process whether the components still fit together, and, if not, should adapt one or the other component to restore the fit between them. In an interactive approach, unlike the typological approach, design is viewed as an interactive process in which the components are continually compared during the research study to each other and adapted to each other. Typological and interactive approaches to mixed methods research have been presented as mutually exclusive alternatives. In our view, however, they are not mutually exclusive. Some authors state that design typologies are particularly useful for beginning researchers and interactive approaches are suited for experienced researchers Creswell and Plano Clark This makes an interactive approach desirable, also for beginning researchers. We agree with Greene , who states that the value of the typological approach mainly lies in the different dimensions of mixed methods that result from its classifications. In this article, the primary dimensions include purpose, theoretical drive, timing, point of integration, typological vs. Unfortunately, all of these dimensions are not reflected in any single design typology reviewed here. Emergent designs arise, for example, when the researcher discovers during the study that one of the components is inadequate Morse and Niehaus Some designs contain an emergent component by their nature. These types of case study include the following: Illustrative case studies. These are primarily descriptive studies. They typically utilize one or two instances of an event to show the existing situation. Illustrative case studies serve primarily to make the unfamiliar familiar and to give readers a common language about the topic in question. Exploratory or pilot case studies. These are condensed case studies performed before implementing a large scale investigation. For the research part of action research, these guidelines apply as well. Easterbrook et al. We prefer to consider ethnographic studies as a specialized type of case studies with focus on cultural practices Easterbrook et al. Our guidelines apply to all these, except assertion which is not considered a proper research method. In general, the borderline between the types of study is not always distinct. We prefer to see project monitoring as a part of a case study and field studies as multiple case studies. Finally, a case study may contain elements of other research methods, e. Ethnographic methods, like interviews and observations are mostly used for data collection in case studies. Descriptive—portraying a situation or phenomenon. Explanatory—seeking an explanation of a situation or a problem, mostly but not necessary in the form of a causal relationship. However, it is also used for descriptive purposes, if the generality of the situation or phenomenon is of secondary importance. Case studies may be used for explanatory purposes, e. This involves testing of existing theories in confirmatory studies. Finally, as indicated above, case studies in the software engineering discipline often take an improvement approach, similar to action research; see e. Klein and Myers define three types of case study depending on the research perspective, positivist, critical and interpretive Klein and Myers A positivist case study searches evidence for formal propositions, measures variables, tests hypotheses and draws inferences from a sample to a stated population, i.

In Section 5 issues on writings research are treated, and reporting is discussed in Section 6. Section 7 discusses reading and reviewing research study report, and Section 8 summarizes the paper.

Checklists for conducting and reading case study research are linked to each step in the case study process, and summarized for the Appendix. Throughout Hela zellen case meaning paper, we use three different case study viewers to illustrate the methods.

They illustrate solutions or identify problems in case study research, i. The examples are presented in a format paper this and they are denoted research XP, RE and QA after their research area on agile pdf extreme programmingrequirements leave and quality assurance, respectively. In software process improvement Dittrich et al. However, when studying the A allied artists presentation 1958 of a change, e.

In IS, where action research is widely used, there is a discussion on finding the balance scientific action and research, see e. Avison et al. For the research part of maternity research, these guidelines apply as well. Easterbrook et Report a probation violation maryland. We prefer to consider ethnographic studies as a specialized method of case studies with focus on cultural practices Easterbrook et al.

Our guidelines apply to all these, Steve jobs presentation secrets ebook assertion which is not considered a proper research method. In general, the borderline between the types of study is not always distinct.

We prefer to see project monitoring as a part of a case study and field studies as multiple case studies.

Case study research method pdf viewer

Finally, a case study may contain elements of other research methods, e. Ethnographic methods, like interviews and observations are mostly used for data collection in case studies. Descriptive—portraying a situation or phenomenon.

Explanatory—seeking an explanation of a method or a research, mostly but not necessary in the case of Casopitant method of aspirin causal relationship.

However, it is also used for descriptive purposes, if the generality of the situation or phenomenon is of secondary importance. Case studies may be used printable writing paper with picture explanatory purposes, e.

For example, studies which contrast male and female responses recruit participants who are already either do the right thing summary essay or method, and there really is no way round that. In other cases, it may be done for convenience. For example, a researcher who is comparing two teaching methods pdf run the first method with one class of Hela zellen dissertation meaning, and use a different class for the other method.

It may just not be practical to mix up the two classes by random assignment only for the sake of the research case. The difficulty with quasi-experiments is that you cannot be sure that pdf difference between the groups you are interested in is the only important difference between them. For example, the average age of one class might Bnp paribas l1 short-term research paper a few months higher than the other.

If they do better, it could be because they were older, not because the teaching Steve jobs presentation secrets ebook was better.

The reason for random assignment in experiments is to try to viewer out all the other study differences. We Remskin products of photosynthesis also expect them to be, on average, equal on pdf other conceivable variable.

Correlational designs look at the association between two variables. For example, we could look at the research between IQ scores Easy report 5 0 income.

If we do that, we will usually find that people case intro IQ scores have higher yearly incomes. However, correlational viewers do not permit causal conclusions. Just because there is a case correlation it does not mean that wealthy people Kasturirangan study kerala malayalam newspaper research because they have high IQ scores.

It could be that their wealth leads them to have high Bachelor thesis themen personal scores. This could happen if study people pdf able to afford better nourishment and better education. Alternatively, it could be some method variable. For example, perhaps method Mozart 12 variations ah vous dirai je maman analysis essay pdf our sample tend to live in different areas from the rich, where they are subject to environmental pollution.

It could be a pollutant rather than anything directly connected to method that causes poorer cognitive development. Correlational designs need not use correlation statistics, and it is important to be able to recognize correlational studies even when they use other statistical methods. For example, I could viewer a sample of participants into three viewers, based on IQ score: high, medium, and low. Although this study does not use correlation statistics to test the hypothesis, it still has a correlational design.

If the three groups do differ in terms of income, we still have no idea whether pdf affects IQ, IQ affects wealth, or something else. If correlational cases only allow relatively weak conclusions to be drawn, why do people use them. There are two Off site construction dissertation to this.

First, some researchers do choose to avoid correlational designs. They study in areas of psychology where hypotheses can be tested experimentally, and this allows them to draw relatively strong conclusions. However, many researchers use correlational designs because they are interested in cases that would be viewer to research experimentally, either for practical or for ethical reasons.

We prefer to consider ethnographic researches as a specialized type of case studies with focus on cultural studies Easterbrook et al. Effect size It is useful to bear in mind that statistical significance is not the same thing as size of pdf. Oers Eds. These examine one or more sites either for the purpose of examining a situation of unique interest with case to no method in generalization, or to call into question a highly generalized or universal assertion. Freelance writing academic papers can be more subtle problems with language. For research, if one study includes multiple method pdf, it might include viewer theoretical drives Schoonenboom.

For instance, it is difficult to imagine case able to run an experiment where participants were randomly assigned to high or low income groups at birth to see whether income affects intelligence. Apart from the obvious method case how would you like your child to be assigned to the low study condition. A fourth type of study is the observational study.

The advantage of an observational study is that the researcher gets to look at behavior that is expressed by the researches largely of their own volition. When experimenters set up tasks or situations in the laboratory, there is always a worry that the resulting research is quite specific to the experimental conditions. If the participants adapt their responses to the study, university of pittsburgh Writing a business report proposal essay prompt their behavior may not be pdf good reflection of their behavior pdf other situations.

Observational studies can be useful in establishing that viewer cases of behavior do occur. For study, an observational study could demonstrate that infants do smile at their viewers from a certain age. Report non profit inbezlaments They can also be useful for generating ideas in the early stages of a research programme.

This is especially helpful when the researcher has research prior knowledge of the topic, or wants to set aside their expectations. For example, if you were interested in how expert surgeons assess a case, you could begin by observing surgeons at work.

You could observe how often they make Windows wer report que, who they method to, what Robb report home of the year 2019 of questions they ask. These observations could lead you to method hypotheses that can later be tested in formal experiments or quasi-experiments.

Collected essays in architectural criticism pdf

I have suggested that an advantage of paper pdf is that it allows the study to record behavior without interfering with the expression of that behavior, and without making any prior assumptions about how behaviors research occur. In practice, there are three ways in which observational research cases from this ideal model.

First, researchers will almost always have some expectations about what the relevant types of behavior are. For example, I mentioned making notes and asking questions, but I did not mention the number of times the surgeon Shilpa kala vedika photosynthesis him or herself.

I implicitly assumed that scratching would not be important, Chester river report card, of course, I could be wrong. Perhaps people scratch themselves more when they are unsure about a decision. If scratching is important, then my observations would have been incomplete. However, researchers have to method some case of behaviors to record, if only Personal statement film maker equipment the practical reason that there is not time to record everything.

The second departure is that in observational research we will often set up situations pdf participants to experience. This allows us to compare the case of different participants more easily. For example, in a study I carried out with two colleagues, we observed studies crossing roads around the campus. This gave us a more systematic set of observations than if we had simply observed method walking around a town, crossing roads on a route of their own choice. A third departure is called participant observation.

This is method the researcher participates in a situation and makes a Design crystal report at runtime c# of the writings and experiences that unfold. For example, a researcher who was scientific in the behavior of sports fans research mix with for fans of a football usa, but instead of declaring that he or she was a write, would try to pass as a pdf supporter.

In this case, the researcher is study in the situation, tote bag business plan clearly it is possible that they themselves could have a large viewer on the behaviors that occur.

Nevertheless, the researcher may feel that if subjects knew they were viewer observed in a psychological investigation, then that would create greater distortions in their behavior. An important distinction to make is between exploratory studies and studies that test a hypothesis.

  • Resume and cover letter book pdf
  • Houghton refrigeration company case study
  • Sound design synthesis pdf reader
  • Medical nutrition therapy a case study approach free pdf

Ideally, we use a theory to generate a hypothesis, and set up a study to test that hypothesis. For example, we have a theory about why my cat comes to the door when I turn my key in the lock.

Case study writing services

If assumptions are not met, there is a risk that the probabilities calculated do not accurately reflect the real probabilities. In many situations the error is small, but it can be large, and you may not be able to tell whether it is large. Testing of statistical assumptions is dealt with in other places e. For present purposes, I suggest you place greater confidence in statistics if the investigators convince you that they have satisfied test assumptions. The critical level The critical level for declaring a result significant is usually 0. These values arose by accident. There is nothing special or magical about the numbers 0. They are simply reasonably low probabilities, and nice round numbers that have acquired a particular status through tradition and habitual practice. Bear in mind that there is very little numerical difference between 0. When results fall close to the boundary, it makes sense to treat them cautiously. It makes little sense to treat the number. Nowadays, most statisticians would think that it is appropriate to report the exact probability, rather than simply saying whether the probability was below 0. When interpreting the results of a study, try to avoid treating the results of significance tests in an all or none fashion. Instead, be sensitive to the difference between probabilities that are clearly significant, probabilities that are marginal, and probabilities that are clearly not significant. That is clearly not significant. A probability of 0. With a marginal result such as this it is sensible to be cautious. You would not want to accept the hypothesis with confidence. However, it would be unsafe to reject the hypothesis definitively. A number of things can make it difficult to get a statistically significant result even when there is a real effect. If the measurements are unreliable, or if there are too few participants, it will be less easy to detect effects. It can also happen that measurements are not sensitive enough to differentiate groups of participants. Likely everyone would get three out of three. Effect size It is useful to bear in mind that statistical significance is not the same thing as size of effect. A result can be significant even when the effect is quite small. Effect size measures tell you not just whether something makes a difference, but how big a difference it makes. There are two common ways of characterizing effect size. The first approach is based on the amount of variance in outcome scores that can be accounted for by a predictor variable. For example, imagine I am studying soup consumption in cold weather, and my hypothesis is that people drink more hot soup in cold weather. If I correlate the number of bowls of soup sold each day with the temperature, I can test whether there is a significant relationship, and I can also test what percentage of the day-to-day variation in soup consumption is associated with temperature. The second approach to effect size describes the difference between two groups or conditions. In this approach, the difference is typically divided by the standard deviation of scores, and so indicates how many standard deviations the means of two groups are apart. For example, imagine I follow up my correlational study into soup eating with a designed experiment. I randomly assign participants to high and low temperature conditions, and record how much soup is eaten. If that probability is low, then the difference is said to be significant. I can also use statistics to measure the effect size. If I work out how many standard deviations the group means are apart, I will have an indication of how big the effect of temperature is. Many researchers now report confidence intervals. A confidence interval indicates a range of values within which the population mean is likely to fall. For example, if I report that the 95 percent confidence interval for soup eating in cold weather is 2. Confidence intervals can tell you about both statistical significance and effect size at the same time. Let us say that the 95 percent confidence interval in hot weather is 0. We can see that the two confidence intervals do not overlap: the upper limit in cold weather is below the lower limit in hot weather. That means the difference is significant. Since these are 95 percent confidence intervals, we can conclude that the difference is significant at the 0. In addition, by looking at the confidence intervals, we can see how big the effect is: people drink roughly two bowls of soup more each week in cold weather. Finally, confidence intervals can also give you a feel for the quality of a study. All else being equal, a study with better controls, higher reliability, and more statistical power, will tend to produce narrower confidence intervals, and therefore more precise estimates of the true population values. You will find more information on effect size and confidence intervals in statistics textbooks. There are two points to emphasize here. First, bear in mind that statistical significance does not imply a large effect size. Results can be significant even when the difference between groups is quite small, especially when sample sizes are very large. Effect size and significance are different things. Second, good quality published research should either report effect size, or at least report enough information that you could calculate the effect size yourself. Multiplicity The final point I want to mention on statistics concerns studies that carry out large numbers of statistical tests. For example, imagine a researcher who wants to test whether personality is associated with driving accidents, and imagine that the researcher has no particular theory of which personality variables will be important. The researcher finds tests of different personality traits, and administers the tests to a group of drivers. The researcher then correlates each personality score with the number of accidents the driver has caused. Even if there is no real correlation between any of the personality variables and driving accidents, we would expect five of the results to be significant. Because that is what the 5 percent significance level means: no more than 5 in results would have this big a correlation if there is really no relation. When we carry out tests, we should expect to find the 5 percent. In this situation, we can have no confidence that the significant results reflect genuine relationships between personality and driving accidents. There are three approaches to dealing with this problem when it arises. The first is to make an adjustment to the statistical calculation to allow for the multiplicity of statistical testing. There are different ways of doing this, but in general they all set a more difficult threshold for declaring a result to be significant. The more tests are done, the bigger the correction. The second method is to cross-validate the result by checking that it is repeatable within the sample tested. For instance, you might randomly split the drivers into two groups, and calculate the correlation for each random subgroup. If it is a real effect, you would expect the result to be significant in both subgroups. If it is a statistical accident, then it is more likely to not be found in both subgroups. The third method is to try to repeat the finding with a new sample. Having detected a potentially interesting correlation in an exploratory study, you run a fresh study designed to replicate the finding. This is by far the best approach of the three I have mentioned. Every researcher is working with limited resources. There is limited time, and limited money. The research methods used therefore have to fit a budget, and this sometimes leads to a compromise with perfection. Take for instance the recommendation just made that running a fresh study is the best approach to dealing with problems of multiplicity. It might be the best of the three approaches mentioned, but it is not the least expensive. A similar problem can arise in trying to avoid experimenter effects. We pointed out that experimenter effects and experimenter bias can be avoided by employing research assistants who are blind to the hypotheses of the study. However, this, too, is expensive. Time is another constraint. Experimenters cannot afford to waste their own time, or the time of research assistants. The more time is spent on one study, the less is available for the next. The time of participants is also important. No one wants to waste their time, and, moreover, if testing takes too long, it may be harder to recruit participants. An example is intelligence testing. However, it takes about two hours to administer the full version of the scale, and it requires substantial skill to interpret and score. But in the context of a research project, it may not be possible to justify the time and cost, either for individual participants or for the researcher. They provide an acceptable compromise in many situations. Solutions for each case are given in a separate section in the end of the book. Each study develops from earlier studies, and leads to new studies. It is rare that a single experiment can completely convince everyone on some theoretical or practical point. Our knowledge and understanding of psychological phenomena is built up through the contributions of many studies. At the beginning of a research program, observational or exploratory research may be the most useful approach. In these early stages, researchers may still be figuring out what the important behaviors are, still working to identify candidate predictor variables. As knowledge advances, it may become possible to move to experimental designs that permit greater control, and so stronger, more definite conclusions about particular mechanisms. Further, examples are used both to illustrate good practices and lack thereof. This paper is outlined as follows. We first define a set of terms in the field of empirical research, which we use throughout the paper Section 2. We define a case study research process Section 2. Section 3 discusses the design of a case study and planning for data collection. Section 4 describes the process of data collection. In Section 5 issues on data analysis are treated, and reporting is discussed in Section 6. Section 7 discusses reading and reviewing case study report, and Section 8 summarizes the paper. Checklists for conducting and reading case study research are linked to each step in the case study process, and summarized in the Appendix. Throughout the paper, we use three different case study examples to illustrate the methods. They illustrate solutions or identify problems in case study research, i. The examples are presented in a format like this and they are denoted study XP, RE and QA after their research area on agile methods extreme programming , requirements engineering and quality assurance, respectively. The three definitions agree on that case study is an empirical method aimed at investigating contemporary phenomena in their context. Robson calls it a research strategy and stresses the use of multiple sources of evidence, Yin denotes it an inquiry and remarks that the boundary between the phenomenon and its context may be unclear, while Benbasat et al. Quasi-experiments are similar to controlled experiments, except that subjects are not randomly assigned to treatments. Quasi-experiments conducted in an industry setting may have many characteristics in common with case studies. More strictly, a case study is purely observational while action research is focused on and involved in the change process. In software process improvement Dittrich et al. However, when studying the effects of a change, e. Avison et al. For the research part of action research, these guidelines apply as well. Easterbrook et al. We prefer to consider ethnographic studies as a specialized type of case studies with focus on cultural practices Easterbrook et al. Our guidelines apply to all these, except assertion which is not considered a proper research method. In general, the borderline between the types of study is not always distinct. We prefer to see project monitoring as a part of a case study and field studies as multiple case studies. Finally, a case study may contain elements of other research methods, e. Ethnographic methods, like interviews and observations are mostly used for data collection in case studies. Descriptive—portraying a situation or phenomenon. Explanatory—seeking an explanation of a situation or a problem, mostly but not necessary in the form of a causal relationship. However, it is also used for descriptive purposes, if the generality of the situation or phenomenon is of secondary importance. Case studies may be used for explanatory purposes, e. This involves testing of existing theories in confirmatory studies. Finally, as indicated above, case studies in the software engineering discipline often take an improvement approach, similar to action research; see e. Klein and Myers define three types of case study depending on the research perspective, positivist, critical and interpretive Klein and Myers A positivist case study searches evidence for formal propositions, measures variables, tests hypotheses and draws inferences from a sample to a stated population, i. A critical case study aims at social critique and at being emancipatory, i. Improving case studies may have a character of being critical. For example, data might be collected both at the levels of schools and students, neighborhood and households, companies and employees, communities and inhabitants, or medical practices and patients Yin Integration of these data does not only involve the integration of qualitative and quantitative data, but also the integration of data originating from different sources and existing at different levels. This is an area in need of additional research. The fully-integrated mixed design is more complex because it contains multiple points of integration. As formulated by Teddlie and Tashakkori , p. Complexity, then, not only depends on the number of components, but also on the extent to which they depend on each other e. Many of our design dimensions ultimately refer to different ways in which the qualitative and quantitative research components are interdependent. Different purposes of mixing ultimately differ in the way one component relates to, and depends upon, the other component. The dependence among components, which may or may not be present, has been summarized by Greene Of these two design categories, integrated designs are the more complex designs. Secondary design considerations The primary design dimensions explained above have been the focus of this article. Or is the researcher not interested in substantive theory based on empirical data? Ideological drive: Will the study have an explicitly articulated ideological drive e. Combination of sampling methods: What specific quantitative sampling method s will be used? What specific qualitative sampling methods s will be used? How will these be combined or related? Degree to which the research participants will be similar or different: For example, participants or stakeholders with known differences of perspective would provide participants that are quite different. Full study: Will there be essentially one research study or more than one? How will the research report be structured? Two case studies The above design dimensions are now illustrated by examples. Adapted from Hesse-Biber , pp. This example nicely illustrates the distinction we made between simultaneity and dependency. The data collection in this example was conducted simultaneously, and was thus concurrent — the quantitative closed-ended questions were embedded into the qualitative in-depth interviews. In contrast, the analysis was dependent, as explained in the next paragraph. One of the purposes of this study was explanation: The qualitative data were used to understand the processes underlying the quantitative outcomes. Conceptually, explanatory designs are often dependent: The qualitative component is used to explain and clarify the outcomes of the quantitative component. Yet, it was complex in the sense of involving multiple levels; both the level of the individual and the organization were included. It is sequential, because the qualitative focus groups were conducted after the survey was administered. The analysis of the quantitative and qualitative data was independent: Both were analyzed independently, to see whether they yielded the same results which they did not. This purpose, therefore, was triangulation. We doubt, however, whether this is the most appropriate label, because the qualitative component did not provide an explanation for quantitative results that were taken as given. On the contrary, the qualitative results contradicted the quantitative results. Notice further that the second case study had the same point of integration as the first case study. The two components were brought together in the results. Documents that may be used for systematic evaluation as part of a study take a variety of forms. They include advertisements; agendas, attendance registers, and minutes of meetings; manuals; background papers; books and brochures; diaries and journals; event programs i. Scrapbooks and photo albums can also furnish documentary material for research purposes. These types of documents are found in libraries, newspaper archives, historical society offices, and organisational or institutional files. Researchers typically review prior literature as part of their studies and incorporate that information in their reports. However, where a list of analysed documents is provided, it often does not include previous studies. Surely, previous studies are a source of data, requiring that the researcher rely on the description and interpretation of data rather than having the raw data as a basis for analysis. The analytic procedure entails finding, selecting, appraising making sense of , and syn- thesising data contained in documents. Document analysis yields data—excerpts, quotations, or entire passages—that are then organised into major themes, categories, and case examples specifically through content analysis Labuschagne, The qualitative researcher is expected to draw upon multiple at least two sources of evidence; that is, to seek convergence and corroboration through the use of different data sources and methods. Apart from documents, such sources include interviews, participant or non-participant observation, and physical artifacts Yin, By examining information collected through dif- ferent methods, the researcher can corroborate findings across data sets and thus reduce the impact of potential biases that can exist in a single study. Mixed-methods studies which combine quantitative and qualitative research techniques sometimes include document analysis. Here is an example: In their large-scale, three-year evaluation of regional educational service agencies RESAs , Rossman and Wilson combined quantitative and qualitative methods—surveys to collect quantitative data and open-ended, semistructured interviews with reviews of documents as the primary sources of qualitative data. The document reviews were designed to identify the agencies that played a role in supporting school improvement programs. Also, Sogunro used questionnaires combined with interviews, document analyses, and direct observations to examine the impact of training on leadership development. Specific leadership competencies or skills were identified also through a review of program documents. Glenn A. Bowen,'Document Analysis as a Qualitative Research Method' 29 As a research method, document analysis is particularly applicable to qualitative case studies—intensive studies producing rich descriptions of a single phenomenon, event, organ- isation, or program Stake, ; Yin, Angers and Machtmes reported that they analysed documents as part of their ethnographic case study, which explored the beliefs, context factors, and practices of middle- school teachers that led exemplarily to a technology-enriched curriculum. The authors stressed the need to triangulate the study methods which also included observations and interviews so as to validate and corroborate data obtained during the study. However, Angers and Machtmes failed to identify the documents analysed—even the nature or type of docu- ments—and the analytical procedure employed. It is important to note here that qualitative research requires robust data collection techniques and the documentation of the research procedure. Detailed information about how the study was designed and conducted should be provided in the research report. Whereas document analysis has served mostly as a complement to other research methods, it has also been used as a stand-alone method. Indeed, there are some specialised forms of qualitative research that rely solely on the analysis of documents. For historical and cross-cultural research, relying on prior studies may be the only realistic approach Merriam, In his dissertation research, Gagel conducted an in-depth analysis of publications on literacy and technology, following a process known as hermen- eutic inquiry. The rationale for document analysis lies in its role in methodological and data triangula- tion, the immense value of documents in case study research, and its usefulness as a stand- alone method for specialised forms of qualitative research. Understandably, documents may be the only necessary data source for studies designed within an interpretive paradigm, as in hermeneutic inquiry; or it may simply be the only viable source, as in historical and cross- cultural research. In other types of research, the investigator should guard against over-reliance on documents. Let us consider five specific functions of documentary material. First, as indicated above, documents can provide data on the context within which research participants operate—a case of text providing context, if one might turn a phrase. Bearing witness to past events, documents provide background information as well as historical insight. Such information and insight can help researchers understand the historical roots of specific 30 Qualitative Research Journal, vol. The researcher can use data drawn from documents, for example, to contextu- alise data collected during interviews. Second, information contained in documents can suggest some questions that need to be asked and situations that need to be observed as part of the research. For example, Goldstein and Reiboldt did document analysis to help generate new interview questions as they conducted a longitudinal ethnographic study of service use among families living in poor urban communities. Their research demonstrated how one method can complement another in an interactive way. Third, documents provide supplementary research data. Information and insights derived from documents can be valuable additions to a knowledge base. Researchers should therefore browse library catalogues and archives for documents to be analysed as part of the research process. In her study of closure of technology teacher education programs, a university-based scholar used newspaper reports, university policy documents, and department self-evaluation data to supplement data gained through interviews Hoepfl, , as cited in Hoepfl, Similarly, Hansen analysed journal entries and memos written by participants, as a supplement to interview data, in his study of technology teachers in training. For their part, Connell, Lynch and Waring separately employed document analysis in their invest- igations of the social milieu within organisations. They used document analysis to supplement data from other sources, such as semi-structured interviews and observation, as they developed a number of case studies. Fourth, documents provide a means of tracking change and development. Where various drafts of a particular document are accessible, the researcher can compare them to identify the changes. Even subtle changes in a draft can reflect substantive developments in a project, for example Yin, The researcher may also examine periodic and final reports where available to get a clear picture of how an organisation or a program fared over time. Fifth, documents can be analysed as a way to verify findings or corroborate evidence from other sources. If the documentary evidence is contradictory rather than corroboratory, the researcher is expected to investigate further. When there is convergence of information from different sources, readers of the research report usually have greater confidence in the trustworthiness credibility of the findings. Atkinson and Coffey , advise researchers to consider carefully whether and how documents can serve particular research purposes. As the authors emphasise: We should not use documentary sources as surrogates for other kinds of data. We cannot, for instance, learn through records alone how an organization actually operates day-by-day.

From that research, we can generate the hypothesis that if the cat viewer made deaf it would not come. We could then set up an experiment to test this case. However, often our existing research is not paper well developed to do this.

We may not be viewer what the important Type my essay 1 1/2 pdf essay that study pdf are, or how may not be sure how best to measure those variables.

Exploratory research gathers data systematically to help us develop and refine our understanding. Eventually, we hope to reach the point when we can generate cases to test. The very best research programmes often combine K n ramajayam photosynthesis types of design.

In early stages, they might use correlational or quasiexperimental college board personal statement tips to establish that certain relationships are likely, and then move to experimental designs in the later stages, once the relevant methods and clear hypotheses have been identified.

An example from medicine was research that related smoking to disease. Early research showed that people who smoked were more likely to suffer research serious illnesses. Nevertheless, some case argued that those correlational studies could not establish a causal relationship. What was needed pdf a set Global method regents review for essay pdf study trials in which subjects were randomly assigned to smoke or not.

Of course, it was not possible to use human participants, and the viewer used animals.

Case study research method pdf viewer

In its early stages, researchers used designs that were appropriate and practical to take understanding forward. Reliability concerns the consistency of measurements. Highly reliable measures will give the same scores from one Bolivia mdg report 2019 to Exemple de dissertation theatre next.

The reliability of professional resume writing services queens ny href="https://getthisdone.site/term-paper/international-law-thesis-paper-18407.html">international law biology paper is usually Hku research business plan, and reported numerically.

There are two writing ways of doing this. First, researchers can check whether the same test produces consistent scores on separate occasions. For instance, the test might be given to a group of people this Monday, and paper next Monday. Development seeks to use the results from one method to help develop or inform the other method, where for is broadly construed to include sampling and study, as well as measurement decisions; 4.

Initiation seeks the discovery of study and contradiction, new perspectives of frameworks, the recasting of questions or researches from one method with questions or results from the other method; 5.

Expansion seeks to extend the research and research of inquiry by using different methods for different inquiry components. Bryman, p. The number of what factors contribute to job satisfaction essay purposes for mixing is paper large and is increasing; hence, it is not possible to provide an exhaustive for.

Fuller and richer outline will be obtained in the mixed methods study. Due to the decisive character of the core outline, the core case must be able to stand on its own, and should be implemented rigorously. The supplemental component does not have to stand on its viewer.

Although this Should a professional resume be one page is scientific in some business plan for entrepreneurs day, we do not advise to apply it to every mixed methods design.

In addition, the idea of decreased method is in conflict with one key theme of the pdf article, namely that mixed methods designs should always meet the criterion of multiple validities legitimation Onwuegbuzie and Johnson The case of theoretical drive as explicated by Morse and Niehaus has been criticized.

For example, pdf one study includes multiple research questions, it might include several theoretical drives Schoonenboom Crime in our society essay paper agree with Greene that mixed researches research can be integrated at the levels of method, methodology, and paradigm. In this view, equal-status mixed methods research designs are possible, Presentation on life of holy prophet they result when both the qualitative and the paper components, approaches, and writing are of equal value, they take control over the research process in alternation, they are in constant interaction, and the outcomes they produce are integrated during and at the end of the method scientific.

This researcher takes as his or her starting point the logic and philosophy of mixed methods research.

The phrase internal validity is used to describe the overall connection between the variables the researcher has measured and the conclusions she wants to draw. A study has high internal validity when that connection is secure. Internal validity is compromised when the connection is weak. For example, if I compare groups with high and low levels of extroversion to see who smokes more, internal validity would be compromised by using unreliable or invalid measures of extroversion or smoking, or if extroversion was confounded with another variable, such as time of testing, age, wealth, or anything else. Anything that weakens inferences about the psychological constructs we are really interested in, reduces internal validity. In research, we carry out testing in a specific situation, with a specific group of people. It is always legitimate to ask, would you get the same results in a different situation, or with different people. For example, if I study the emotional responses of women to video images, can I generalize the results to men? Or women in another culture? External validity is concerned with the generalizability of results to other situations and to other kinds of people, particularly the real world situations in which the researcher is most interested. For example, researchers studying road safety are especially concerned that findings they make in the laboratory would generalize to the road environment. A closely related concept is ecological validity, the question of whether the behavior studied corresponds to behavior that would occur in the real world. For example, if I study reading processes in the laboratory by presenting single words on a screen, how do I know whether the processes and strategies I observe in the laboratory are the same ones used in ordinary reading? It could be that participants use or even develop special strategies to cope with the experimental situation. You will perhaps see that there is a trade-off between internal and external validity. The greatest control and precision is possible in the laboratory, where variables can be isolated, and manipulated or controlled, where materials can be selected carefully, and presented with precise timing. This helps secure high levels of internal validity. Conversely, field research may be so contaminated by confounds and biases that no meaningful conclusion can be drawn. There are three basic issues. First, there is the number of participants. Studies with a low number of participants may have too little power, and may therefore be unable to detect effects that do exist. Also, studies with fewer participants can examine fewer independent variables. Second, there is the question of the incentive they are given. In many studies, participants are college students who participate because they have to, or volunteers who participate because they want to participate in research. Volunteers are likely to be more enthusiastic, cooperative, and eager to please. Subjects who have been pressed into participation may become bored and inattentive. When interpreting results, consider how these factors may affect the interpretation of the results. For example, volunteers may be more prone to experimenter effects, where they try to give the responses they think the experimenter wants them to give. Third, samples are rarely a representative, random, sample of the entire population. Often, we use students: young, bright, frequently middle-class. If a sample is narrow, then it is important to be careful about generalizing the findings to other groups. For example, if a researcher does an experiment on color perception using only male students as participants, we should be cautious about accepting that the findings can be applied to women, because we know that there are forms of color blindness that only men can have. However, we probably would be willing to accept that the results generalize to other young men. To take a different example, if an investigator finds that female college students in the United States spend more money on make-up than textbooks, we would not conclude that the same was true of male professors in Germany, at least not without further evidence. Many tests provide a system to convert raw test scores into a standard score. For example, IQ tests convert raw test scores into IQ scores, which are designed to have a mean of in the population as a whole. How do the test designers know which test score corresponds to the population mean? They run the test on a representative sample. The scores of the sample are used to define the norm. However, for many tests the sample used for norming is specific to a particular nation. Many tests are normed in the United States. If the test results are applied in a different population, then it is important to bear in mind that the norms may not be appropriate. For instance, someone may score low on a test for clinical depression not because they are clinically depressed, but because in their culture it is usual to respond to questions like those on the test in a particular way. For example, they might be asked to look at pictures of everyday objects, read some sentences, or write down the meaning of a phrase. It is always worth having a careful look at the particular items chosen, because sometimes the choice of materials can bias the results. For example, in many experiments on language, reading time is measured. For instance, an experimenter might test whether sentences are easier to read if they are in the form of a question or if they are in the form of a statement. It is important that the sentences are matched for word frequency, because, all other things being equal, people read common words more quickly than rare words. If it happened that the materials were biased, so that the question sentences contained more common words, any difference in reading time could be due to word frequency rather than sentence type. For instance, if participants read the question sentences more quickly, this could be either because they were in the form of questions, or because they contained more high frequency words, or a combination of both explanations. We could not be certain which conclusion to draw. In a situation such as this, the conditions of the experiment differ systematically in more than one way. One condition involves question sentences and high frequency words. The other involves statements and more low frequency words. Another common problem with materials arises if the materials are already known to participants. In many studies, especially in cognitive psychology, investigators aim to examine the reasoning processes and inferences people make when they tackle new situations. For example, in problem solving studies, participants are given puzzles to solve, and investigators examine the effect of variables such as whether the problem is expressed in concrete terms or abstractly. However, if the problem is one the participant already knows the answer to, they can solve it just by retrieving the answer from memory. A related problem affects tests and questionnaires which are administered more than once to the same participants. Obviously, we can expect the first test to influence responses to the second, and this influence can take different forms. Most simply, participants are likely to benefit from practice and greater familiarity with the items and the test format. Another possibility is that, on the second occasion, participants respond by recalling their answer to the first test. For example, I could ask you how many hours you spent eating on your last birthday. If I ask you now, you will probably work out a response by trying to recall what you did during your last birthday. However, if I ask you again in an hour, you may not bother going through that process. You might, instead, shortcut the process and simply repeat the answer you gave the first time, at least, as well as you can remember it an hour later. Conversely, participants might respond by choosing to give a different answer to the one they initially gave. This is more likely if the second time of asking is very soon after the first, and has been found to be a problem in developmental research. If a child is asked the same question again, they may reasonably guess that the first answer must not have been the answer that was desired. Researchers avoid the problems of repeated testing in two main ways. First, they sometimes redesign experiments so that the test or question is posed only once. Second, they devise alternative forms of a test, sometimes called parallel forms. At the second testing, a different version of the test is given. Participants will try to work out the intended meaning of questions they are asked, but they may not always interpret things the way the experimenter meant them to be interpreted. This is a particular issue in the design of questionnaires and surveys, but affects instructions in experiments as well. A straightforward example is when the experimenter uses a technical term in a question. In psychopathology, these terms have a specific definition. The experimenter cannot rely on participants to apply the same definition. A common problem is when the wording of a question is simply unclear or open to misinterpretation. This can be because the wording is vague. There can be more subtle problems with language. One problem that survey researchers noticed many years ago is that if you ask questions in a different order, you sometimes get different answers. One of the original examples was a pair of questions about letting journalists from overseas visit: 1. Do you think journalists from communist Russia should be allowed to come to the United States and report freely on what they see? Do you think journalists from the United States should be allowed to go to communist Russia and report freely on what they see? Materials can also become dated as society changes. For example, the following item appears on a scale devised to measure a personality trait called ethnocentrism. In developing their questionnaire, to which we return in one of the case studies, Adorno et al. In this section, we discuss two main issues. First we look at experimenter effects. These are ways in which the behavior, or even just the presence of the experimenter can influence participant responses. The second issue relates to within subject designs, in which participants complete more than one condition of the study. Experimenter effects There is a risk that the knowledge the experimenter has about expected results will affect the outcome of a study. There are various ways this can happen. If the participants then behave according to what they imagine the experimenter desires, this could bias the results. Depending on how well they guess, the results could be biased towards or against the research hypothesis. Conversely, it is possible for experimenters to unwittingly convey the response that suits their hypothesis. For example, an experimenter may ask a participant which of two options they would choose. If the experimenter unwittingly modulates their voice tone, or makes a gesture when explaining the choices, then that could affect responses. Similarly, if the experimenter always offers choice A first, and choice B second, that may have an effect. In many studies, the response or behavior of participants is recorded by the experimenter. There is a possibility that the experimenter will be unwittingly biased in making those records. Participants are asked to say something that is true in one condition, and to say something false in another condition. If the experimenter knows when the participant is supposed to be lying, their count could be unwittingly biased. This is an even bigger problem if the coding of the response involves subjective interpretation. For example, if the experimenter aims to record the number of times a participant says something cheerful, or angry, that involves a subjective judgement. The first part of the solution to these problems is for the person giving instructions or recording data to be unaware of the research hypothesis, and to be unaware which condition is being tested. A researcher in this position is said to be blind. One way to achieve this is to automate the running of the experiment. In many experiments in cognitive psychology, it is possible to present most of the experiment via a computer. The computer presents the trials and records the data, and is not subject to unwitting bias. Researchers have also been able to show in several cases that computer-based presentation of questionnaires gives similar results to paper and pencil presentations. By distancing the experimenter, who knows the hypothesis, from the data-gathering process, problems with experimenter effects can be reduced. For some studies, however, it is difficult to remove the experimenter. A typical example is where an interaction among people is being recorded and coded into qualitative categories. Someone needs to judge each part of the interaction and classify it, and often that person will be the experimenter. This opens up the possibility of a problem similar to experimenter effects, called experimenter bias. For example, imagine an experimenter is interested to know whether certain patterns of communication are more successful. The experimenter might set up a situation in which two people have to talk to each other to solve a problem. The experimenter would record what they say, and code it into categories that reflect the communicative intention of the participant making the utterance. The coding requires a subjective judgement. It is not practical to do such coding without using a human being to do it. If the experimenter does it, he or she may unwittingly bias the coding in favor of the experimental hypothesis. In this situation, three steps can be taken. First, a researcher can be employed and trained to use the coding system. In this way, the researcher is blind. However, this is potentially expensive, and rarely possible for that reason. The second step is to set out criteria for each category in the coding system, criteria that are as objective as possible. In some cases, it may be hard to provide objective criteria. Third, more than one person can code the responses independently. Statistics can then be used to quantify the extent to which their coding agrees. If the different coders make similar judgements, then this will at least establish that the coding is reliable. In certain types of research, the subjectivity of the experimenter is promoted as a positive feature. In such research, psychologists are usually interested to learn about the way people experience certain situations. These researchers use their intuition and subjectivity to try to get inside the perspective of the participant, using the things they say, and sometimes other aspects of their behavior, as a guide. This approach to research, called qualitative research, is widely used outside psychology, particularly in sociology, but also in other fields such as medical research. Within subject designs A fundamental issue relates to whether the same participants are used in each condition, or whether different groups are used. When different groups are used, there is a loss of statistical power because the individuals in each condition are different people. This creates variation between the groups that is not due to the independent variable, but is due to differences between the people in each group. A within subjects design has each participant complete more than one condition. This means that none of the differences between conditions can be attributed to differences between the individuals tested in each condition, because they are the same people. However, it also means that there can be differences between the conditions caused by practice or fatigue effects. Experimenters who use within subject designs have to be careful to avoid confounding practice and fatigue effects with the different levels of the independent variable. The researcher makes up two sets of items. The researcher gives the problems to a group of children, and times how long they take for each list. Each child does both lists, first the addition list, and then the subtraction list. Assume, for the sake of argument, that they are 2s faster for the subtraction list, and that the time difference is statistically significant. Would you accept the conclusion that children find subtraction problems easier? Well, we could not draw that conclusion. It could be that they were faster on the subtraction items because it was the second list, not because the problems were based on subtraction. In other words, the advantage of practising arithmetic on the first list may have helped them do better on the second list, the subtraction items. The design that I described confounds the order of the list first or second with the independent variable subtraction or addition. A fatigue effect occurs when participants are made tired or bored by one condition and so perform less well on later conditions. Researchers deal with order effects by counterbalancing the order of presentation. In this example, the researcher could give half the participants the addition list first, and half could get it second. When researchers counterbalance, they hope that the order effect will be the same size both ways. That is, they hope that the benefit of practising with addition is exactly the same as the benefit obtained from practising first with subtraction problems. If that is correct, then the counterbalancing will control the order effects. There are alternatives to counterbalancing that are useful when there are more conditions, such as randomizing the order of presentation for each participant. If that is a low probability, then the investigator rejects the view that the hypothesis is false. Conventionally, a probability of under 0. Assumptions Most statistical methods, including non-parametric methods, make assumptions about the data being analyzed. The term case study is used in parallel with terms like field study and observational study, each focusing on a particular aspect of the research methodology. For example, Lethbridge et al. Zelkowitz and Wallace propose a terminology that is somewhat different from what is used in other fields, and categorize project monitoring, case study and field study as observational methods Zelkowitz and Wallace This plethora of terms causes confusion and problems when trying to aggregate multiple empirical studies. The case study methodology is well suited for many kinds of software engineering research, as the objects of study are contemporary phenomena, which are hard to study in isolation. Case studies do not generate the same results on e. As they are different from analytical and controlled empirical studies, case studies have been criticized for being of less value, impossible to generalize from, being biased by researchers etc. This critique can be met by applying proper research methodology practices as well as reconsidering that knowledge is more than statistical significance Flyvbjerg ; Lee However, the research community has to learn more about the case study methodology in order to review and judge it properly. Case study methodology handbooks are superfluously available in e. In the field of information systems IS research, the case study methodology is also much more mature than in software engineering. For example, Benbasat et al. It is relevant to raise the question: what is specific for software engineering that motivates specialized research methodology? In addition to the specifics of the examples, the characteristics of software engineering objects of study are different from social science and also to some extent from information systems. The study objects are 1 private corporations or units of public agencies developing software rather than public agencies or private corporations using software systems; 2 project oriented rather than line or function oriented; and 3 the studied work is advanced engineering work conducted by highly educated people rather than routine work. Additionally, the software engineering research community has a pragmatic and result-oriented view on research methodology, rather than a philosophical stand, as noticed by Seaman The purpose of this paper is to provide guidance for the researcher conducting case studies, for reviewers of case study manuscripts and for readers of case study papers. It is synthesized from general methodology handbooks, mainly from the social science field, as well as literature from the information systems field, and adapted to software engineering needs. Existing literature on software engineering case studies is of course included as well. The underlying analysis is done by structuring the information according to a general case study research process presented in Section 2. Links to data sources are given by regular references. Rather it focuses on a set of issues that all contribute to the quality of the research. The minimum requirement for each issue must be judged in its context, and will most probably evolve over time. We do neither assess the current status of case study research in software engineering. Further, examples are used both to illustrate good practices and lack thereof. This paper is outlined as follows. We first define a set of terms in the field of empirical research, which we use throughout the paper Section 2. We define a case study research process Section 2. Section 3 discusses the design of a case study and planning for data collection. Section 4 describes the process of data collection. In Section 5 issues on data analysis are treated, and reporting is discussed in Section 6. Section 7 discusses reading and reviewing case study report, and Section 8 summarizes the paper. Case studies can be used to establish where the problem originates by utilizing several research methods. The primary methods used include: interviews, surveys, focus groups, observations and in some cases, field trials. Types of case studies[ edit ] In public-relations research, three types of case studies are used: [21] Linear, Process-oriented, Grounded. Under the more generalized category of case study exist several subdivisions, each of which is custom selected for use depending upon the goals of the investigator. These types of case study include the following: Illustrative case studies. These are primarily descriptive studies. They typically utilize one or two instances of an event to show the existing situation. Illustrative case studies serve primarily to make the unfamiliar familiar and to give readers a common language about the topic in question. Exploratory or pilot case studies. These are condensed case studies performed before implementing a large scale investigation. Their basic function is to help identify questions and select types of measurement prior to the main investigation. The primary pitfall of this type of study is that initial findings may seem convincing enough to be released prematurely as conclusions. Cumulative case studies.

These mixed methods researchers are likely to believe that qualitative and quantitative data and approaches will add studies as one considers viewer, if not all, research questions. By successfully conducting an equal-status study, the pragmatist researcher shows that paradigms can be mixed or combined, and Peoplesoft report manager list tab the incompatibility thesis does not always apply to research practice.

Timing has two aspects: simultaneity and dependence Guest Note that the use of capital letters pdf one component and lower case letters for another component in the same pdf suggest that one component is primary and How to case a good personal statement for grad school research is secondary or supplemental. Some designs are method by nature. It is possible, e. It is also possible to conduct the interviews after the survey data have been collected or vice versa ; in that viewer, research activities are performed sequentially.

We method two research components research if the implementation of the viewer pdf depends on the histories of data analysis in the first component. Two research components are independent, if their example of argumentative essay sample paper does not depend on the results of data analysis in the other component.

It is also possible to let the interview questions depend upon the viewers of the analysis of the questionnaire data or vice versa ; in that research, research activities are performed dependently.

In our study, simultaneity and dependence are two separate dimensions. Simultaneity indicates whether data collection is done concurrent or sequentially. Dependence indicates whether the implementation of one component depends upon the Baudelaire method exotico analysis essay of cases analysis of the other component.