QRM - Week 4 - Ch. 14, 15, 16, 17

20 flashcards · Other · by Daniel F.

approach to designing case studies
Case study method requires a design, but there is little information on how to do this. not recognized as a separate research method less scientific, limited generalization, Complexity and variability, no standardized approach, Publication bias
research design
Logical plan for logical plan for getting from here to there, where here may be defined as the initial set of questions to be answered, and there is some set of conclusions about these questions what questions to study, what data are relevant, what data to collect, and how to analyse the results
Components of research design for case studies
Study's questions: Appropriate for 'how' and 'why' questions. Develop questions in three stages using literature, examining key studies, and refining potential questions. Study's propositions: Direct attention to what should be examined within the scope of the study. If exploratory, state the purpose. Unit(s) of analysis: Define what the case is (individuals, events, companies, etc.). Discuss with others to avoid incorrectly identifying the unit of analysis. Logic linking the data to the propos…
role of theory in design work
Theory development is essential for propositions and 'rival' theories. Theory provides a blueprint for the study and guides data collection and analysis strategies. Analytic generalization uses a previously developed theory as a template to compare empirical results of the case study. multiple case studies are based on the same theory show the same results means its credible lvl 1 statistical generalization making inferences about a population based on empirical data collected from a sample of…
Criteria for judging the quality of research designs
Construct validity: Identify correct operational measures for concepts being studied. name the specific concepts and then match operational measures to these, preferable with relation to literature tactics Use multiple sources of data to understand concepts Create a clear link between the research questions, the data collected, and the conclusions drawn Share draft reports with key informants to gather feedback Internal validity: Establish a causal relationship between conditions. tactics Compa…
Case study designs single multi case - holistic embedded
Single-case holistic investigate a single case in its entirety case is unique, bounded, complex-interaction-heavy NO breaking down the case into smaller units or subunits for separate analysis aim: gain an in-depth understanding of the complex interactions and dynamics within the case Single-case embedded investigate one case in its entirety break it down, investigate subunits detailed analysis of particular components aim: explore specific aspects of a case that may have broader implications o…
Potential single-case designs
Critical case case is crucial for understanding the phenomenon theory holds true in this critical case, it is likely to be applicable in other, less critical cases as well Extreme or unique case case is outlier or an exception in the context of the phenomenon explore the limits of a theory or uncover new insights Representative or typical case case is accurate representation of the broader population or phenomenon Revelatory case case offers offers unique access to or insight into the phenomeno…
Potential multiple-case designs
require high levels of resources and time, sampling might bet better Replication logic selecting cases that allow for the replication of findings from one case to another strengthen the internal and external validity of the research by demonstrating consistent patterns Literal replication cases are very similar in terms of context and characteristics goal: show that the same results occur in each case Theoretical replication cases that differ in context and characteristics but are predicted to…
advice in selecting case study designs
Consider multiple-case designs for replication or contrast possibilities. 2 cases better than 1 Balance closed and flexible designs. Consider mixed methods designs for addressing broader or more complicated research questions.
case study investigator: desired skills
Asking good questions - create a rich dialogue with the evidence Being a good 'listener' - listening between lines Exercising adaptiveness and flexibility - cases might change, researcher must stay unbiased Having a firm grasp of the issues being studied Avoiding bias
Preparation and training for a specific case study
Human subjects protection Gain consent Protect them from harm Protect privacy Protect vulnerable groups Case study training as a seminar experience Addressing problems within the case study plan
The case study protocol
document that guides the researcher in carrying out the data collection from a single case instrument and the procedures and general rules to be followed in using the protocol Increases reliability Sections An overview of the case study project Background info, issues being investigates, relevant readings Field procedures - different from an experiment or survey Gaining access to key organizations or interviewees Having sufficient resources while in the field Developing a procedure for calling…
Screening the candidate ‘cases’ for your case study
Assessing cases against prewritten criteria Collecting limited data about the cases
The pilot case study
Select case for convenience or criticality Make scope a bit wider to get more insight Reports should be explicit and clear
Six sources of case study evidence
Documentation: Personal documents, reports of events, administrative documents, formal studies or evaluations, and mass media or community newspaper reports. Pros: direct evidence, analysed at convenience Cons: biased or incomplete, limited, authenticity not guaranteed Archival records: Public use files, service records, organizational records, maps, charts, and survey data. Pros: systematic and structured source, covers time, allows for temporal comparison Cons: may lack context, be incomplete…
Three principles of data collection
Use multiple sources of evidence: Develop converging lines of inquiry through triangulation and corroboration. Four types of triangulation include data sources, evaluators, perspectives, and methods. Create a case study database: Separate the data or evidentiary base from the report of the investigator. Include case study notes, documents, tabular material, and narratives in the database. Maintain a chain of evidence: Allow an external observer to follow the derivation of any evidence from init…
General Strategies for Analysing Case Study Evidence
follow the theoretical propositions that led to your case study develop a descriptive framework for organizing the case study Using both qualitative and quantitative data Examining rival explanations
analytic techniques for case studies
Pattern matching: Compare an empirically based pattern with a predicted one to increase internal validity. Explanation building: Analyze case study data by building an explanation about the case, focusing on causal links. Time-series analysis: Trace changes over time by matching the time-series trend with a theoretically significant trend or a rival trend. Logic models: Stipulate a complex chain of events over an extended period, matching empirically observed events to theoretically predicted e…
Ensure high quality analysis of case studies
Attend to all evidence when assessing research questions and propositions. Address all major rival interpretations. Focus on the most significant aspects to solve the main issue. Use and show prior, expert knowledge.
Types of triangulation
Data sources triangulation using multiple data sources to gather information about the same phenomenon Evaluators triangulation using multiple evaluators or researchers to analyze and interpret the data Perspectives triangulation collecting data from different stakeholders or participants with diverse perspectives