Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Fall 12: INF1240 Research Methods: Case Studies & MultiMethod
Transcript of Fall 12: INF1240 Research Methods: Case Studies & MultiMethod
Case Studies & Mixed-Methods
Moving from methods to research strategies
Robert K. Yin (a.k.a. King of Case Studies)
Frequent confusion between:
types of evidence (qualitative/quantitative)
types of data collection (methods)
types of research strategies
Apparent in critiques and discussions of specific studies, methods, etc.
When to use Single-Case Study?:
When it represents a critical case in testing an established theory (i.e. it meets all of the conditions set forth/proposed by the theory)
When it represents an extreme or unique case (i.e. rare, outlier, exceptional, or perhaps indicative of a very new/emerging phenomenon)
When it is a revelatory case (i.e. elusive, difficult to gain access to, previously inaccessible, etc.)
single vs. multiple (unit + sub-units) units of analysis.
e.g. Holistic case study of a specific Facebook group
e.g. Embedded case study of a specific Facebook group (unit), its individual members (sub-units)
Your plan of action
wherein action = data collection using the appropriate method or methods, selected to answer your research question(s)
Multiple strategies available
Example: Case studies
Yin (1981) describes: "As a research strategy, the distinguishing characteristic of the case study is that it attempts to examine:
a) a contemporary phenomenon in its real-life context, especially when:
b) the boundaries between phenomenon and context are not clearly evident" (p.59).
Not usually linked to a particular type of evidence or method
Once again: Criticisms about scope/generalizability
When to use Multiple-Case Study Designs?:
Traditionally seen (by canonical researchers) as a different "methodology" (e.g. political science and anthropology = different sets of rationales for "classic" (single) vs. "comparative" (multiple) studies)
When "replicability" is an important component of your research question/topic of inquiry (***replicability is NOT equivalent to random sampling/generalizability)
When each of the cases selected either: a) suggests similar trends/features/results (literal replication) or b) suggests contradictory trends/features/results, but for predictable reasons (theoretical replication)
Doesn't establish causality, or generalizable evidence - but rather: compelling support for the theory or propositions that you've set out to explore
In both instances: development of a rich, theoretical framework is crucial (Yin, 1984).
This framework must:
"state the conditions under which a particular phenomenon is likely to be found"
"the conditions when it is not likely to be found" (p.54)
Yin, R.K. (1984). Case Study Research: Design and Methods. Sage Publications.
Many research designs in STS (science and tech studies), esp. those following the SCOT (social construction of technology) approach, use a "case study" strategy (e.g. historical case studies), combining a variety of data collection methods. (e.g. Pinch &Bijker, 1987). Hence the Beaulieu et al. (2007) article.
Research strategy for examining phenomena that happened in the past - where methods such as interviews may not be possible (or even the most reliable), can't do ethnography/observation, etc.
Not just a temporal classification for text/artifact analysis, but instead an entire way of approaching and organizing data.
as research design
as research strategy
Simons (2009): to understand the "distinctiveness of [an] individual case" (p.3)
Why a case study?
Smith (1978) ...of a "bounded system"
MacDonald and Walker (1975) ...of an "instance in action"
Studies of tech development/history:
Studies of contemporary tech:
Text analysis, discourse, etc.
e.g. of multi-method strategy
"methodological pluralism" is the approach, strategy, framework - not just the way the data is collected/analyzed
Strives to bring qualitative and quantitative methods "into epistemological alignment as research methods with the same claims to knowledge and explanatory power, although with different, complementary roles to play in social and cultural analysis" (Schroder et al., 2003, p.45)
Belief that social reality "exists independently of language" = but language is also our only way of accessing knowledge.
So everything, all data, all research (empirical and interpretative), is mediated by a discursive layer
Goal is to be as holistic as possible - tackling a phenomenon or question from a broad, multiplicity of angles, applying induction and deduction (and abduction - creation of new knowledge).
Framework that highly encourages mixed method design - oftentimes combining qualitative and quantitative in complimentary, synergistic ways
The question drives the data collection - narrowed down to a case study or a case history, the scope/volume of data becomes more manageable.
Theory drives analysis: Looking at Pinch & Bijker: they delineate where to look for data through the application of key constructs that form the core of their theory/approach:
e.g. the notions of "interpretative flexibility" (the same technology can mean different things to different people) and the "principle of symmetry" (for any successful tech, there are also failures/unsuccessful: VHS vs. BETA):
e.g. the identification of "relevant social groups": institutions, organizations, groups of individuals concerned with the artifact - attach to it a (shared) meanings/importance that impact upon decisions about how and where it is used:
INF1240: Research Methods
Week 10: Nov. 12, 2012
Holistic vs. embedded
Single vs. Multiple Case Study Designs:
Yin sees them as two different ways to carry out the same type of strategy. Answer different questions, involve different levels of analysis.
drives them to examine both obvious groups (e.g. consumers, users, designers) and not so obvious groups (e.g. protesters, religious groups, excluded groups)
Bias Alert: Major criticisms about how "relevant" was operationalized in early applications of this theory - e.g. dis- empowered groups omitted from written "history" (press, corporate memos, etc.) also omitted from analysis. SCOT scholars now try to account for this as much as possible.
drives them to examine both successful and non-successful variations of a particular technology (e.g. bicycle), and
consider how "selection" plays a role in technological development