Case Selection

Question 1

Case Selection and External Validity

Answer 1

• External validity is the extent to which research findings can be generalized beyond the specific cases studied.
• To maximize external validity, researchers should ideally study the entire population of cases. However, this is often not practical, so researchers work with samples.
• Random sampling techniques are commonly used in large-n (quantitative) research designs. They include random samples and stratified random samples. These techniques aim to minimize selection bias and increase external validity.
• Non-random sampling techniques like convenience sampling and snowball sampling are more commonly used in small-n or qualitative research. While they may introduce some level of selection bias, they are often more practical for studying specific cases or populations.

Question 2

Small-n Comparative Case Selection

Answer 2

• Small-n research involves studying a limited number of cases in-depth to understand relationships and causal factors.
• To select cases in small-n comparative research, researchers aim to identify cases that share relevant characteristics or variables to reduce the number of variables they need to control for.
• The comparative approach is valuable for identifying patterns, understanding causation, and testing theories.

Question 3

Mill’s Methods (causation)

Answer 3

• Mill’s methods are used in small-n comparative research to determine causation. They include the method of difference and the method of agreement.
• Method of Difference: Researchers look for cases with different values on the dependent variable while keeping other relevant variables constant. By examining the potential differences, they infer what causes variations in the dependent variable.
• Method of Agreement: Researchers find cases with the same value on the dependent variable but differing on other key independent variables. They explore similarities between these cases to understand why they share the same outcome, even with varying independent variables.
• These methods help researchers establish causal relationships by systematically examining variations in variables.

Question 4

Selecting on the Dependent Variable

Answer 4

• Selecting cases based on the dependent variable can introduce selection bias, which is a distortion in research findings. For example, if cases are chosen because they exhibit the outcome of interest, it may not be possible to understand the impact of various independent variables on that outcome.
• Researchers should avoid selecting cases based solely on the dependent variable when attempting to make causal inferences or generalize findings.

Question 5

Case Selection: Single-Case Studies

Answer 5

• Single-case studies vary in their objectives and research designs. They can serve different purposes.
• Atheoretical Case Study: This is a purely descriptive approach where the researcher provides a detailed account of the case without relying on established theories or hypotheses.
• Interpretative Case Study: These case studies select cases of interest and use existing theories to interpret and explain the phenomena observed.
• Hypothesis-Generating Case Study: In these studies, researchers select cases to develop new theories or hypotheses in areas where knowledge is limited.
• Most-Likely Case: Researchers select cases that are expected to be the most likely to exhibit the outcome they are studying. This is analogous to a “pass-fail” test for a theoretical proposition.
• Least-Likely Case: Researchers select cases that are expected to be the least likely to exhibit the outcome. This is a “tough test” for a theory and can only marginally weaken a generalization if the theory holds.
• Deviant or Outlier Case: These cases are known to deviate from established generalizations and can provide valuable insights into the boundaries and limitations of existing theories.