POLI 311 Lecture Notes - Lecture 11: External Validity, Statistical Significance, Statistical Process Control
2 Apr 2016
School
Department
Course
Professor
Document Summary
We can use case studies to mimic the design of the classical experimental design. Types of case studies: atheoreical case studies: descripive studies of a single case. The purpose is not to propose or test hypotheses but to gain deep and nuanced understanding of a case study. This can simulate curiosity or interest in an anomaly which leads to theory building. The drawback is that it can spawn a lot of hypotheses, but likely most of them won"t hold more generally: plausibility probe case studies: designed to provide a preliminary assessment of the plausibility of a theory. Used to see if it"s worth devoing more ime and resources to something. Case has to be well chosen so that the results actually mater: deviant case studies: studies of cases that are known to deviate from an established generalisaion, designed to igure out why they deviate.
