PO217 Lecture Notes - Lecture 7: Design Of Experiments, Quasi
Document Summary
Overview: why is research design so important, the nature of casual inferences, the classical experimental design, variations on the classic experimental design, quasi- experiment designs. Purpose: to impose controlled restricted on our observations of the empirical world. A good research design: allows the researcher to draw casual inferences with confidence, defines the domain of generalizability of those inferences. We can never be certain that one variable "causes" another-- but we can increase confidence in our casual inferences if we are able to: In research we try and determine the causation. Correlation does not imply causation demonstrate co-variation. If a change in factor 1 does not lead to a change in factor 2, there is no co-variation. The variables must change together in order to identify a relationship establish time order. Has to occur during the same time eliminate sources of spuriousness. Show that there are not other factors affecting the relationship.