POLSCI 2NN3 Lecture Notes - Lecture 9: Internal Validity, Demand Characteristics
Document Summary
How to structure research to be confidence in asserting a casual relationship. The key is to have a good research design. If your research is targeted at demonstrating cause-effect relationships: your research design should allow you to convincingly demonstrate three things. Changes that occur in both variables are parallel. Time aspect; difficult because things are happening simultaneously. Can see that two variables that move together, but exist in a context with a lot of other variables. There are many types of experimental design. Consider the simplest: the pre-test/post-test laboratory experiment. A hypothesis is stipulated that there is a relationship between two variables. Any change in the dependent variable is attributed to the independent variable. All experimental designs (including this one) must reflect on two issues: internal. In the form above, this is not a strong design. Framework for the on going example for the lecture.