FMST 210 Lecture Notes - Lecture 2: Confounding, Internal Validity, Random Assignment
Document Summary
Skepticism, or critically evaluating knowledge claims, requires an analysis of validity: validity: a matter of degree, 2 types of validity: internal and external. A single research study will never achieve absolute (100%) validity/truth: some studies have a higher degree of validity than others. Longitudinal design: measure variables at least twice (are the results. Internal validity can be increased with random assignment. This criterion demands that you are able to demonstrate that there is a relationship between the variables in your research question. How do you provide evidence of this criterion? (statistics) If you measure all the variables (survey, observational, case study), you can calculate a correlation. As long as the correlation is significant and not zero, you have evidence of an association: strength doesn"t matter. This criterion requires you to demonstrate that the association is unidirectional; associations are assumed to be bidirectional. Researchers must state the results for a unidirectional effect only.