CMST 2TM6 Lecture Notes - Lecture 19: Statistical Power, Face Validity, Content Validity
Document Summary
The standards used to evaluate whether particular evidence represents an effect way to support a claim. Precision: detailed accuracy in defining and measuring variables. Power: findings are generalizable- therefore more powerful descriptions/explanations of communication statistical power. Parsimony: valuing precise, powerful and succinct descriptions of variables and their relationships. Necessity of cause: when the effect occurs when the cause is present and the effect cannot occur when the cause is absent. Sufficiency of cause: when the cause the best possible explanation for the event. Expected to influence a change in another variable. Systematically varied by researcher - research wishes to explain x. Dependent variable - y. expected to change as a result of actions of the independent variable. Correlation = changes that occur together, but not necessarily x causes y. Positive correlation; as x goes up, y goes up. Negative correlation; as x goes up, y goes down. Internal validity: precision and accuracy with observations in the study.