MANA 420 Lecture Notes - Lecture 4: Dependent And Independent Variables, Internal Validity, Design Of Experiments
Document Summary
Review in experimental designs it"s difficult to isolate causes. What is causing what: we need to manipulate independent variable, randomly assign people to condition of independent variable, and we control for extraneous variable. Internal validity: the confidence with which you"re saying that the causal relationship is a result of the manipulation of the independent variable: experimental design offers use the highest degree of internal validity. Random assignment is the most vital part because it wipes out all the extraneous variables. Use the t-test: 1 variable = nominal with 2 levels (2 groups, variable 2 = interval or ratio. T test: on average, where is the midpoint. Interval and ratio scales are treated the exact same way in terms of analysis.