Lecture 4: Experiment 2011, 9, 15
How many pairs of shoes do you have in your closet? Population mean: Sample 1 mean: Sample 2 mean: Sample 3 mean: Sample 4 mean: Why are the sample means different from population? Why different from sample to sample? Why was sample 4 mean so small?
Lecture 4 Topics Research goal of an experiment Internal validity of an experiment** –What is a confounder? External validity of an experiment*
Research Goal of an Experiment Determine cause-effect relationships between the independent variable (cause) and the dependent variable (effect) –Example: Hawthorne experiment –Independent variable: –Dependent variable:
Internal Validity of Experiment The extent to which your independent variable is the only possible explanation of the changes shown in your dependent variable –Example: Hawthorne experiment
Confounder An uncontrolled extraneous variable in an experiment that alter the true relationship between the independent variable (cause) and the dependent variable (effect) IVLightingConditions DVProductivity Motivation Cause
Statistical Research Story
How to improve the internal validity? How to Control Confounders? Will attending a review session be helpful? –Randomized experiment: The participants are randomly assigned to experimental conditions –Control confounders: Make the groups approximately equal in all respects except for the independent variable, which is purposely manipulated Test the effect of a newly-developed drug –Double-Blind Design: Control experimenter bias –Widely used at clinical settings
External Validity of an Experiment The extent to which the results of an experiment can be generalized to outside world –Example: Test the effect of a newly-developed drug How to improve the external validity? Trade-Off between Internal and External Validity
Lecture 4 Recap Research goals: Cause-effect relations Internal validity of an experiment** –Confounders –How to control confounders Randomization Double-blind experiment External validity of an experiment*