Designing Experiments Section 5.2
Experimental Units Individuals on which the experiment is done If the individuals are human then we call them subjects
Treatment A specific experimental condition applied to the units (subjects)
Factors Explanatory variables in an experiment
Note: Experiments can give good evidence for causation They allow specific factors to be studied They study the combined effects of several factors
Placebo Effect Response to a dummy treatment
Control Group Group that receives the placebo
Principles of Experimental Design Control Randomize Replicate
Statistically Significant An observed effect so large it would rarely occur by chance
Completely Randomized When all experimental units are allocated at random among all treatments
Cautions About Experimentation
Lack of Realism Subjects, treatments, or setting of an experiment may not realistically duplicate the conditions
Double-blind Experiment Neither the subjects nor the people who have contact with them know which treatment a subject received
Randomized Comparative Experiment One of the most important ideas in statistics Able to give convincing evidence for causation
Matched Pairs Design Compares 2 treatments Blocks of 2 units that are as closely matched as possible Randomize the order for each subject Example of block designs
Block A group of experimental units or subjects that are known before the experiment to be similar in some way that is expected to affect the response to the treatments A form of control
Block Design Random assignment of units to treatments is carried out separately within each block Matched pairs is a form of block design
Practice Problems pg. 306 #5.49-5.58