M10: Experiments Flashcards
Response variable
Variable that we measure to draw conclusions about it
Experimental units
Individuals/subjects involved in the experiment
Treatments
Experimental conditions given to the subjects
How to avoid lurking variables and confounding effects?
Make sure conditions are as similar as possible
Comparative experiment
Compare two or more groups to eliminate confounding variables
Placebo
Dummy treatment
Control group
Group that receives placebo
Blind study
Subject does not know what treatment they receive
Double blind
Neither the subject nor the person who has contact with them know who gets treatment
Advantages of random samples
a) Can use probability to analyze the results
b) Avoid selection bias
of Replications
Number of experimental units that get each treatment
Statistical significance
Difference between two or more treatments are statistically significant if they are too large to be attributed to chance
Factor definition
Categorical explanatory variable in an experiment
How to get the number of treatments based on factor levels
treatments = Factor 1 x Factor 2
How to get the number of replications based on the number of treatments and number of experimental units
Experimental units (Subjects) / # Treatments
How are replications determined in observational studies?
The breakdown depends on the nature of the sample
Matched pair design
Each experimental unit is matched with another one on every possible confounder you can think of.
One person from each pair gets randomly assigned to one treatment
Identical twins
Cross-over design
When each person serves as their own “perfect match” and receives the two treatments in random order.
Blocked design
Matched pairs extended to three or more treatments
Each set of matched experimental units is a block
