Ch 4 Section 2&3 Flashcards
Experiment
Deliberately imposes some treatment on individuals to measure their responses
(Actually doing something to people)
Confounding
Occurs when 2 variables are associated in such a way that their effects on a response variable cannot be distinguished from each other.
Treatment
Specific condition applied to the individuals in an experiment
(What you are doing)
Experimental units
Smallest collection of individuals to which treatments are applied
(Doing on/ to)
Subjects
Units are human beings, they are called this
Factors
explanatory variables in an experiment
Level
Each treatment formed by combining a specific value of each of the factors
Random assignment
Experimental units are assigned to treatments using a chance process.
Completely randomized design
Experimental units are assigned to the treatments completely by chance.
All experiments will have what?
Comparison Random assignment Control Replication (doing it on enough subjects) CCRR
What are the simplest statistical designs for experiments? What is this like?
Completely randomized designs for experiments
Like SRS are for samples
Stratifying is to surveys as what is to what
Blocking is to experiments
When should you block??
Block things only when there are things that will affect response variable
Block
Group of experimental units known before experiment to be similar in some way that’s expected to affect the response to the treatment
Randomized block design
Random assignment of experimental units to treatments is carried out separately within each block.
____ what you can, ____ on what you want, and then ____ the create comparable groups
Control what you can, block what you want, and then randomize to create comparable groups.
Matched pairs design
Common type of randomized block design for comparing 2 treatments
-create blocks by matching pairs of similar experimental units
Ex of matched pairs design
Twins, shoes, match w self
What can influence the response when dealing with matched pairs design? What should we do about this problem? Example?
Order of the treatments
So we randomize the order for each experimental unit
Cookie A & B, do in different order so the first cookie doesn’t taste better bc you were hungry
Observational study
Observes individuals & measures variables of interest, but does NOT attempt to influence the responses. (Survey).
DOES NOT PROVE CAUSATION.
What type of influence (or conclusion) can be made from a particular study?
What do they allow?
Answers depend on design of study.
- random selection of individuals, which allows inference about the population
- random assignment of individuals to groups, which permits influence about cause & effect
Criteria for establishing causation when an experiment cannot be done?
Association must be strong
Association must be consistent
Larger values of the explanatory variable are associated with stronger responses
Alleged cause precedes the effect in time
Alleged cause is plausible