Sampling
The good type of sampling is...
Random Sampling
What is the technology called that we use to randomly sample and randomly assign to treatments?
Random Number Generator
Is statistical significance good or bad for the results of an experiment?
Good!
The way we assign experimental units to treatment groups is called the...
Larger sample sizes are better because...
They give a more accurate estimate of the true value in the population
The group of interest that we sample from is called the...
Population
Do experiments completely eliminate confounding variables?
No
Does a difference in results between experimental groups always mean that the treatment is effecting the response variable?
Observational studies can only make inferences about...
What is the equation for Margin of Error?
2 time the Standard Deviation (of the simulation)
We avoid using convenience and voluntary samples because they can create...
Bias
If an experiment is double-blind, it means that...
Neither the subject nor the experimenter interacting with them know whether they get the treatment or placebo
Run a simulation
Random sampling allows us to make inferences about...
The population
Is "How long will you be in line when you get your lunch from the cafeteria?" a valid statistical question?
Yes!
What is the difference between nonresponse and response bias?
Nonresponse is when subjects in the sample don't respond, while response bias is a consistent inaccurate response
Why do we use random assignment of treatments to subjects in an experiment?
To create roughly equivalent groups
Too large to be explained by chance variation in random assignment
Random Assignment allows us to make inferences about...
Cause and Effect
What are confounding variables? What do they make it harder to determine?
Variables that are not the treatment but that can affect the response variable. They make it harder to determine cause and effect.
What is the term used to explain why samples only give us an estimate of a population's true value?
Sampling Variability
Just thinking you are getting a treatment can have an effect on the response variable
How do we use a simulation to determine statistical significance?
See how often the sample difference (or greater) occurs in the simulation. If it does not happen often (low %) then it is significant
Most experiments have what type of sampling and what type of assignment to treatment?
Non-random sampling and random assignment
What needs to be included in outlines of completely randomized designs? (5 things)
Sample, "random assignment", groups & their size, treatments, comparisons of response variable