AP Stat Unit 4

What is the essential difference between an experiment and an observational study?

An experiment imposes treatments on the subjects, but an observational study does not.

You can only make inferences on a ____________.

Controlled experiment

What two methods are poor at representing the population well? Why?

Convenience: consistently under or overestimates a response. Voluntary response: attracts people who feel strongly about an issue

Non sampling errors

No response: subjects can't be contacted or refuse to participate

When controlled experiments are impractical or unethical, what is necessary to establish a cause-and-effect relation between two variables?

Strong association between variables, an association between variables is observed in many different settings, the alleged cause is plausible, and there are no lurking variables that would affect the response.

What is the principle use for controls in designing experiments?

They distinguish a treatment effect from the effects of confounding variables.

Sampling errors

Under coverage: when a portion of the population is left out when collecting data

Matched-pair design

When the matched pair is you to yourself

What is the principle use of random assignment in designing experiments?

It creates approximately equal groups for comparison

Why is a random sample essential to sampling and surveying?

It reduces bias and errors, helps with time management, gives every sample the same possibility to be chosen, and makes trustworthy inferences in the population.

What is the principle reason for replication in designing experiments?

It reduces sampling variability