UNIT 2.1-2.8 [AP STATISTICS]

What are the steps of a Simple Random Sample?

1. Label

2. Randomize

3. Select

What can one do to sample a population?

1. Convenience Sample

2. Voluntary Response

3. Simple Random Sample (SRS)

How can one take a Stratified Random Sample?

Split the population into groups (strata) and then choose simple random samples from each strata.

What are strata?

The groups selected from a population in a sample.

What does strata have?

Each strata has individuals with shared attributes or characteristics.

What are good estimates?

A sampling method that produces the best estimates with low bias and low variability.

What is a Stratified Random Sample?

When some are sampled from all homogeneous groups.

What is a Cluster Random Sample?

When all are sampled from some heterogeneous groups.

What is a Systematic Random Sample

When one chooses a random starting point and selects based on random intervals. (Every n^th one.)

What is undercoverage?

When some people are less likely to be chosen.

What is nonresponse?

When participants cannot be reached or refuse to answer.

What is response bias?

When there are problems in the data-gathering instrument or process.

What is an explanatory variable/factor?

The factor causing something.

What is a response variable?

The result of what the explanatory variable poses.

What is a confounding variable?

Another factor that is related to the explanatory and influences the response.

What is an observational study?

When no treatment is imposed.

What is an experiment?

When treatments are imposed, can showcase causation.

What are experimental units?

What/Who the treatment is imposed on.

What are treatments?

What is done/not done to experimental units.

What would be a well-designed experiment?

1. Comparison (Two or more treatments.)

2. Random Assignment

3. Replication (More than one in each treatment group.)

4. Control (Keep all other labels constant.)

How can random assignment be done?

1. Label

2. Randomize

3. Assign

Utilized in order to show causation.

What is the placebo effect?

When a fake treatment works.

What is a placebo?

A fake treatment.

What is a single blind?

When subjects don’t know about treatments.

What is a double blind?

When subjects and/or experimenters don’t know about treatments.

What is a Block?

A group of experimental units that are similar.

How can a Randomized Block Design be organized?

Separate subjects into blocks and then randomly assign to treatments within each block.

How can a Matched Pairs Design be made?

Subjects are paired and then randomly assigned to treatments. (Each subject receives two treatments, the number of the treatments must be randomized.)

What is Statistical Significance?

When results of an experiment are unlikely (less than 5%) to happen purely by chance.

What happens when something is statistically significant?

If statistically significant, there is convincing evidence that treatment caused the difference.

How can one calculate statistical significance?

1. Find the mean for each group and subtract the means.

2. 
   

What can a random sample do?

Allows us to generalize a conclusion to the population from which was sampled.

What can random assignment do?

Allows us to conclude a treatment causes changes in the response variable.