Stats Vocab Unit 1

Systematic Random Sample (SRS)

Selecting a sample from an ordered arrangement of the population by randomly picking one of the first k individuals and then choosing every kth individual thereafter.

Cluster Random Sample

A population is already divided into groups. Then, a SRS is used to choose some of the groups and ALL the individuals from the selected groups are included in the sample.

Stratified random sample

A population is divided into similar groups. Then, a SRS is used to choose some individuals from EACH group.

Census

When data is collected from EVERYONE in the entire population.

Observational Study

When data is collected from individuals without assigning treatments.

Experiment

A study where researchers assign treatments to subjects and then measure a response.

Association vs. Causation

Observational studies can be used to show if variables are related (Associated). Experiments can be used to show if a treatment CAUSES the observed outcome.

Completely randomized experiment

When the subjects are randomly assigned to the different treatment groups.

Randomized block experiment

When subjects at first split into similar groups (blocks) to control for natural differences. Then WITHIN EACH BLOCK, subjects are randomly assigned to the different treatment groups.

Matched pairs experiment

When TWO similar experimental units are grouped together (such as twins, people with the same blood pressure, or the same person twice) Then, they are randomly assigned to treatments. One treatment to the first unit in the pair and the second treatment to the other unit (or the order of treatment is randomized if using the same person).