319 Lecture 1: Data, Sampling, and Statistical Thinking

Parameter

A numerical characteristic describing the population; usually unknown

Statistic

A numerical summary calculated from the sample, or is observed from the data

Population

The entire collection of objects or outcomes about which information is sought

Sample

A small selection of the population that is actually observed

Simple Random Samples (SRS)

A simple random sample of size n is selected so that every possible collection of n population items is equally likely to be choosen

What makes Sampling variation different from Sampling bias?

SV: random differences among properly selected samples, and can usually be reduced by increasing sample size.

SB: systematic error caused by how the observations are selected or measured, and a larger sample does not necessarily fix it.

Convenience samples

samples that are obtained in an easy way, and not in a well-defined random method

Variables

Is a characteristic that can take different values for different objects or outcomes

Controlled experiment

The investigator controls or assigns factor values. Which a well designed randomized experiment can provide strong evidence for cause and effect

Observational study

The investigator observes existing factor values without assigning them. Associations may be affected by confounding variables.