ST 311 Exam 1

Parameter

Numerical description of the population

Statistic

Numerical characteristic of a sample

Simple Random Sampling (SRS)

Every possible sample in the population has a chance of being selected

Stratified Sample

Divide population into groups, then chose randomly from every group

Cluster Sample

Splits the population into areas and randomly selects which are sampled

Systematic Sample

Starting point, then every x item

Multistage Sample

Combining Methods

Bad Sampling Frame

Hard to list everyone

Undercoverage

Missing or underrepresented groups

Non-response bias

Hard to contact

Response Bias

Untruthful

Wording and Order

Elicit a certain response

Experiment

Apply a treatment and observe

Observational study

Observing what is always happening

Response Variable

Measures an outcome (experiment)

Explanatory Variables

Explains changes in response

Completely Randomized Designs

Random people to random treatments

Randomized Block Designs

Divide into subgroups

Matched Pair Design

Like randomized block design but there is only 2 treatment groups

Histogram

Graph of frequency distribution

Dotplot

Each value is a dot

Central Limit Theory

The larger the sample size, the better the approximation when the parent population is normal