U2 Stats Test

Convenience Sample

Asking people readily available, easy to reach

Voluntary Response

People choosing to respond. Skewed results.

Simple Random Sample

1. Label Individuals

• Assign #s, write names on slips of paper

2. Randomize

• Random # generator, draw from a shuffled hat (NO REPEATS)

3. Select

Stratified Random Sample

Split the population into groups, then chose a simple random sample from each group

Cluster Random Sample

Split the population into groups, then sample an entire group

Systematic Random Sample

Chose a random starting point, use equal intervals to pick the sample

Stratified Sample Benefits

Guaranteed to receive data from all groups

Benefits of Cluster sample

Easy and fast to collect

Non Response Bias

people couldn’t be reached or refuse to answer

Undercoverage

Some people are less likely to be chosen (ex. calling landlines)

Response Bias

Problems in the data gathering or instrument

Well Designed Experiment

1. Comparison

2. Random Assignment

3. Replication

4. Control

Randomized Block Design

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

Matched Pairs Design

subjects are paired on similar characteristics and then randomly assigned to treatment

In this context, statistically significant means the observed changes in [context] were too large to attribute to chance

Statistically Significant