#### Math Final

Simple Random Sample

Every combination of units has an equal chance of being selected

Stratified Random Sample

The study population is divided into easily identifiable strata (subpopulations) and then each strata is surveyed. Use when there is a mitigating factor

Systematic Sample

All subjects are listed and assigned sequential numbers from 1 to N. The required sample size n is determined and a sampling fraction (f) is calculated by dividing N by n. Every f(th) person is then selected.

Cluster Sampling

The sample population is divided into naturally occurring clusters (groups). All clusters are listed and a sample of clusters is randomly selected. All subjects in the selected clusters are then interviewed

Multistage Sampling

The selection of subjects to be selected for stratified or cluster samples may involve several stages.

Convenience Sampling

Using a sample of people who are readily available to participate (easy to do)

Volunteer Sampling

A sample of participants produced by a sampling technique that relies solely on inviting people to take part

Census

Surveying ALL members of a population. Only use with small populations and easy access to all of them

Undercoverage

When certain populations are not represented (excluded)

Volunteer Response Bias

Volunteers are relied on (mostly get polar opinions)

Nonresponse Bias

Bias introduced to a sample when a large fraction of those sampled fails to respond

Response Bias

People responding to the survey are influenced by the question or the surveyor

Observational Study

Researchers don't assign choices but instead observe them

Randomized, Comparative Experiment

Researchers take a group and randomly assign treatments

Retrospective Study

Identify subjects first that have already done something and then collect data on that thing

Prospective Study

Identify subjects in advance and then collect data from what they do

Factor

An explanatory variable thats manipulated in the experiment

Response variable

A variable measured in the experiment (dependent variable)

Levels

The specific values that the experimenter chooses for a factor

Treatment

Combinatiom of specific levels from all the factors that an experimental unit receives

Control

Controls services of variation by making conditions equal

Randomize

equalize the effects of unknown or uncontrollable sources of variation

Replicate

Using multiple subjects or repeating the experiment

Block

Targeting a blocking variable (attribute causing variation)

the four principles of experimental design

control, randomize, replicate, block

Blinding

When people don't know the difference between the variables being tested

Double blinding

When both the subject and administrators are unaware of which treatment the subject is receiving

Single blinding

a technique in which the subject doesn't know whether he or she is receiving a treatment or a placebo

Placebo

A fake treatment that looks just like the one being tested

Placebo effect

When people report having results from the placebo

Blocking

Grouping similar individuals with traits we aren't interested in to reduce variability due to subsets of a sample reacting differently

Confounding

A variable associated in a noncausal way with a factor and affects the response

Lurking

A variable associated with both x and y that makes it appear that x may be causing y