Stats

Population

All individuals in a defined group.

Sample

Subset of individuals selected from a population.

Census

A method that gathers information from every individual in a population.

Inference

The process of drawing conclusions about a population based on a sample.

Simple Random Sample (SRS)

A sampling method where all groups of size n have an equal chance of being chosen.

Randomization Methods

Techniques such as calculators, tables of random digits, or drawing names from a hat to achieve randomness.

Stratified Sampling

A method where individuals in groups are alike, and individuals are selected from each group.

Cluster Sampling

A method where individuals in groups represent the entire population, and entire groups are selected.

Bias

A systematic error that causes estimates to be consistently too high or too low.

Convenience Sampling

A sampling method where individuals are chosen based on ease of access.

Voluntary Response

A sampling method where only some individuals choose to respond.

Response Bias

Occurs when people lie or when questions are poorly worded.

Non-response

A situation where individuals cannot be reached or choose not to reply.

Undercoverage

A sampling issue where part of the population is excluded.

Observational Study

A study that examines the relationship between two variables without manipulation.

Retrospective Study

A study that looks back at past data.

Prospective Study

A study that looks forward and collects data over time.

Experiment

A study where treatments are assigned to observe cause and effect.

Explanatory Variable

The variable that is manipulated or studied in an experiment.

Response Variable

The variable that is measured in an experiment.

Confounding Variables

Other variables that may influence the results of an experiment.

Purpose of Randomization

To create roughly equivalent groups for comparison.

Experimental Units

The subjects or entities that receive the treatment in an experiment.

Replication

The process of having enough individuals in a study to ensure valid results or repeating the study with more individuals.

More Data

Leads to a smaller margin of error in results.

Control

Keeping other variables constant to isolate the effect of the treatment.

Control Group

A group that receives a placebo or old treatment for comparison purposes.

Placebo Effect

A phenomenon where individuals experience a dummy response to a fake treatment.

Single Blind

An experimental design where one side (either individuals or experimenters) does not know the treatment given.

Double Blind

An experimental design where both sides do not know the treatment given.

Statistically Significant

Results that are unlikely to occur by chance alone, typically with a probability of less than 5%.

Blocking

A technique similar to stratifying, but used specifically for experiments.

Matched Pairs

A blocking technique where individuals are grouped in pairs for comparison.

Random Selection

Allows for conclusions to be made about the entire population.

Random Assignment

Allows for conclusions to be made about cause and effect relationships.

Four Principles of Experimental Design

Randomization, control, comparison, and replication.