Unit 4: Sampling and Experimental Design Vocabulary

Bias

Bias

Any systematic failure of a sampling method to represent its population.

Blinding

Any individual associated with an experiment who is not aware of how subjects have been assigned.

Blocking

Grouping experimental units by some factor that we are not testing if we believe that factor could affect the outcomes.

Census

A sample that consists of the entire population.

Cluster Sample

A sampling design that groups the population in a convenient way and then randomly selects groups to sample and collects data about the entire group.

Confounding

A variable that has an effect on the experiment in a way that cannot be separated from the actual treatments; it is a reason we block.

Control

A principle of experimental design; controlling any factor we can when designing an experiment.

Control Group

The experimental units assigned to a placebo or standard treatment; it allows for comparison.

Convenience Sample

A sample taken of individuals that are easily selected.

Double Blind Experiment

An experiment in which neither those who could influence the results nor those who evaluate the results know who was assigned to which treatment.

Experiment

A process that imposes treatments on randomly selected or assigned subjects then compares the responses across treatment levels.

Experimental Units

Individuals on whom an experiment is performed; called subjects when people.

Factor

A variable whose levels are controlled by the experimenter.

Heterogeneous

Different

Homogeneous

Alike.

Level

The specific value that an experimenter chooses for a factor.

Matched Pairs

A type of design that pairs subjects that are very similar in order to compare them to each other.

Multistage Sample

A sampling method that combines several sampling methods.

Nonresponse Bias

Bias introduced to a sample when a large portion of those sampled do not respond.

Observational Study

A study that gathers data from a sample without imposing a treatment.

Placebo

A treatment that has no effect.

Placebo Effect

The tendency of experimental units to respond to treatment regardless of the type of treatment (even for control group).

Population

The entire group of individuals or instances about whom we hope to learn.

Prospective

A study in which individuals are selected and monitored for future behavior.

Randomization

A principle of experimental design; the process of selecting a subject randomly.

Replication

A principle of experimental design which requires many experimental units in each treatment.

Response

A variable whose values are compared across different treatments.

Response Bias

A type of bias resulting from false information being given.

Retrospective Study

A study in which individuals are selected and surveyed about past behavior.

Sample

A representative subset of a population.

Sampling Frame

A complete list of individuals/items from which the sample is drawn.

Simple Random Sample

A sample in which each item has an equal chance of being selected and every combination has an equal chance of being selected.

Simulation

A sequence of random outcomes that model a situation.

Statistically Significant

An observed difference that is too large to likely have occurred naturally.

Stratified Random Sample

A sampling design that divides the population into homogeneous groups and random samples are selected from each group.

Single Blind Experiment

An experiment in which either those who could influence the results or those who evaluate the results know who was assigned to which treatment but not both.

Systematic Sample

A sampling design that randomly selects the first item and then uses a pattern to select each additional item.

Sampling Variability

The natural tendency of randomly drawn samples to differ from one another.

Treatment

The process or controlled circumstance applied to randomly assigned experimental units.

Trial

The sequence of outcomes representing events that we are simulating.

Under-coverage

A form of bias that samples in a way that part of the population gets less representation than it has in the population.

Voluntary Response Bias

Bias introduced to a sample because the individuals choose on their own to participate or not.