Stat U4

Experiment

A study where researchers impose treatments to measure their effects and investigate cause and effect.

Response Variable

The variable that measures the outcome of a study (Y-variable).

Explanatory Variable

A variable that may explain or predict changes in the response variable (X-variable).

Factors

The explanatory variables manipulated by the experimenter.

Direct Control

Keeping extraneous variables constant across treatment groups.

Random Assignment

Using chance to assign experimental units to treatments.

Levels

The specific values or categories of each factor.

Treatment

A specific experimental condition or combination of factor levels.

Experimental Units

The smallest units to which treatments are applied.

Subjects

Experimental units that are human beings.

Observational Study

A study that observes and measures without imposing treatments.

Retrospective Study

An observational study using past events or existing data.

Prospective Study

An observational study that follows individuals into the future.

Survey

A study that collects self-reported information from a sample.

Extraneous Variable

Another variable that could influence the response variable.

Confounding Variable

A variable associated with the explanatory variable that also affects the response.

Random Samples

Samples selected using a chance process.

Simple Random Sample (SRS)

A sample where every possible group of n individuals has an equal chance of selection.

Sampling with Replacement

After selection, an individual is returned and can be selected again.

Sampling without Replacement

After selection, an individual is removed and cannot be selected again.

Strata

Non-overlapping groups whose members share similar characteristics.

Block / Blocking Variable

A block contains similar experimental units; a blocking variable defines those groups.

Stratified Sampling

Divide into similar strata and take an SRS from each stratum.

Cluster

A nearby, mixed subgroup that ideally represents the population.

Clustered Random Sample

Randomly select whole clusters and survey everyone in the selected clusters.

Randomization Distribution

Distribution of a statistic from shuffling treatment labels under the null hypothesis.

Systematic Sampling

Choose a random start, then select every kth individual.

Control Group

A comparison group receiving a placebo, standard treatment, or no treatment.

Placebo Effect

A response caused by believing a treatment is being received.

Placebo

An inactive treatment designed to resemble the real treatment.

Convenience Sample

Selecting individuals who are easiest to reach.

Bias

Systematic error that consistently favors certain results.

Randomized Block Design

Divide units into similar blocks and randomly assign treatments within each block.

Voluntary Response Bias

Bias when people choose to participate, often overrepresenting strong opinions.

Matched Pairs

A block design using matched pairs or the same subject receiving both treatments.

Undercoverage

Some population groups are left out or underrepresented in sampling.

Nonresponse

Selected individuals cannot be contacted or refuse to respond.

Statistically Significant

A result so extreme that it would rarely occur by chance if the null hypothesis were true.

Double Blind

Neither subjects nor people interacting with or evaluating them know treatment assignments.

Single Blind

Either subjects or the people interacting with/evaluating them do not know assignments.

Response Bias

Inaccurate survey answers caused by wording, interviewer influence, or dishonesty.

Completely Randomized Design

Assign all experimental units to treatments entirely by chance.

Replication

Apply each treatment to many experimental units to estimate response variability.