Data Collection and Experimental Design Vocabulary

Vocabulary flashcards covering key terms for data collection, sampling methods, experimental design, and data ethics.

Experiment

Study where researchers give treatments and measure results.

Experimental unit

Person or thing that gets the treatment.

Subject

A person in an experiment.

Treatment

A condition given to an experimental unit.

Factor

Variable that researchers change in an experiment.

Level

A specific value of a factor.

Response variable

What is measured as the result.

Explanatory variable

Variable used to explain or predict the response.

Observational study

Study where researchers observe without giving treatments.

Retrospective study

Study that looks at past information.

Prospective study

Study that collects information going forward.

Survey

Study that collects information by asking people questions.

Extraneous variable

Other variable that could affect the result.

Confounding variable

Outside variable connected to both the explanatory and response variables.

Random sampling

Choosing people/items using chance.

Investigative question

Question that includes the variable being studied, type of inference, and type of conclusion.

Simple Random Sample (SRS)

Sample where every possible sample of the same size has an equal chance of being chosen.

Without replacement

Sampling condition where a person/item cannot be chosen again.

With replacement

Sampling condition where a person/item can be chosen again.

Stratified random sample

Sampling method where people are split into groups and some are randomly chosen from every group.

Stratum

One group in a stratified sample.

Cluster random sample

Sampling method where people are split into groups and whole groups are randomly chosen.

Cluster

A whole group chosen in a cluster sample.

Systematic random sample

Sampling method where you pick a random starting point, then choose every kth person.

Sampling variability

The concept that different samples can give different results.

Convenience sample

Choosing people who are easiest to reach.

Bias

A sampling problem that causes results to be too high or too low.

Voluntary response sample

Sample where people choose whether to participate.

Voluntary response bias

Bias resulting when people who choose to respond have stronger opinions.

Undercoverage

Condition where some people have little or no chance of being chosen.

Undercoverage bias

Bias occurring because the people left out may be different from the rest.

Nonresponse

Condition where a selected person does not respond.

Nonresponse bias

Bias occurring when people who do not respond may be different from the population.

Response bias

Bias resulting when people give answers that are different from the truth.

Control group

Group used for comparison to see if the treatment caused a change.

Placebo

Fake treatment with no active ingredient.

Placebo effect

Feeling better even after receiving a placebo.

Blinding

Keeping people from knowing which treatment was given.

Single-blind

Study design where either the subjects or researchers don't know the treatment.

Double-blind

Study design where subjects and researchers don't know the treatment.

Random assignment

Using chance to assign treatments.

Completely randomized design

Experimental design where treatments are randomly assigned with no restrictions.

Replication

Using multiple subjects/units for each treatment.

Direct control

Keeping certain outside variables the same to prevent confounding and reduce variation.

Block

Group of similar experimental units.

Randomized block design

Experimental design where researchers make similar groups, then randomly assign treatments within each group.

Matched pairs design

Experimental design where two treatments are given to similar pairs or to the same person in random order.

Statistically significant

A result that is unlikely to be caused by chance.

5% significance level

Less than a $5\%$ chance of getting the result randomly.

Institutional Review Board (IRB)

Group that checks a study is safe for participants before the study begins.

Informed consent

Participants agree to take part after being told about the study.

Confidentiality

Keeping participants' personal information private.