Chapter 4: Sampling and Experimental Design (Lessons 4.1a–4.3)

Vocabulary flashcards covering key terms and concepts from sampling and experimental design topics (population, sampling methods, bias, designs, and inference).

Population

The entire group of individuals or items of interest from which we want to draw conclusions.

Sample

A subset of individuals selected from a population to study and infer about the population.

Census

Data collected on every member of the population.

Simple Random Sample (SRS)

A sample in which every individual has an equal chance of being selected, and every possible sample of the same size is equally likely.

Biased Sampling

A sampling method that systematically overestimates or underestimates the population parameter.

Nonresponse bias

Bias that arises when individuals selected do not respond, and respondents differ from nonrespondents.

Undercoverage

When a portion of the population is not represented in the sample.

Voluntary response bias

Bias that occurs when individuals self-select to participate, often attracting those with strong opinions.

Wording bias

Bias caused by the way questions are phrased or ordered.

Stratified Random Sample

Population divided into strata, a random sample drawn from each stratum, then combined.

Cluster Random Sample

Population divided into clusters; a random set of clusters is chosen; all individuals in those clusters are included.

Systematic Random Sample

Start at a random point and select every kth individual on a list.

Blocking

Grouping experimental units into blocks of similar size or characteristics to reduce variation.

Randomized Complete Block Design

Experimental units are blocked by a factor; within each block, treatments are randomly assigned.

Completely Randomized Design

Experimental units are assigned to treatments completely at random, with no blocks.

Matched Pairs Design

Pairs of similar experimental units are formed; within each pair, treatments are randomly assigned.

Treatments

The different levels or types of the explanatory variable applied in an experiment.

Explanatory Variable (Factor)

The variable purposely manipulated to observe its effect on the response.

Response Variable

The outcome measured and compared across treatment groups.

Control

A baseline treatment used for comparison to assess the treatment effect.

Placebo

An inactive treatment used to blind subjects and control for placebo response.

Placebo Effect

Improvement due to belief in the treatment rather than the treatment itself.

Blinding

A design where either participants, researchers, or both are unaware of which treatment is given.

Single-blind

Only the subjects or the researchers are unaware of treatment assignment.

Double-blind

Neither the subjects nor the researchers know who receives which treatment.

Experimental units

The smallest unit to which a treatment is applied (person, animal, etc.).

Random Assignment

Randomly assigning units to treatments to balance confounding variables.

Replication

Repeating the experiment with multiple units per treatment to reduce sampling error.

Confounding

A situation where the effects of two variables on the response cannot be distinguished.

Observational study

A study that measures variables without applying treatments; cannot establish causation.

Retrospective study

An observational study that looks backward in time at existing data.

Prospective study

An observational study that follows individuals forward in time to observe outcomes.

Generalization

Extending conclusions from a sample to a larger population.

Inference

Drawing conclusions about a population from sample data; what the data allow us to claim.

Statistical significance

An observed effect so large that it would be unlikely to occur by chance.

Line of Table D

A method using a table of random digits to select an SRS by choosing a random start and moving along lines.

Hat method

An SRS method where names are written on slips of paper in a hat and drawn.

Technology method

An SRS method where each member is assigned a number and a random number generator selects the sample.

Digits method

Using random digits with equal-length labels to select a sample; ensure uniform digit lengths to avoid bias.

Convenience Sample

A sample selected by choosing individuals who are easiest to reach; often leads to biased results.