Introduction to Statistics and Experimental Design

Question-and-answer flashcards covering key terminology, experimental components, design types, and illustrative examples from Chapter 1 on statistics and data collection.

What is an explanatory variable in an experiment?

The variable that is intentionally manipulated by the researcher to observe its effect on the response variable.

What is a response variable in an experiment?

The outcome or measurement that is recorded to assess the effect of the explanatory variable.

In experimental terminology, what is an individual (or subject)?

A single person or object upon which data are collected.

What major advantage do experiments have over observational studies?

They allow researchers to control conditions and establish cause-and-effect relationships.

Define a lurking variable.

A variable not included in the study that can influence both the explanatory and response variables, potentially misleading conclusions.

Define a confounding variable.

A variable that is related to both the explanatory and response variables and whose effects are indistinguishable from those of the explanatory variable.

In the folic-acid pregnancy study, what is the explanatory variable?

Whether the pregnant woman took the folic-acid supplement or the placebo.

In the folic-acid pregnancy study, what is the response variable?

The birth-defect rate among the newborns.

Give one plausible lurking variable in the folic-acid pregnancy study.

Examples include overall maternal nutrition, prenatal care quality, or socioeconomic status.

Define a factor in an experiment.

A categorical explanatory variable whose levels are deliberately set by the experimenter.

Define a treatment in an experiment.

A specific combination of factor levels administered to the experimental units.

What is an experimental unit?

The smallest entity to which a treatment is applied (often called a subject when it is a human participant).

What is the purpose of a control group?

To provide a baseline for comparison, helping to isolate the effect of the treatment.

What is a placebo?

An inert or inactive treatment given to the control group to mimic the experience of the treatment group.

Differentiate single-blind and double-blind designs.

Single-blind: subjects do not know which treatment they receive. Double-blind: neither subjects nor those measuring the response know which treatment was given.

In the blood-pressure drug experiment, what is the explanatory variable?

Whether the patient receives the new drug or the placebo.

In the blood-pressure drug experiment, what is the response variable?

The patient’s blood-pressure readings over the eight-week period.

Who are the experimental units in the blood-pressure experiment?

The 200 hypertension patients.

Identify the factor and its levels in the blood-pressure experiment.

Factor: medication type; Levels: new drug and placebo.

Is the blood-pressure experiment blinded? If so, how?

Yes, it is single-blind—a double-blind design could also apply if researchers measuring blood pressure are unaware of group assignments.

Name one possible lurking variable in the blood-pressure experiment.

Examples: patients’ baseline diet, exercise habits, or medication adherence.

Describe a completely randomized design.

All experimental units are randomly assigned to treatments without regard to any blocking variable.

Describe a randomized block design.

Experimental units are first divided into homogeneous blocks, then randomly assigned to treatments within each block.

Describe a matched-pairs design.

Pairs of similar units (or the same unit measured twice) receive different treatments; responses are compared within each pair.

Which design type best fits the store marketing-strategy experiment?

Completely randomized design (stores are randomly assigned to Strategy A or B).

Which design type best fits the job-role productivity experiment?

Randomized block design (employees blocked by job role, then assigned to work location).

Which design type best fits the paired employee training experiment?

Matched-pairs design (employees paired on pre-training scores, each pair split across Training A and B).