PBSI 245 Chapter 5: Experiments, Good and Bad

Vocabulary flashcards based on the lecture notes 'PBSI 245 Chapter 5: Experiments, Good and Bad' to help understand key terms related to experimental design and causal inference.

Non-experimental study

Measures variable(s) without intervention or treatment; also called observational designs.

Experiment

Measures variable(s) after imposing intervention or treatment, used to determine cause-and-effect relationships.

Response variable

Measures the results of a treatment; also known as outcome or dependent variable (DV).

Explanatory variable

What is thought to cause changes in the response variable; also known as predictor or independent variable (IV).

Treatment group / Control group

Refer to specific levels of the explanatory variable.

Subjects

The individuals participating in the sample of a study.

Quasi-experiments

Involve manipulation of an independent variable (sometimes naturally occurring) and the existence of groups/conditions, but lack random assignment to those groups.

Confounded variables

Variables whose effects on the outcome variable cannot be distinguished from each other.

Lurking variable

A potentially confounding variable that was not examined in a study.

Placebo effect

Subjects respond favorably to a bogus treatment, even one without active ingredients, or when active ingredients are present.

Control group (Placebo Effect Context)

A condition in which subjects do not receive the actual treatment or receive a dummy treatment (e.g., a placebo pill).

Randomized comparative experiment

Subjects are randomly assigned to one of two or more treatment groups.

Random assignment (in experiments)

Chance chooses which individuals are assigned to which treatment groups, assuming subject characteristics are equally distributed across groups.

Control (Principle of Experiment)

Removing effects of lurking variables by ensuring all subjects are equally affected by them.

Randomize (Principle of Experiment)

Ensuring chance chooses which individuals are assigned to which treatment groups.

Use enough subjects (Principle of Experiment)

Helps to reduce chance variation and sampling variability in results.

Statistical significance

An effect on the response variable that is of a size that would rarely occur by chance.