Glossary
Blocking Design
An experimental design where experimental units are first grouped into 'blocks' based on a characteristic known or suspected to affect the response, and then treatments are randomly assigned within each block.
Example:
When testing a new teaching method, students are first grouped by their current GPA (high, medium, low) to create blocks, and then within each GPA group, students are randomly assigned to the new method or the traditional method.
Blocking Variable
A characteristic used to group experimental units into blocks in a blocking design, chosen because it is known or suspected to influence the response variable.
Example:
In a study comparing two pain relievers, patient age might be a blocking variable if older and younger patients are expected to respond differently to the medication.
Completely Randomized Design
An experimental design where all experimental units are assigned to treatment groups entirely by chance, ensuring each unit has an equal probability of being in any group.
Example:
To test a new energy drink, a researcher randomly assigns 100 volunteers to either receive the new energy drink or a placebo, without considering any other factors.
Confounding Variable
An extraneous variable that influences both the dependent variable and the independent variable, making it difficult to determine the true effect of the independent variable on the dependent variable.
Example:
If a study on coffee's effect on alertness doesn't account for participants' sleep habits, sleep could be a confounding variable because it affects both coffee consumption (perhaps tired people drink more) and alertness.
Control Group
A group of experimental units that does not receive the active treatment being studied, often receiving a placebo or the standard existing treatment, serving as a baseline for comparison.
Example:
In a clinical trial for a new medication, the control group might receive a sugar pill instead of the actual drug.
Experimental Group
A group of experimental units that receives the specific treatment being tested in an experiment.
Example:
In a study comparing a new teaching method to a traditional one, the students taught with the new method constitute the experimental group.
Experimental Units
The individuals or objects on which an experiment is performed and to which treatments are applied.
Example:
In a study testing different fertilizers, the individual tomato plants receiving the fertilizer are the experimental units.
Matched Pairs Design
A special type of blocking design used to compare two treatments, where units are paired based on similarity (or the same unit receives both treatments), and then treatments are randomly assigned within each pair.
Example:
To compare two types of running shoes, each runner tries both shoe types, with the order of shoes worn being randomized for each runner. This uses the runner as their own matched pair.
Randomization
The process of using chance to assign experimental units to treatment groups or to determine the order of treatments, which helps balance out lurking variables and ensures unbiased results.
Example:
Flipping a coin to decide which of two new apps a participant will test first is an example of randomization.
Treatments
The specific conditions or interventions applied to the experimental units in an experiment, which are designed to measure a response.
Example:
If a study investigates the effect of different amounts of sleep on test scores, the varying hours of sleep (e.g., 6 hours, 8 hours, 10 hours) are the treatments.
Variability
The extent to which data points in a set differ from each other or from the mean; in experimental design, reducing variability within treatment groups leads to more precise results.
Example:
A researcher uses a matched pairs design to reduce variability caused by individual differences when comparing two types of exercise routines.