How experiments, randomization, and control help establish causation
Video Follow-Along Worksheet
↻ Already did this one? It stays open even after it’s checked off — just change any answer and click ✓ Check Answers again (no need to Reset). ✨ Grade with AI only ever raises your grade. Make sure you’re signed in so your changes save.
Learning Objectives:
Identify the components of an experiment, including experimental units, treatments, factors, and response variables.
Explain why observational studies cannot establish causation when confounding variables may be present.
Describe the four elements of a well-designed experiment: comparison, random assignment, replication, and control.
Explain the roles of control groups, placebos, and blinding in experiments.
Essential Knowledge:
Experimental units are the individuals assigned treatments; when they are people, they may be called participants or subjects.
An explanatory variable, or factor, is intentionally manipulated in an experiment, and its levels are treatments.
A response variable is the measured outcome after treatments are applied.
A confounding variable is related to the explanatory variable and influences the response variable, possibly creating a false perception of association.
A well-designed experiment includes comparison, random assignment, replication, and control of potential confounding variables.
Completely randomized designs assign treatments at random to all units, while randomized block designs randomize within similar blocks.
Placebos, control groups, and blinding help reduce bias and isolate treatment effects.
Key Vocabulary
Experiment
A study in which treatments are intentionally imposed on experimental units.
Experimental Unit
An individual or object that receives a treatment in an experiment.
Participant / Subject
A human experimental unit.
Factor
The explanatory variable that is intentionally manipulated in an experiment.
Treatment
A level or combination of levels of the explanatory variable.
Response Variable
The outcome measured after treatments are administered.
Confounding Variable
A variable related to the explanatory variable that also influences the response variable.
Comparison
Using at least two treatment groups so responses can be compared.
Random Assignment
Assigning treatments to experimental units by chance.
Replication
Using enough experimental units in each treatment group.
Control
Holding other variables as constant as possible to reduce confounding.
Control Group
A group that does not receive the treatment of interest or receives an inactive treatment.
Completely Randomized Design
A design in which all treatments are assigned completely at random to all units.
Randomized Block Design
A design in which units are first grouped into similar blocks, then randomized within each block.
Placebo
A fake treatment similar to the treatment being tested.
Placebo Effect
A response by experimental units to a placebo.
Double-Blind
Neither the subjects nor the interacting researchers know which treatment each subject receives.
Matched Pairs Design
A special case of a block design that uses pairs or has each subject receive both treatments.
Video 1: Introduction to Experimental Design
[0:00 – 6:21]
1.[0:04]
Topic 3.5 introduces the effect of variables and the basic components of an .
2.[0:37]
The opening example asks whether students who regularly take during AP Statistics earn higher grades.
3.[1:22]
Even if note-takers have a higher average grade, that does not automatically prove .
Take-Note Example[0:56 – 3:12]
The video compares students who take notes with students who do not. A difference in grades could be caused by note-taking, but it could also be explained by some other factor that affects both note-taking and grades.
4.[2:00]
A variable related to the explanatory variable that also influences the response variable is a variable.
5.[2:32]
In this example, academic may influence both note-taking and grades.
6.[3:12]
Because the original study only observes what students already do, it is an observational .
Components of an Experiment[3:43 – 4:20]
An experiment intentionally imposes treatments on experimental units.
If the experimental units are human, they may be called participants or subjects.
The explanatory variable is also called a factor, and its levels are treatments.
The response variable is the measured outcome of the study.
7.[3:43]
In an experiment, treatments are intentionally imposed on experimental .
8.[3:43]
When the experimental units are people, they may also be called participants or .
9.[4:04]
The explanatory variable in an experiment is also called a .
10.[4:20]
The response variable measures the of the study.
11.[5:11]
Letting students decide whether to take notes still allows confounding to be .
12.[5:59]
Observational studies cannot determine because confounding variables may still exist.
13.
Reflection: Students who regularly take notes in AP Statistics tend to earn higher grades than students who do not. Explain why this observational study cannot show that note-taking causes higher grades. Identify the explanatory variable, the response variable, and a possible confounding variable.
Video 2: Well-Designed Experiments
[0:00 – 6:34]
14.[0:29]
A well-designed experiment must compare at least two treatment .
15.[0:50]
Random assignment helps possible confounding factors across treatment groups.
16.[1:08]
means using enough experimental units in each treatment group.
17.[1:33]
Researchers should also potential confounding variables as much as possible.
Principles of a Well-Designed Experiment[0:29 – 1:33]
Comparison: use at least two treatment groups.
Random assignment: assign treatments to units by chance.
Replication: use enough experimental units in each group.
Control: hold other variables as constant as possible.
Bullseye: Botswana Cattle Experiment[1:40 – 3:59]
Cattle were randomly assigned to receive painted eye spots, painted cross marks, or no markings. None of the 683 cattle with eye spots were attacked, while 4 cross-marked cattle and 19 unmarked cattle were killed by lions or leopards.
18.[2:23]
The cattle were randomly assigned to receive eye spots, cross marks, or remain .
19.[2:59]
Because treatments were randomly assigned, the Botswana cattle study was an .
20.[4:06]
One method for random allocation is to use a random number .
21.[5:34]
After treatments are assigned, researchers compare the variable across groups.
Video 3: Experimental Designs and Methods
[0:00 – 10:13]
Melanoma Treatment Study[0:37 – 2:56]
The video describes a randomized, placebo-controlled, double-blind experiment comparing a standard treatment with a new treatment combination. The response variable was recurrence-free survival.
22.[1:36]
In a completely randomized design, treatments are assigned completely at .
23.[3:13]
In a randomized block design, the units within each block should be with respect to the blocking variable.
24.[4:16]
After units are placed into blocks, treatments are randomized each block.
25.[5:28]
A placebo is a treatment that is similar to the treatment being tested.
26.[7:40]
In a double-blind experiment, neither the subjects nor the who interact with them know which treatment is being given.
Design Vocabulary from Video 3[2:20 – 9:30]
Completely randomized design: all units are randomly assigned to treatments.
Randomized block design: group similar units into blocks, then randomize treatments within each block.
Placebo: a fake treatment used to help detect whether the treatment effect is real.
Single-blind: either the subjects or the interacting researchers do not know the treatments.
Double-blind: neither the subjects nor the interacting researchers know the treatments.
Matched pairs design: a special case of a block design that uses pairs or gives each subject both treatments.
27.
Reflection: Explain the difference between a completely randomized design and a randomized block design. Include what blocking does and how treatments are assigned in each design.
Exit Ticket
A dermatologist wants to test whether a new anti-itch cream works better than a placebo cream for patients with eczema. Patients are first separated into two groups based on severity: mild eczema and severe eczema. Within each severity group, patients are randomly assigned to receive either the new cream or an identical-looking placebo cream. Neither the patients nor the nurses who rate the symptoms know which cream each patient receives.
(a) Identify the experimental units.
(b) Identify the explanatory variable, the treatments, and the response variable.
(c) Explain how blocking and random assignment are used.
(d) Identify the control group and explain why this is a double-blind experiment.
Class Responses (Esc to close)
Click a "Class" button next to a question to see responses.