In AP Statistics, treatment groups are the groups of subjects or experimental units that each receive a different treatment (or level of a treatment) in an experiment, so researchers can compare responses across conditions and draw cause-and-effect conclusions.
A treatment group is a set of experimental units (people, plants, plots of land, whatever you're experimenting on) that all receive the same treatment, while other groups receive different treatments. The whole point of an experiment is comparison. If everyone got the same thing, you'd have nothing to compare, so you split your units into groups and deliberately impose different conditions on each one.
The key word is impose. In an observational study, you just watch what people already do. In an experiment, the researcher decides who gets what, and random assignment decides which units land in which treatment group. That randomization is what balances out lurking variables across groups and lets you make causal claims. A control group (one that gets no treatment, a placebo, or the standard treatment) often serves as one of the treatment groups, giving you a baseline to compare against.
Treatment groups live in Unit 3 of AP Statistics (Collecting Data), specifically in the topics on experimental design. The CED expects you to identify the experimental units, the treatments, the response variable, and how units are assigned to treatment groups, and to explain why random assignment matters. This is also where the famous AP Stats rule comes from. Well-designed experiments with random assignment to treatment groups support cause-and-effect conclusions; observational studies do not. That distinction shows up constantly in multiple choice and is a classic point in FRQ rubrics. Treatment groups also echo forward into inference (Units 6-7), because a two-sample test comparing treatment groups is how you actually analyze the experiment you designed back in Unit 3.
Keep studying AP® Statistics Unit 7
Control Group (Unit 3)
The control group is usually one of your treatment groups. It receives a placebo, no treatment, or the current standard, and it exists to give the other groups a baseline. Without it, you can't tell whether the new treatment actually did anything or whether everyone improved on their own.
Random Assignment (Unit 3)
Random assignment is how units get sorted into treatment groups. It spreads lurking variables roughly evenly across the groups, so any difference in the response can be attributed to the treatments. The 2017 FRQ literally asked about randomly assigning two men and two women into a treatment group and a control group, so know your assignment methods.
Experimental Design (Unit 3)
Treatment groups are the building blocks of every design you'll name on the exam. A completely randomized design assigns all units to treatment groups at random, a randomized block design forms treatment groups within each block, and a matched pairs design makes each unit (or pair) its own tiny comparison.
Null Hypothesis (Units 6-7)
When you analyze an experiment, the null hypothesis is usually that the treatment groups have the same true mean or proportion response. The whole inference machinery, like a two-sample t-test, exists to decide whether the observed difference between treatment groups is bigger than chance variation from random assignment.
Treatment groups appear in two main ways. In multiple choice, you'll identify the treatments and treatment groups in a described study, decide whether a study is an experiment or observational study, and judge whether a causal conclusion is justified. In FRQs, you design or critique experiments. The 2022 FRQ (Q5) described researchers comparing reductions in blood pressure across groups in a flavonoid study, and the 2017 FRQ (Q6) asked about randomly assigning four people to a treatment group and a control group, then reasoning about the probability of different group compositions. To earn points, you have to describe a concrete random assignment method (slips of paper, random number generator), name the treatments and response variable, and explain that comparing treatment groups under random assignment is what permits a cause-and-effect conclusion.
Every control group is a treatment group, but not every treatment group is a control group. 'Treatment groups' is the umbrella term for all the comparison groups in an experiment. The control group is the specific one that gets no treatment, a placebo, or the existing standard so the other groups have something to be measured against. An experiment can compare two active treatments with no control group at all and still be a valid experiment.
Treatment groups are the groups of experimental units that each receive a different treatment, and comparing their responses is the entire point of an experiment.
Units must be assigned to treatment groups by random assignment, which balances lurking variables and is what makes cause-and-effect conclusions possible.
A control group is just a special treatment group that provides a baseline, like a placebo or the standard treatment.
Treatments are imposed by the researcher; if subjects chose their own conditions, you have an observational study and no causal claim is allowed.
On FRQs, describe the random assignment method concretely (like numbering subjects and using a random number generator) instead of just saying 'randomly assign.'
The difference you observe between treatment groups gets tested later with inference, where the null hypothesis says the treatments produce equal responses.
Treatment groups are the groups of experimental units that receive different treatments (or different levels of a treatment) in an experiment. Comparing the response variable across treatment groups is how researchers measure the effect of each treatment.
Not exactly. A control group is one specific kind of treatment group, the one that receives a placebo, no treatment, or the standard treatment as a baseline. The other treatment groups receive the experimental conditions being tested.
It needs at least two conditions to compare, but they don't both have to be 'new' treatments and one doesn't have to be a control. An experiment comparing dose A versus dose B is still a valid experiment, and a matched pairs design can even apply both treatments to the same subjects.
By random assignment, such as numbering subjects and using a random number generator or drawing names from a hat. The 2017 AP Stats FRQ asked exactly this, with two men and two women randomly assigned to a treatment group and a control group of two people each.
A random sample is about how you select subjects from a population (it supports generalizing results). Treatment groups are about how you split subjects within an experiment (random assignment to them supports causal conclusions). An experiment can have random assignment to treatment groups without a random sample, and vice versa.
Connect this key term to the AP exam workflow: review the course, practice questions, and check related study tools.
Review units, study guides, and course resources.
Check this vocabulary in multiple-choice context.
Apply key concepts in written AP responses.
Estimate the exam score you are working toward.
Review the highest-yield facts before practice.
Put the full course together before test day.