HSS.IC.B.3Common CoreMathStatistics and ProbabilityGrades 9-12
HSS.IC.B.3: Sample Surveys, Experiments and Observational Studies
In plain English: HSS.IC.B.3 is the Common Core statistics standard that asks students to tell sample surveys, experiments and observational studies apart by purpose and design, and to explain what randomization does in each: random selection lets a survey speak for a population, and random assignment lets an experiment show cause and effect. It is usually taught in Algebra II or Statistics.
Recognize the purposes of and differences among sample surveys, experiments, and observational studies; explain how randomization relates to each.
Common Core State Standards for Mathematics · Domain: Making Inferences and Justifying Conclusions (IC) · Cluster: Make inferences and justify conclusions from sample surveys, experiments, and observational studies Also written as HSS-IC.B.3 or S-IC.3 · Official standard
Statistical studies answer different kinds of questions, and the kind of question decides the design. A sample survey estimates something about a population, such as the proportion of students who favor a later start time. An experiment tests whether a treatment causes a change, such as whether a study method raises quiz scores. An observational study examines a relationship as it happens, without assigning anything, such as whether students who play a sport have different grades from those who do not.
The lesson centers on randomization, which plays a different role in each design. Random selection makes a sample representative, so its results can be generalized to the population. Random assignment makes treatment groups alike in everything except the treatment, so a difference in results can be attributed to the treatment. Observational studies can use random selection but not random assignment, which is why they show association and not causation.
Learning Objectives
By the end of this lesson, students will be able to:
State the purpose of a sample survey, an experiment and an observational study
Classify a study as a sample survey, an experiment or an observational study and justify the choice
Explain the difference between random selection and random assignment, and which design uses each
Carry out random selection and random assignment with a random digit table or random number generator
Decide what conclusion a study supports: generalization to a population, cause and effect, or association only
Prior Knowledge Required
Students should already be comfortable with:
Using a random sample to make inferences about a population 7.SP.A.1
Populations, parameters, samples and statistics HSS.IC.A.1
The difference between correlation and causation HSS.ID.C.9
Show the headline and ask students to write two ways a researcher could have produced it:
Warm-Up Prompt
"A headline says: 'Teens who eat breakfast get better grades.' Describe one way researchers might have collected the data. Based on your description, could they say that eating breakfast causes better grades? What would they have to do differently to find out?"
Students typically describe asking teens about breakfast and grades. Ask what else might differ between teens who eat breakfast and those who don't (sleep, family routines, time to get ready). Some students will suggest making one group eat breakfast; name that idea as an experiment. Tell the class that today's lesson sorts studies into three kinds and explains what each kind is allowed to conclude.
Direct Instruction20 minutes
Introduce the three designs with Diagram 1, then the summary table. Stress the question to ask first: did the researcher impose a treatment? If yes, it is an experiment. If no, it is a sample survey (when the goal is to estimate a population value) or an observational study (when the goal is to compare groups or relate variables as they occur).
The three study designs and the role of randomization
Design
Purpose
Randomization used
What it allows
Sample survey
Estimate a population value (a proportion or mean)
Random selection of who is asked or measured
Generalizing from the sample to the population
Experiment
Decide whether a treatment causes a change in a response
Random assignment of subjects to treatments
A cause-and-effect conclusion for the subjects
Observational study
Describe a relationship without influencing it
Random selection when possible; no random assignment
Evidence of an association, not causation
Sample survey
The student council wants the proportion of the school's 1,500 students who favor a later start time. It randomly selects 120 names from the enrollment list, and 78 of those students favor the change.
Equation: p̂ = 78/120 = 0.65. Random selection lets the council estimate that about 65% of the school's students favor a later start.
Experiment
Thirty student volunteers are randomly assigned, 15 to study vocabulary with a flashcard app and 15 with a printed list. The app group's mean quiz score is 17.2 and the list group's is 15.6.
Equation: Difference 17.2 - 15.6 = 1.6 points. Random assignment made the groups alike except for the method, so a difference larger than chance would explain points to the app as the cause, for students like these volunteers.
Observational study
From school records, a researcher compares 85 varsity athletes with 140 students who play no sport. The athletes' mean GPA is 3.3; the others' is 3.1.
Equation: Difference 3.3 - 3.1 = 0.2. Nobody was assigned to play, so other variables (time management, family support, eligibility rules) could explain it: association, not causation.
Choosing a design
Match each question to a design: (a) What percent of a city's voters approve of the mayor? (b) Does a new fertilizer increase tomato yield? (c) Do teens who sleep less have lower grades?
Equation: (a) sample survey; (b) experiment; (c) observational study, because assigning teens to sleep less would be unethical.
Carrying out random selection
Choose 4 of 60 students labeled 01-60, using the random digit line 81 47 03 47 92 15 66 38 29 54. Skip numbers above 60 and repeats.
Equation: Selected: 47, 03, 15, 38 (skip 81, the repeated 47, 92 and 66).
Use Diagram 2 to separate the two kinds of randomization. Random selection answers "who is in the study?" and supports generalizing. Random assignment answers "who gets which treatment?" and supports cause and effect. An experiment on volunteers uses assignment but not selection, so its causal conclusion applies to subjects like the volunteers; a survey uses selection but has no treatments, so it cannot show causation.
Guided Practice15 minutes
Pairs classify each study, name the randomization it uses, and state the strongest conclusion it allows. Discuss each one before moving on.
A cell phone company randomly selects 400 of its customers and asks how satisfied they are with their service. (Sample survey; random selection; estimate for all customers.)
A physical therapist randomly assigns 36 patients with knee injuries to two stretching routines and measures recovery time. (Experiment; random assignment; the routine can be the cause of a difference.)
Researchers compare the reading scores of children who attended preschool with those who did not, using a random sample of district records. (Observational study; random selection only; association.)
A teacher lets students choose whether to use a practice website, then compares test scores of users and non-users. (Observational study, even though it compares groups: the students chose, so motivated students may be over-represented among users.)
A food scientist randomly assigns 50 tasters to try one of two recipes and rate them. (Experiment.)
Listen for students who label any comparison of two groups an experiment. Ask, "Who decided which group each person was in?"
Independent Practice15 minutes
Question for students: does listening to music while doing homework affect accuracy? Tasks: (1) Describe an observational study that could explore the question and one weakness of it. (2) Describe an experiment with 24 volunteers labeled 01-24. (3) Use the random digit line 19 04 88 31 12 04 23 07 56 15 02 19 10 21 44 08 17 93 14 06 to assign 12 volunteers to the music group, skipping numbers above 24 and repeats. (Answer: 19, 04, 12, 23, 07, 15, 02, 10, 21, 08, 17, 14; the other 12 form the no-music group.) (4) Explain what random assignment protects against in your experiment. (5) Explain why a survey that asks students whether they listen to music could not answer the question.
Closure5-10 minutes
Exit ticket: (1) Name the design that can show cause and effect, and the kind of randomization that makes it possible. (2) A study compares the health of people who chose to become vegetarians with that of people who did not. What kind of study is it, and why can't it prove that diet causes the difference? (3) Give an example of a question best answered by a sample survey.
Differentiation Strategies
For Struggling Students
Give a two-question flowchart: Did the researcher impose a treatment? If not, is the goal to estimate a population value or to compare groups as they are?
Color-code the two kinds of randomization on every example: one color for random selection, another for random assignment
Start classification practice with studies that name the groups explicitly before moving to studies described in a news-story style
For Advanced Students
Introduce blocking: design an experiment on a study method that first separates students by grade level, then randomly assigns within each grade, and explain the benefit
Ask students to explain why a double-blind design matters in an experiment on an energy drink, and what it adds to random assignment
Have students find a news article about a health study, classify it, and judge whether the headline's wording matches the design
Assessment Guidance
What to Look For
The key discriminator is whether the researcher imposed a treatment: students should use it to separate experiments from observational studies, and should not call a study an experiment just because it compares two groups. Look for precise language about randomization: "random selection" for who is studied and "random assignment" for who gets which treatment. Strong answers match the conclusion to the design: estimates for a population from a random sample, cause and effect only from random assignment, and association only from observational data.
02
Classroom Activities
3 Activities
1
Study-Type Card Sort
20 minGroups of 3-4
Groups sort 9 study cards into three columns (sample survey, experiment, observational study), then add the randomization each study uses and the conclusion it supports.
The 9 Cards
A newspaper randomly selects 800 registered voters in the state and asks which candidate they prefer (survey)
A school randomly chooses 50 of its laptops and records their battery health (survey)
A park service randomly selects 200 visitors at the entrance and asks how far they traveled (survey)
Forty volunteers are randomly assigned to drink water or a sports drink before a 1-mile run, and their times are compared (experiment)
A farmer randomly assigns 20 plots of a field to two irrigation schedules and measures the corn harvested (experiment)
A teacher randomly assigns half of her four classes to receive quizzes on paper and half online, then compares scores (experiment, with classes as the units)
Researchers compare asthma rates of children living near highways with those living farther away (observational)
A study follows 1,000 randomly selected adults for 10 years and records how much coffee each drinks and whether they develop heart disease (observational)
A coach compares the injury rates of players who chose to do a warm-up routine with those who skipped it (observational)
Procedure
Sort the cards and write on each: who decided the groups, if there are groups, and whether randomization was used for selection, assignment or neither
For each observational card, name one variable that could explain the relationship other than the one being studied
For one observational card, rewrite it as an experiment, or explain why an experiment would be unethical or impractical
Modification for Distance Learning
Put the cards on a shared slide with three drop zones. Groups drag the cards in a breakout room and add sticky notes with their reasons.
2
Does Random Assignment Balance the Groups?
20 minGroups of 4
Before running an experiment on whether a 5-minute breathing exercise before a quiz improves scores, the class checks whether random assignment makes the two groups alike on a variable that could matter: hours of sleep last night.
Procedure
Each student writes last night's hours of sleep on an index card (or use the invented set below for a class of 20)
Shuffle the cards and deal them into two groups of equal size. Compute each group's mean hours of sleep and the difference
Repeat the shuffle 4 more times and record the 5 differences on the board
Discuss what would happen if students chose their own group, for example if well-rested students were more willing to try something new
Teacher Key for the Invented Data
Invented sleep hours for 20 students, numbered 1-20: 6, 7.5, 8, 5.5, 7, 9, 6.5, 7, 8.5, 6, 7, 7.5, 5, 8, 6.5, 7, 9.5, 6, 7.5, 8 (mean 7.15 hours). One random split put students 1, 2, 6, 7, 10, 12, 16, 17, 18, 20 in group A: mean 7.30 hours, versus 7.00 for group B, a difference of 0.3. Four more random splits gave differences of 0.2, -0.1, -0.2, -0.6. The differences are small and change sign, so no group is favored on average.
Discussion Questions
Random assignment did not make the group means exactly equal. Why is it still useful?
Sleep is one variable you measured. What about variables you did not measure?
How would bigger groups change the size of the differences?
3
One Question, Three Studies
25 minGroups of 3
Each group gets one question and designs all three kinds of study for it, then presents which design best answers the question and why. The activity makes the purpose of each design concrete.
Questions to Assign
Is using a phone in the hour before bed related to how long teens sleep?
Does a short video lesson help students learn to solve a type of puzzle faster than a written explanation?
What proportion of the school's students would use a new after-school tutoring room?
Are students who take music lessons more likely to take an advanced math class?
Procedure
For your question, describe a sample survey, an observational study and an experiment. For each, say who is studied, how randomization is used and what could be concluded
Mark any design that would be unethical or impractical and explain why
Choose the design that best fits the question and present it in 2 minutes
Other groups score each presentation on a checklist: correct design names, correct randomization, conclusion matches design
Challenge Variation
Groups add a second version of their experiment that uses blocking (for example, by grade level) and explain when blocking gives a clearer comparison than complete randomization.
03
Diagrams & Visual Aids
2 diagrams
Diagram 1: Three Kinds of Statistical Study
Each design has its own purpose and its own use of randomization. Only an experiment imposes treatments, and only random assignment supports a cause-and-effect conclusion. Random selection is what lets a sample survey, or an observational study that uses it, speak for a larger population.
Diagram 2: Random Selection Versus Random Assignment
Top: random selection chooses who is studied, here 120 of the school's 1,500 students, so results generalize to the population. Bottom: random assignment splits 40 volunteers into two groups of 20 to receive different treatments, so a difference in responses can be attributed to the treatment.
04
Homework Assignment
~30 min
HSS.IC.B.3 Homework: Surveys, Experiments and Observational Studies
Directions: For every study, name the design, say what randomization it uses (random selection, random assignment, both or neither) and state the strongest conclusion it supports. Show your work for any computation. All data are invented for practice.
Part 1: Purposes and Differences (Problems 1-3)
Classify each study as a sample survey, an experiment or an observational study, and name the randomization it uses. (a) A city randomly selects 500 households from its utility records and asks how many people live in each, to estimate mean household size. (b) A botanist randomly assigns 24 pepper plants to two watering schedules and measures the weight of peppers each produces. (c) Researchers review the records of 2,000 randomly chosen adults and compare the heart health of those who drink coffee daily with those who do not.
A school district wants to know whether a new after-school tutoring program raises algebra test scores. One board member suggests comparing the scores of students who signed up for the program with those who did not. Explain why this is an observational study, what could go wrong with its conclusion, and how an experiment would answer the question better.
A town randomly selects 250 of its 5,000 adult residents and asks whether they would use a proposed bus line; 95 say yes. (a) Compute the sample proportion and estimate the number of adult residents who would use the bus line. (b) What role did randomization play? (c) Could this survey show that building the bus line would reduce traffic? Explain.
Part 2: Randomization in Practice (Problems 4-6)
Sixteen volunteers, labeled 01-16, will test a new sunscreen against a standard one. (a) Use the random digit line 07 22 13 07 19 02 15 11 04 16 09 30 01 to choose 8 volunteers for the new sunscreen, skipping numbers above 16 and repeats. (b) Why is this better than letting volunteers choose which sunscreen to use?
In a random sample of 300 adults, the 120 who walk at least 30 minutes a day have a mean resting heart rate of 68 beats per minute, and the 180 who do not have a mean of 74. (a) What is the difference in means? (b) Is this a survey, an experiment or an observational study? (c) Can we conclude that walking lowers heart rate? Name a variable that could explain the difference. (d) What does the random selection of the 300 adults allow?
A student wants to know whether chewing gum during a test improves scores. (a) Design an experiment with 40 volunteers: describe the treatments, how you would randomly assign them and what you would measure. (b) Explain why a survey asking students whether they chew gum and what grades they get could not answer the question.
Rubric
Criterion
Full Credit (2 pts)
Partial Credit (1 pt)
No Credit (0 pts)
Identifying the Design
Every study classified correctly with a reason
Most classified correctly or reasons missing
Mostly incorrect
Role of Randomization
Random selection and random assignment named correctly and explained
Randomization named but roles confused
Randomization missing or incorrect
Conclusions
Every conclusion matches the design (population, cause, or association)
One conclusion overstated
Causal claims from non-experiments
Carrying Out and Designing
Correct selection from the digit line and a complete experimental design
Minor error in selection or design
Missing or incorrect
05
Quiz: 20 Questions
Interactive, with answers
Instructions
Answer each question, then read the explanation. Your score updates as you go, and Reset quiz clears everything so you or your students can try again.
Multiple choice: pick an option to check it. Short answer: write your answer, then reveal the model answer.
0 of 20 answered · 0 correct
Question 1 of 20 · Multiple Choice
Which of these studies is an experiment?
Answer: C
Only in choice C does the researcher impose the treatments (standing desk or regular desk), and assignment is random. Choice A is a sample survey. Choice B compares groups, but the students chose their own course, so it is an observational study. Choice D is also observational.
Question 2 of 20 · Multiple Choice
What is the main purpose of a sample survey?
Answer: A
A sample survey collects data from a sample to estimate a population value. Choice B is the purpose of an experiment. Choice C describes an experiment's design, and choice D describes a census, not a sample.
Question 3 of 20 · Multiple Choice
Researchers track 1,200 adults for 15 years, recording how often each eats fish and whether each develops memory problems. They find that frequent fish eaters had fewer memory problems. What kind of study is this?
Answer: B
The adults chose their own diets; researchers only observed and recorded. That makes it an observational study, and it shows an association, not causation. Choice A is a common error: comparing groups does not make a study an experiment. Neither size nor length decides the design.
Question 4 of 20 · Multiple Choice
In an experiment, what is the purpose of random assignment?
Answer: D
Random assignment spreads other variables, measured or not, evenly across groups on average, so a difference in results can be attributed to the treatment. Choice A describes random selection. Choice B overstates: groups are similar on average, not identical, which is why chance variation must be considered. Choice C would undo the point of assignment.
Question 5 of 20 · Multiple Choice
In a sample survey, how is randomization used?
Answer: C
Random selection gives each member of the population a known chance of being included, so results can be generalized. Choice A describes an experiment. Choice D is false: without random selection, a survey's results may not represent the population.
Question 6 of 20 · Multiple Choice
A well-designed observational study finds that people who own a dog have lower blood pressure on average. What conclusion is justified?
Answer: A
Without random assignment, other variables (for example, activity level or income) could explain the difference, so only an association is justified. Choices B and C claim causation. Choice D goes too far: observational studies can show associations and suggest questions for experiments.
Question 7 of 20 · Multiple Choice
An experiment randomly assigns 50 volunteers from one high school to two study methods and finds a clear difference in scores. To whom does the cause-and-effect conclusion apply?
Answer: C
Random assignment supports a causal conclusion, but the subjects were not randomly selected from a larger population, so the conclusion applies to these volunteers and similar students. Choice A requires random selection from all students. Choice D is too strict: experiments with volunteers are common and valid for the subjects.
Question 8 of 20 · Multiple Choice
Why do researchers use observational studies, not experiments, to study whether heavy teen smartphone use is linked to anxiety?
Answer: B
When a treatment might be harmful, researchers cannot ethically impose it, so they observe people who already differ. Choice A reverses the strength of evidence: a randomized experiment is the design that supports causation. Choice C is false, and choice D is about cost, not design.
Question 9 of 20 · Multiple Choice
An observational study finds that students who eat lunch in the cafeteria have higher attendance than students who leave campus. Which is a possible confounding variable?
Answer: D
A confounding variable is related to both the explanatory variable (where students eat) and the response (attendance). Having a car could make leaving campus and missing class both more likely. Choice A is the sample size, not a confounder, and choice C is unrelated to either variable.
Question 10 of 20 · Multiple Choice
Students are labeled 01-40. Using the random digit line 33 05 71 05 18 64 29, and skipping numbers above 40 and repeats, which 3 students are selected?
Answer: C
Read pairs in order: 33 (keep), 05 (keep), 71 (skip, above 40), 05 (skip, repeat), 18 (keep). Choice A keeps 71, which is not a label. Choice B keeps the repeated 05. Choice D skips 33 for no reason.
Question 11 of 20 · Multiple Choice
A county randomly selects 400 of its 8,000 households, and 148 of them have a home garden. What is the best estimate of the number of households in the county with a home garden?
Answer: B
p̂ = 148/400 = 0.37, and 0.37 × 8,000 = 2,960 households. Random selection is what justifies applying the sample proportion to the whole county. Choice A is the sample count. Choice C multiplies 0.37 by 10,000 instead of 8,000. Choice D is the proportion, not the number of households.
Question 12 of 20 · Multiple Choice
Which design best answers the question "Does a new reading app increase reading speed for ninth graders?"
Answer: D
The question asks whether the app causes a change, and only a randomized experiment supports that conclusion. Choice B is an observational study: students who chose the app may already read more. Choices A and C do not compare a treatment at all.
Question 13 of 20 · Multiple Choice
Which study uses both random selection and random assignment?
Answer: A
Choice A chooses the 200 students from the district by chance and then uses chance again to decide which policy each one gets. Choice B uses random selection only (a sample survey). Choice C uses random assignment only, because the subjects are volunteers. Choice D uses neither: the students chose their own groups.
Question 14 of 20 · Multiple Choice
A study compares the grades of students who chose to join a school's robotics club with those of students who did not, and finds higher math grades among club members. Which statement is correct?
Answer: C
Students selected themselves into the club, and students who already like math may be more likely to join, so the groups may differ in interest in math and other ways; this is an observational study that shows association only. Choices A and D treat a self-chosen activity as an imposed treatment. Choice B misnames the design and overgeneralizes.
Question 15 of 20 · Short Answer
A company randomly selects 60 of its 900 employees and asks how many days a week they work from home. Classify the study, state its purpose, and explain the role of randomization.
It is a sample survey. Its purpose is to estimate a population value: the mean number of work-from-home days for all 900 employees. Random selection makes the 60 employees likely to represent all 900, so the sample mean can be used as an estimate for the company. No treatment is imposed.
Question 16 of 20 · Short Answer
Thirty volunteer runners will test whether a new running shoe lowers their 5K times compared with their usual shoes. Describe how to carry out the random assignment and explain why it matters.
Label the runners 01-30. Use a random number generator (or random digits, skipping repeats and numbers above 30) to choose 15 runners for the new shoe; the other 15 wear their usual shoes. Then compare the groups' 5K times. Random assignment makes the groups similar in fitness, experience and other variables on average, so a difference in times can be attributed to the shoe rather than to who chose it.
Question 17 of 20 · Short Answer
Explain the difference between random selection and random assignment. Which kind of study uses each, and what conclusion does each one support?
Random selection uses chance to choose who is in the study; sample surveys use it (and observational studies can), and it supports generalizing to the population. Random assignment uses chance to decide which subjects receive which treatment; experiments use it, and it supports cause-and-effect conclusions. A study can use both, one or neither.
Question 18 of 20 · Short Answer
A news story reports: "Students who take music lessons score higher on math tests." The data came from school records. Explain what kind of study this is and why the headline should not say that music lessons cause higher scores.
It is an observational study: students or families chose music lessons; nobody assigned them. Other variables, such as family income or time for extra activities, could be linked to both music lessons and math scores. Without random assignment, the data show an association, not causation.
Question 19 of 20 · Short Answer
In an experiment, 24 volunteers were randomly assigned to take a 10-minute walk or to sit for 10 minutes before a memory test. The walking group's mean score was 14.5 and the sitting group's was 12.0. Compute the difference and state what the experiment can show if the difference is larger than chance would explain.
Difference: 14.5 - 12.0 = 2.5 points. Because the volunteers were randomly assigned, the groups were similar except for the walk, so a difference larger than chance would explain is evidence that the walk caused higher memory scores for volunteers like these. (Deciding whether 2.5 is larger than chance would explain is the work of HSS.IC.B.5.)
Question 20 of 20 · Short Answer
A random sample of 500 adults is asked two questions: how many days a week they exercise and how stressed they feel. Those who exercise more report less stress. Is this a survey, an observational study or both? Can it show that exercise reduces stress?
It is both: a sample survey is the data-collection method, and because it relates two variables without imposing a treatment, it is also an observational study. Random selection lets the association be generalized to the adult population sampled, but with no random assignment it cannot show that exercise reduces stress; less stressed people might simply have more time to exercise.
0 of 20 answered · 0 correct
06
Frequently Asked Questions
10 Questions
What does HSS.IC.B.3 mean?
It means students can tell sample surveys, experiments and observational studies apart and explain what randomization does in each. A survey estimates a population value, an experiment tests whether a treatment causes an effect, and an observational study describes a relationship without intervening.
What is the easiest way to tell an experiment from an observational study?
Ask whether the researcher imposed a treatment. If the researcher decided who got which treatment, it is an experiment. If the subjects already differed, or chose their own group, it is an observational study, even if two groups are compared.
What is the difference between a census and a sample survey?
A census collects data from every member of the population; a sample survey collects data from part of it. A census is often too slow or costly for a large population, and a well-chosen random sample can estimate a population value closely, with a known amount of chance error.
How can students carry out random assignment without technology?
Write each subject's name or number on identical slips, mix them in a bag, and draw half for the first treatment. A random digit table also works: label the subjects, read the digits in groups, and skip repeats and unused labels, as in the lesson's examples. Flipping a coin for each subject is random too, but it can produce groups of unequal size.
Can an observational study ever show cause and effect?
Not on its own. Without random assignment, a confounding variable could explain the relationship. Strong, repeated observational evidence, with a plausible mechanism and confounders measured and accounted for, can make a causal explanation convincing, but a single observational study shows association.
Why not always run an experiment?
Some treatments would be unethical to impose (such as smoking), impractical (such as where people live) or impossible (such as a person's age). For those questions, researchers use observational studies and interpret the results carefully.
Is HSS.IC.B.3 taught in Algebra 2 or Statistics?
It is usually taught in Algebra II or an introductory Statistics course, and in Math III in integrated pathways. It follows HSS.IC.A.1 on random samples and prepares students for HSS.IC.B.4 and HSS.IC.B.5, which use data from surveys and experiments.
What mistakes do students make on this standard?
A common one is calling any comparison of two groups an experiment. Others include confusing random selection with random assignment, claiming causation from an observational study, and thinking a large sample fixes the lack of random assignment.
What is a control group, and does every experiment need one?
A control group gives a baseline for comparison: it receives no treatment, a placebo or the current method. Every experiment needs some comparison, but it can be between two real treatments, such as two study methods, instead of a treatment and no treatment.
Is this on the SAT?
Related ideas can appear in the Problem-Solving and Data Analysis domain of the digital SAT, for example a question asking what conclusion a study's design supports, or whether results can be generalized. Knowing the roles of random selection and random assignment is the key idea.
07
Related Standards
6 standards
These standards connect to HSS.IC.B.3: prerequisites to review first, parallel standards at the same level, and next steps that build on it.
Before this lesson
7.SP.A.1Prerequisite
Use a sample to learn about a population; random samples tend to be representative
