School Survey Report and Carnival Game Fairness Check
How can a small sample tell us about a whole school, and is that game really fair?
The mystery: A coin landed heads 7 times in a row. Is the next flip more likely to be tails?
Questions we will figure out together in this unit:
tell the difference between a population and a sample, and explain why survey scientists ask a sample instead of everyone.
Phase 1: Spark
Do Now. Sample puzzle: a student flips a coin at lunch and it lands heads 7 times in a row. Her friend says, “Tails is due next.” Do you agree? Write your answer and one reason. We will not settle this today. We will test it with real experiments later in the unit.
On the eighth flip, which do you think is true?
The reading opens over the lesson; close it to come back to exactly where you were.
Welcome, survey scientists. This unit has a real job for you. You will make a School Survey Report and Carnival Game Fairness Check for our school community: students, families, the student council and teachers. The report will tell the community what students think and do, and the fairness check will tell families whether the spring carnival games are really fair. Our driving question: How can a small sample tell us about a whole school, and is that game really fair?
Population and sample. The population is the whole group we want to learn about, such as every student in our school. A sample is the part of the population we actually ask or measure. The government’s job survey in our reading uses a sample of about 110,000 people to learn about a nation of millions.
Why not ask everyone? Asking everyone is called a census. A census of a big group takes a long time and costs a lot. A good sample can give an answer that is close, much faster. The key word is good: the sample has to look like the whole population.
Worked example 1: name the parts. Keisha wants to know how many minutes students in grades 7 and 8 read each night. Step 1: the population is all grade 7 and 8 students in the school. Step 2: she plans to ask 40 of them, so the sample is those 40 students. Step 3: the thing she measures, minutes of reading, is the variable. Step 4: she will use the sample’s answers to estimate the answer for the whole population.
Worked example 2: how big is the sample? Sample data: a school has 480 students, and Luis plans to survey 60 of them. Step 1: write the part over the whole: 60/480. Step 2: simplify: 60/480 = 1/8. Step 3: change to a percent: 1/8 = 0.125 = 12.5%. Step 4: so each student in his sample stands for about 8 students in the school, because 480 ÷ 60 = 8.
A warning from our reading. The job survey says its estimate is probably within about 300,000 people of the true number. A sample gives a close answer, not a perfect one. Good survey scientists say how close they think they are.
Our Wonder Board. Today we start a board of questions we could answer with math. Some will be about surveys; some will be about chance, like the seven heads. We will come back to it all unit.
Use the reading and the two worked examples. Name the population and the sample before you answer.
Start your survey scientist notebook.
Sample data: a team surveys 75 of the 300 families at a school. What percent of the families did they survey?
Write the population, the sample and the variable for this plan: “We will ask 50 students in grades 7 and 8 which carnival game they like best.”
Write one question about our school or the carnival that you could answer with a survey or an experiment. Put it on a sticky note for the Wonder Board.
We met our puzzle and our audience. Next time we find out what the community wants to know and look at how the Census Bureau chooses who to ask.
Exit ticket: in your own words, what is the difference between a population and a sample? Give a school example.
Pacing: Do Now 5 min, reading 8, mini lesson 12, guided practice 10, Wonder Board and exit ticket 5.
Accuracy: 60,000 households, about 110,000 people, about 2,000 areas with about 800 chosen, one-fourth replaced each month, no more than 4 months in a row, polls of fewer than 2,000, and “90 out of 100” within about 300,000 are from the BLS page opened 2026-11-20. School sizes (480, 300 families) are sample data. Checked: 60/480 = 1/8 = 12.5%; 480 ÷ 60 = 8; 40/480 = 1/12; 480 ÷ 40 = 12; 75/300 = 25%.
The coin puzzle: do not resolve it today. The answer (each flip is independent, so heads and tails stay equally likely for a fair coin) belongs to Part 2 (CC.2.4.7.B.3). Accept all predictions and record the class tally for later.
Likely misconceptions: calling the people asked the population; thinking a sample must be huge to be useful; thinking a sample answer is exact.
Supports: a two-circle picture (big circle = population, small circle inside = sample). Stretch: the job survey asks about 110,000 people each month; if one-fourth of the households are replaced, about how many people are new each month? (About 27,500, if households are about the same size.)
Answer guide: carnival plan: population = all grade 7 and 8 students, sample = the 50 asked, variable = favorite game. The 300,000 explanation should say a different sample could give a slightly different answer, but the estimate is still close to the truth for a very large group.
U.S. Bureau of Labor Statistics, How the Government Measures Unemployment · U.S. government work, public domain
Every month the government reports how many people in the United States are looking for work but cannot find a job. How can it know this for a whole country? The Bureau of Labor Statistics explains that it does not ask every home. Contacting every home would cost far too much and take far too long.
Instead, the Current Population Survey asks a sample, a smaller group chosen to stand for everyone. The sample has about 60,000 households, or about 110,000 people, each month. The page says many public opinion polls ask fewer than 2,000 people.
The sample is planned with care. The country is split into about 2,000 areas, and about 800 of them are chosen so that every state, city areas, farm areas and other kinds of places are included. Each month one-fourth of the households are replaced, and no household stays in the survey for more than 4 months in a row.
Because the survey uses a sample, its answer can be a little off. The page says the chances are 90 out of 100 that the monthly estimate is within about 300,000 people of what a count of everyone would show.
Source: U.S. Bureau of Labor Statistics, How the Government Measures Unemployment (bls.gov)