use the results of a random sample and proportional reasoning to estimate a count or percent for the whole population.
Phase 2: Investigate
Do Now. Sample data: in a random sample of 40 students from a school of 480, 14 say they would come to a Saturday carnival. Estimate how many students in the whole school would come. Show your thinking.
About how many of the 480 students would come on Saturday?
The reading opens over the lesson; close it to come back to exactly where you were.
The big idea. If a sample is random, its percent is our best estimate of the population’s percent. We use a proportion: the part of the sample over the sample size equals the part of the population over the population size.
Three ways to do the same step. (1) Find the sample percent and take that percent of the population. (2) Find how many people each sample member stands for and multiply. (3) Write and solve a proportion. All three give the same answer, so use the one that is easiest for the numbers.
Worked example 1: the Saturday carnival. Sample data: 14 of 40 randomly chosen students would come on Saturday; the school has 480 students. Step 1: sample percent: 14/40 = 0.35 = 35%. Step 2: 35% of 480 = 0.35 × 480 = 168. Step 3: check with “stands for”: 480 ÷ 40 = 12, and 14 × 12 = 168. Estimate: about 168 students.
Worked example 2: snack planning. Sample data: Jada surveys a random sample of 25 of the 300 families coming to the carnival; 9 say they will buy dinner there. Step 1: write a proportion: 9/25 = x/300. Step 2: 300 ÷ 25 = 12, so x = 9 × 12 = 108. Step 3: as a percent, 9/25 = 36%, and 36% of 300 = 108. Estimate: about 108 families will buy dinner.
Say it like a scientist. Write “about 168” or “an estimated 168,” not “exactly 168.” Our reading shows even a national survey of about 110,000 people gives an estimate that can be off by a little.
When not to infer. This step only works when the sample is random (or carefully weighted, like the job survey). Using the gym sample from lesson 2.03 to estimate the whole school would carry its bias into the estimate.
Today’s hands-on. Each team gets a sample-school bag with an unknown mix of colored cards standing for students’ favorite games. You will draw a random sample of 20 and estimate the count for the whole bag of 200 cards.
Hands-on. Your bag holds 200 cards in three colors (ring toss, bean bag throw, spin the wheel). Do not look. Draw a random sample of 20, record the colors, then return the cards. Use your counts and the problems below.
Estimate, then check with a second method.
Sample data: a random sample of 60 students from a school of 480 shows 21 would join a carnival planning team. Estimate the number for the whole school two different ways.
Sample data: 3 of 20 randomly chosen cards are spin the wheel. About how many of 200 cards are?
Your team's estimate from the bag and another team's estimate are different. Write one reason why, and one way to get a better estimate.
We turned a sample into an estimate. Next time we look at what happens when many teams take samples from the same bag.
Exit ticket: a random sample of 30 of 360 students shows 12 want a dance at the carnival. Estimate the number for the whole school and show your proportion.
Pacing: Do Now 5 min, reading 7, mini lesson 10, bag sample and guided practice 13, independent practice and exit ticket 5.
Materials: per team, a bag of 200 colored cards or counters. Suggested hidden mix: 80 ring toss, 90 bean bag, 30 spin the wheel (40%, 45%, 15%). Keep the mix secret until lesson 2.06, where teams pool samples.
Accuracy: about 60,000 households, weighting by age, sex, race, Hispanic ethnicity and state, 75 percent month-to-month and 50 percent year-to-year overlap, the 90-out-of-100 statement within about 300,000, and 7 to 15 million total unemployment are from the BLS page (opened 2026-11-20). Checked: 14/40 = 35%, 0.35 × 480 = 168, 480 ÷ 40 = 12; 9/25 = 36%, 0.36 × 300 = 108; 8/20 × 200 = 80; 15/50 × 600 = 180; 21/60 × 480 = 168; 3/20 × 200 = 30; exit ticket 12/30 × 360 = 144.
Likely misconceptions: setting the proportion upside down (50/15); reporting the sample count (14) as the school count; claiming the estimate is exact.
Supports: a three-row table: sample part, sample size, sample percent; then population size and estimate. Stretch: if 168 students come and each plays 3 games, how many game tickets are needed? (504.)
Answer guide: 21/60 = 35%, 35% of 480 = 168; or 480 ÷ 60 = 8 and 21 × 8 = 168. Teams differ because different random samples vary by chance; larger samples, or pooling samples, usually give a better estimate.
U.S. Bureau of Labor Statistics, How the Government Measures Unemployment · U.S. government work, public domain
The Current Population Survey asks about 60,000 households each month. The Bureau of Labor Statistics uses their answers to estimate facts about everyone in the country, such as how many people are out of work.
To do that, each person in the sample has to stand for many people who were not asked. The page explains that the sample results are adjusted, or weighted, using the Census Bureau’s estimates of the population. The weights are based on age, sex, race, Hispanic ethnicity and state, so that the sample matches the country on those traits.
The survey also keeps many of the same households from month to month. About 75 percent of the sample is the same from one month to the next, and 50 percent is the same from one year to the next. The page says this makes the estimates of change more reliable.
An estimate from a sample is close, not exact. The page says the chances are 90 out of 100 that the monthly estimate of unemployment is within about 300,000 people of the result a total census would give. Over the past ten years total unemployment has been about 7 to 15 million, so that error does not change the big picture.
Source: U.S. Bureau of Labor Statistics, How the Government Measures Unemployment (bls.gov)