show what I know about random samples, inferences and comparing two groups, and plan a fair, private survey of our school for Part 2.
Phase 3: Make Sense
Do Now. Look back at our Wonder Board. Which questions can we now answer with a survey? Star two. Which ones still need the probability tools of Part 2? Circle two.
Our survey will ask students about their habits. What must the report never show?
The reading opens over the lesson; close it to come back to exactly where you were.
Part 1 in one page. A population is everyone we want to learn about; a sample is who we ask. A random sample gives every member the same chance and lets chance choose. From a random sample we can estimate for the whole population with proportions, and report a likely range. To compare two groups, we use centers (mean, median), spreads (range, IQR, MAD) and the gap measured in MADs.
Our survey rules. (1) Ask clear, fair questions. (2) Use a complete list and a random method. (3) Keep answers anonymous: no names on answer sheets, and report only groups. (4) Ask nothing private, such as family money or health. (5) Students may skip any question.
Worked example 1: a plan line. Sample plan: Priya’s team asks “How many minutes did you read last night?” Step 1: population = grade 7 and 8 students. Step 2: frame = the class lists, numbered 001-480 (a sample school of 480). Step 3: method = random digits, skipping repeats and numbers above 480. Step 4: size = 40, so each student stands for 480 ÷ 40 = 12. Step 5: compare grade 7 and grade 8 with box plots and the gap in MADs.
Worked example 2: checking a plan. Sample plan: Marcus’s team will ask 40 students at the bus loop which carnival day they prefer. Step 1: is it random? No, it leaves out walkers and students who are driven. Step 2: fix: draw 40 numbers from the full list. Step 3: if 16 of 40 choose Friday, then 16/40 = 40% and the estimate is 40% of 480 = 192 students.
The check-in. The items below cover every Part 1 target. Work alone, then check with a partner.
Part 1 check-in. Answer on your own first.
Survey plan (team draft). This plan goes into Part 2.
Write your team's survey plan: one categorical question, one numerical question, the population, the sampling frame, the random method, the sample size, and the two groups you will compare.
Which question should our survey not ask?
Write the sentence our report will use to explain to families how answers were kept anonymous.
Part 1 is done: we are ready to build. In Part 2 we run our survey, test the carnival games with probability, and solve the seven-heads puzzle.
Exit ticket: which Part 1 skill will matter most to families who read our report? Explain with an example.
Pacing: Wonder Board 5 min, reading 6, mini lesson 8, check-in 15, planning and exit ticket 6.
Use as a formative check of all three Part 1 targets: random samples (B.1:1a, items 1 and the plan fix), inference from samples (B.1:2a, items 2, 3 and 4) and comparing two populations (B.2:1a, items 5 and 6). Regroup students who miss two or more items on one target before Part 2.
Accuracy: no sharing of answers, the lifetime oath, the $250,000 fine and 5-year prison penalty, deleting names and identifiers, the methods that prevent identifying people, the list of things never asked, field ID and the legal requirement are from the Census Bureau page opened 2026-11-20. School numbers are sample data. Checked: 480 ÷ 40 = 12; 16/40 = 40%, 0.4 × 480 = 192; 9/25 × 300 = 108; (24 − 18) ÷ 2 = 3; 2, 3, 3, 4, 4, 5, 30 sum 51, mean 7.29, median 4.
Privacy: before Part 2, have the principal approve the final survey questions. Answer sheets carry no names; students may skip any question.
Answer guide: Marcus fix: number all 480 students and use random digits (skip repeats and numbers above 480) or draw 40 numbered slips. Anonymity sentence: “No names were written on answer sheets, and we report only group results.” Survey plans should label the categorical and numerical questions and name two groups to compare (for example grade 7 and grade 8).
U.S. Census Bureau, American Community Survey: Top Questions About the Survey · U.S. government work, public domain
People answer surveys only if they trust that their answers are safe. The Census Bureau explains how it protects people who answer the American Community Survey.
By law, the Bureau cannot share anyone’s answers with anyone else, including companies, other government agencies and police. Every Census Bureau worker takes an oath to protect information that could identify a person, and that promise lasts for life. A worker who breaks it can be fined up to $250,000, sent to prison for up to 5 years, or both.
When the Bureau works with the answers, it deletes names and other details that identify people from the files it uses to make its tables. It also uses statistical methods so that the numbers it releases do not point to any one person or business.
The page lists things the survey never asks for, such as a Social Security number, bank or credit card information, money or donations. Survey workers who visit homes always carry an official ID. The page also says the survey is required by law for addresses that are chosen.
Source: U.S. Census Bureau, American Community Survey: Top Questions About the Survey (census.gov)