compare two data sets with side-by-side box plots using medians and IQRs, and write a careful conclusion that does not claim more than the data show.
Phase 3: Make Sense
Do Now. Two states have NAEP average math scores. State A’s is 1 point higher than state B’s. A news post says, “State A’s students are better at math.” What would you want to know before you agree?
Can a higher average score from a sample still be “no real difference”?
The reading opens over the lesson; close it to come back to exactly where you were.
Box plots. A box plot shows the five-number summary: minimum, Q1, median, Q3 and maximum. The box covers the middle half of the data, so its length is the IQR. Two box plots on the same scale let you compare centers and spreads at a glance.
What to compare. (1) The medians: which is higher, and by how much? (2) The IQRs: which group varies more? (3) The overlap: does one box sit mostly above the other, or do they share most of their range?
Worked example 1: pages read, grade 7. Sample data, pages read last week by 11 random grade 7 students: 10, 15, 20, 22, 25, 30, 32, 35, 40, 48, 60. Step 1: minimum 10, maximum 60. Step 2: median = 6th value = 30. Step 3: lower half 10, 15, 20, 22, 25 gives Q1 = 20; upper half 32, 35, 40, 48, 60 gives Q3 = 40. Step 4: IQR = 40 − 20 = 20.
Worked example 2: pages read, grade 8. Sample data, 11 random grade 8 students: 5, 8, 12, 15, 18, 20, 24, 26, 30, 35, 50. Step 1: minimum 5, maximum 50. Step 2: median = 20. Step 3: Q1 = 12, Q3 = 30, IQR = 18. Step 4: compare: the grade 7 median is 10 pages higher; the IQRs are similar (20 and 18); the grade 8 median (20) equals grade 7’s Q1, and grade 7’s median (30) equals grade 8’s Q3. Step 5: the difference of medians, 10, is half of grade 7’s IQR.
A careful conclusion. “In our random samples, grade 7 students tended to read more pages last week than grade 8 students (median 30 vs 20 pages), but the box plots overlap a lot. With only 11 students in each sample, the difference might not hold for the whole school.” Notice the words tended to and might not.
Not cause and effect. Even a dependable difference does not tell us why. NAEP warns its comparisons cannot prove cause and effect, and neither can our survey.
Draw both box plots on one number line from 0 to 60 pages, grade 7 above grade 8. Use them for every item.
Build and compare.
Sample data, minutes of sleep over 8 hours (extra minutes) for 7 students in each group. Group A: 0, 10, 15, 20, 25, 30, 60. Group B: 20, 30, 35, 40, 45, 50, 55. Find both five-number summaries and compare the box plots.
NAEP says a significant difference is one that is…
Explain why our report should not say why one grade reads more, even if the difference were clear.
We can compare groups with care. Next lesson we check everything from Part 1 and plan our real school survey.
Exit ticket: name the three things to compare in side-by-side box plots.
Pacing: Do Now 4 min, reading 7, mini lesson 11, box plots and guided practice 13, independent practice and exit ticket 5.
Accuracy: the statistical test considering the size of the difference and the standard errors, the four outcomes, the small-sample case, the same-average-different-result case and the meaning of “significant” are from the NAEP page opened 2026-11-20. The 1-point state example in the Do Now is a sample situation, not a real NAEP result. Checked: grade 7 summary 10, 20, 30, 40, 60, IQR 20; grade 8 summary 5, 12, 20, 30, 50, IQR 18; median difference 10.
Likely misconceptions: reading the box length as the number of students; thinking a longer whisker means more students; claiming cause.
Supports: a box plot template with labeled tick marks every 5 pages. Stretch: change two grade 8 values so the boxes no longer overlap, and re-write the conclusion.
Answer guide: group A: 0, 10, 20, 30, 60 (IQR 20); group B: 20, 30, 40, 50, 55 (IQR 20). Group B’s median is 20 minutes higher, the IQRs are equal, and A’s Q3 (30) is below B’s median (40), so B tended to get more extra sleep. Cause: many things affect reading (schedules, sports, jobs), and a survey cannot sort them out.
National Center for Education Statistics, NAEP: Statistical Significance and Sample Size · U.S. government work, public domain
When NAEP reports that one group did better than another, the difference has passed a statistical test. The test looks at how big the difference between the averages is and at how uncertain each average is.
The page lists four possible outcomes when two places, A and B, are compared. A has a higher average than B. A has a lower average than B. No difference is found between A and B. Or the sample does not allow a reliable test, which can happen when a group’s sample is small.
The page warns about a surprising case. One state can have a higher average score than another state, and still the difference is not significant. Another state with the same average might show a significant difference. This happens because the uncertainty is not the same for every state.
NAEP also says “significant” does not mean a difference is large or important for schools. It means the difference is dependable, so it is likely to be real for the populations and not just a result of who was sampled.
Source: National Center for Education Statistics, NAEP: Statistical Significance and Sample Size (nces.ed.gov)