compare two populations by expressing the difference between their means as a multiple of the mean absolute deviation, and decide whether the difference is large or small.
Phase 3: Make Sense
Do Now. The Old Faithful groups differ by 26 minutes. Sample data: grade 7 and grade 8 screen-time samples differ by 10 minutes a day. Which gap seems more important? What else would you want to know?
A gap between two groups is most convincing when…
The reading opens over the lesson; close it to come back to exactly where you were.
A gap needs a yardstick. A 10-minute difference can be huge or tiny. It depends on how much the values in each group vary. So we measure the gap in units of spread: difference of means ÷ MAD. If the two groups’ MADs differ, use the larger one to be careful.
A rule of thumb. If the difference is less than about 1 MAD, the groups overlap a lot and the difference may be just chance. If it is about 2 MADs or more, the groups are clearly different. This is an informal rule for grade 7, not a formal test.
Worked example 1: Old Faithful. From lesson 2.08 sample data: means 68 and 94; MADs 2.6 and 3.4 minutes. Step 1: difference = 94 − 68 = 26. Step 2: use the larger MAD, 3.4. Step 3: 26 ÷ 3.4 ≈ 7.6. Step 4: the gap is more than 7 MADs, so the two groups are very different. The park’s own ±10 agrees: 68 + 10 = 78 is still below 94 − 10 = 84, so the ranges do not overlap.
Worked example 2: screen time. Sample data, minutes a day. Grade 7: 60, 80, 90, 110, 120, 120, 130, 150, 160, 180 (mean 120). Grade 8: 70, 90, 100, 120, 130, 130, 140, 160, 170, 190 (mean 130). Step 1: difference = 10. Step 2: each group’s distances from its mean add to 280, so each MAD = 28. Step 3: 10 ÷ 28 ≈ 0.36. Step 4: the gap is about one third of a MAD. The groups overlap almost completely, so this sample does not show a real grade difference.
Center alone can mislead. A report that says “grade 8 spends 10 more minutes on screens” sounds like news. With the MAD, we can say “the difference is small compared with how much students vary.”
For our report. Every two-group comparison in our School Survey Report will give both centers, a spread for each group, and the gap in MADs.
Show the difference, the MAD you used, and the quotient each time.
Decide whether each gap is big.
Sample data, carnival ring toss scores (out of 10). Grade 7 mean 5, MAD 2. Grade 8 mean 6, MAD 2. How many MADs apart are the means? Is the difference clear? Explain.
Sample data: two groups' means are 40 and 52, and their MADs are 3 and 4. Using the larger MAD, how many MADs apart are the means?
Write one sentence for our report that compares two groups the right way, with centers, spreads and the gap in MADs. Use any sample data from this lesson.
We can judge whether a gap is big. Next we hunt down the most common mistakes people make with surveys and data.
Exit ticket: why is a 10-minute difference big for one pair of groups and small for another?
Pacing: Do Now 4 min, reading 7, mini lesson 11, guided practice 13, independent practice and exit ticket 5.
Accuracy: 94 or 68 minutes ±10, 1.5-5 minute duration, 106-184 feet, Daisy 2 h 45 min ±30, Grand 6 h 30 min ±60, the 2025 table update, irregular geysers, the app and phone options, and the closed periods are from the NPS Current Geyser Activity page (last updated 2026-10-26; opened 2026-11-20). The “about 2 MADs” rule of thumb is a classroom guideline, not a park or PSSA rule. Checked: 26 ÷ 3.4 = 7.65; 26 ÷ 2.6 = 10; screen-time sums 1,200 and 1,300, means 120 and 130, distance sums 280 each, MADs 28; 10 ÷ 28 = 0.357; 12 ÷ 4 = 3; 68 + 10 = 78; 94 − 10 = 84; Daisy 165 ± 30 = 135 to 195 minutes.
Likely misconceptions: dividing the MAD by the difference; deciding from the means only; thinking “not clear” means “the groups are the same.”
Supports: a comparison frame: “The means differ by __. The MAD is __. So the gap is about __ MADs, which is (small / clear).” Stretch: build two small data sets with the same means as the screen-time samples but a gap of at least 2 MADs.
Answer guide: Daisy: 135 to 195 minutes, which is far above Old Faithful’s longest typical wait (94 + 10 = 104), so no overlap. Ring toss: 1 ÷ 2 = 0.5 MAD, not clear. The report sentence should give both means, both MADs and the gap in MADs.
National Park Service, Yellowstone National Park: Current Geyser Activity · U.S. government work, public domain
Yellowstone National Park posts a table about its geysers. For each one it gives the average interval between eruptions, how long an eruption lasts, and how high it goes. The table was last updated in 2025, and the park says geyser activity changes over time.
Some geysers can be predicted, and the table marks them. For Old Faithful, the average interval is listed as 94 or 68 minutes, plus or minus 10 minutes. An eruption lasts from 1.5 to 5 minutes and reaches 106 to 184 feet.
Other predicted geysers have much longer waits. Daisy erupts about every 2 hours 45 minutes, plus or minus 30 minutes. Grand erupts about every 6 hours 30 minutes, plus or minus 60 minutes. Many others are listed as irregular, which means they cannot be predicted well.
The page also says the park’s predictions can be found on the free NPS app, by phone, or on a geyser-watching website. Predictions are not available when the Old Faithful Visitor Education Center is closed, usually early November to mid-December and mid-March to mid-April.
Source: National Park Service, Yellowstone National Park: Current Geyser Activity (nps.gov)