plan our dashboard panel and make a scatter plot from paired survey data, and decide whether the data form a function.
Phase 4: Build
Do Now. In Part 1 you asked whether students who sleep more get better grades, and you built a function model of a local trend. Now we collect our own data. Read today’s short reading, then answer.
The reading says a scatter plot can help answer which kind of question?
The reading opens over the lesson; close it to come back to exactly where you were.
Data scientists, welcome to Build. Our audience is the East Allegheny School Board and the borough councils. They make real decisions about start times, phones, parks and libraries. Our Community Data Dashboard will show them what our numbers say, and what they do not say.
Our six panels. Sleep and Scores, Screens and Sleep, Line of Fit and Predictions, Survey Two-Way Tables, the Community Trend Function from Part 1 and What the Data Can’t Tell Us. Each team owns one panel, and every team checks every panel.
Bivariate data. When we measure two things about each person, such as hours of sleep and a quiz score, we get bivariate data. Each person becomes one ordered pair (x, y) and one point on a scatter plot. Put the variable you think might explain the other on the x-axis.
| Student (sample) | A | B | C | D | E | F | G | H | I | J |
|---|---|---|---|---|---|---|---|---|---|---|
| Sleep on a school night (hours) | 6 | 6.5 | 7 | 7 | 7.5 | 8 | 8 | 8.5 | 9 | 9.5 |
| Quiz score (points out of 100) | 70 | 72 | 75 | 78 | 80 | 82 | 85 | 84 | 88 | 90 |
Worked example 1: Ava plots the sample data. Ava puts sleep on the x-axis and score on the y-axis. Sleep runs from 6 to 9.5 hours, so she numbers the x-axis from 5 to 10 by halves. Scores run from 70 to 90, so she numbers the y-axis from 60 to 95 by 5s and draws a small zigzag near 0 to show the axis skips ahead. Student H becomes the point (8.5, 84): over to 8.5, up to 84. She adds a title, axis labels with units, the word sample and a source line.
Worked example 2: Marcus checks for a function. From Part 1, a relation is a function when each input has exactly one output. Marcus looks at the input 7 hours. Student C has (7, 75) and student D has (7, 78). One input, two outputs, so the sample data is not a function. That is normal for real data: two people can sleep the same amount and score differently. Later we will draw a line that is a function to model the pattern.
Our criteria and constraints.
| Criterion | What it means |
|---|---|
| Correct | Every point and every number is checked a second way. |
| Clear | Title, labeled axes with units, a legend and a source on every display. |
| Useful | A board member can read it in one minute and act on it. |
| Honest | Sample data says sample. Real data names its source and date. We never report an association as a cause. |
| Constraints | One panel per team. Every model is an equation with its slope and intercept explained in words. Survey answers are anonymous and voluntary: no names, no student numbers, and any student may skip any question. |
Use the sample sleep and score table in the lesson.
Start your panel’s first display.
On grid paper, make a scatter plot of the sample sleep and score data. Include a title, labeled axes with units, an axis break, the word sample and a source line.
Write two survey questions for our class survey that each have a number answer (for example, minutes of reading last night). Explain how you will keep the answers anonymous.
Use the reading. Which parts of the NCES list did your scatter plot include? Which part will a school board member look at first, and why?
Every point is a person. Plot each one carefully.
Exit ticket: why can a set of real data points fail to be a function even when there is a clear pattern?
Pacing: warm-up and reading 7 minutes, mini lecture 9, guided practice 13, independent 9, closure 2.
Likely misconceptions: reversing the ordered pair (score, hours); spacing the axis by the data values instead of equal steps; thinking a repeated y-value breaks a function (only a repeated x with a different y does); connecting the dots like a line graph.
Supports: pre-scaled grid paper; a function checklist (find repeated inputs). Stretch: plot the data twice, once with an axis break and once starting at 0, and explain which is easier to read and which could mislead.
Survey: review the survey questions before giving them; avoid questions about health, family income or anything personal. If a survey is not possible, teams use the sample data sets in this part, labeled sample. As in 1.01, do not collect real grades or quiz scores: the Sleep and Scores panel uses the sample table only.
Answer guide: warm-up: whether one thing is related to another; H = (8.5, 84); 2 students (F and G) slept 8 hours; not a function (7 and 8 repeat with different scores); scale 60 to 95 by 5s; the source line; the function is (1, 4), (2, 6), (3, 6), (4, 9). Exit ticket: two people can share an input value and have different outputs, so an input can repeat with a different output.
National Center for Education Statistics, NCES Kids' Zone: Examining X-Y (Scatter) Plots · U.S. government work, public domain
The National Center for Education Statistics, or NCES, runs a Kids’ Zone with a guide to making graphs. One page explains X-Y plots, also called scatter plots. The guide says we use them to see whether one event affects another. Its example question: if you spend more hours studying, will you get better grades?
The guide then names the parts of a good plot. The title gives a short explanation of what the graph shows. The legend tells what each point stands for. In the NCES example, each color is a different country. The source says where the information came from, because it is important to give credit to the people who collected the data.
The y-axis runs up and down. It usually holds the numbers for the amount being measured, often starting at 0 and split into equal parts. The x-axis runs side to side. In the NCES example, the x-axis shows how much money each country spent per child on elementary and secondary schools.
Finally, the guide says the most important part of any graph is the data it holds. Every other part helps a reader understand that data.
Source: National Center for Education Statistics, NCES Kids' Zone: Examining X-Y (Scatter) Plots (nces.ed.gov)