Scatter Plot Activity- AP Statistics Hands-On Guide
Why Scatter Plot Activities Actually Matter in AP Statistics
Scatter plots aren't just another graph you make once and forget. They're the foundation for understanding correlation, regression, and how two variables relate to each other. If you can't read a scatter plot cold, you're going to struggle through entire units.
Most textbooks throw a few examples at you and move on. That's not enough. You need hands-on practice where you're actually creating, interpreting, and analyzing scatter plots yourself. This guide gives you activities that actually work in a classroom or for self-study.
What You'll Need Before Starting
Don't show up empty-handed. Here's the minimum setup:
- Graphing calculator or software (Desmos, GeoGebra, or TI-84)
- Real datasets or the ones provided below
- Pencil and graph paper for sketching (yes, really)
- Ruler for drawing trend lines by hand
Activity 1: The Body Measurement Lab
This is the classic for a reason. It works.
Setup
Collect data from classmates or use the sample dataset below. Measure height (in inches) and arm span (in inches). Aim for at least 20 data points.
What You Do
- Plot each person's height on the x-axis and arm span on the y-axis
- Look at the pattern. Is it positive? Negative? No pattern at all?
- Draw a line of best fit by hand first—don't use a calculator yet
- Calculate the correlation coefficient (r) using your calculator
- Find the regression equation
Why This Works
You're using real data from real people. The relationship between height and arm span is strong and positive, which makes it perfect for learning the basics. You can see the pattern with your own eyes.
Activity 2: The Outlier Challenge
Outliers aren't just interesting—they completely change your regression line. This activity makes that obvious.
Setup
Start with this dataset of study hours versus exam scores:
| Hours Studied | Exam Score |
|---|---|
| 2 | 65 |
| 4 | 72 |
| 5 | 78 |
| 6 | 82 |
| 7 | 85 |
| 8 | 90 |
| 10 | 95 |
| 3 | 45 |
What You Do
- Plot the data with hours on x-axis and score on y-axis
- Notice that last point (3 hours, 45%)—that's your outlier
- Calculate r and the regression equation with the outlier included
- Remove the outlier and recalculate
- Compare the results
The difference will shock you. A single outlier can tank your correlation from strong to weak. This is exactly what the AP exam expects you to recognize.
Activity 3: The Prediction Game
Nothing teaches regression like being wrong. This activity makes you commit to predictions before you see the answer.
Setup
Get a dataset with two quantitative variables. Some good sources:
- Car weight vs. MPG (search for the "cars" dataset)
- Temperature vs. ice cream sales
- Advertising budget vs. product sales
What You Do
- Plot the data and find the regression equation
- Pick a value for x that wasn't in your dataset
- Use the regression equation to predict y
- Then—and this is the important part—discuss whether your prediction is reasonable
Predictions are only valid within the range of your x-values. Predicting outside that range is extrapolation, and it's almost always a bad idea on the AP exam. The question will try to trick you into doing it anyway.
Activity 4: Correlation vs. Causation Drill
This is where most students lose points. The graph looks like A causes B. But does it?
The Setup
Find three scatter plots online that show strong correlation but definitely aren't causal:
- Ice cream sales vs. drowning deaths (both go up in summer—confounding variable is heat)
- Average SAT scores vs. teacher salaries in different states
- Facebook users vs. national debt over time
What You Do
For each scatter plot:
- Describe the correlation (direction and strength)
- Explain why correlation does not mean causation here
- Identify the likely confounding variable
The AP exam loves questions that test this concept directly. If you can't explain why A doesn't cause B when the scatter plot shows a clear pattern, you're losing easy points.
How to Get Started This Week
Stop reading and start doing. Here's your action plan:
- Today: Complete the Body Measurement Lab with real data from 15-20 people
- Tomorrow: Do the Outlier Challenge and calculate how much the outlier changes r
- This week: Find one real dataset online and practice making predictions
- Before the test: Drill correlation vs. causation with three different examples until you can explain it in your sleep
Common Mistakes That Will Cost You
- Drawing the line of best fit by eyeballing it: Your brain is terrible at this. Use the calculator's regression function instead.
- Ignoring the residual plot: A high r² doesn't mean your model is good. Check residuals for patterns.
- Extrapolating outside the data range: The AP exam will test this. Don't fall for it.
- Confusing correlation with causation: Every. Single. Year. Students lose points on this.
The Bottom Line
Scatter plot activities work only if you actually do them. Reading about correlation coefficients isn't the same as calculating one from real data. Pick one activity from this guide, do it today, and move to the next one tomorrow.
Your AP exam score depends on being able to interpret, analyze, and make predictions from scatter plots without hesitation. The only way to get there is practice with real data, real problems, and real mistakes.