Scatter Plots & Bivariate Data

Section 1

Where Did Scatter Plots Come From?

Have you ever wondered how scientists figure out whether one thing is connected to another? For example, does studying more lead to better test scores? Does the temperature outside affect how much ice cream people buy? People have asked questions like these for hundreds of years. Scatter plots were invented to help answer them.

A scatter plot is a graph that shows data about two different measurements at the same time. Instead of just looking at a list of numbers, you can see patterns by placing dots on a grid. Let's look at how this idea developed over time.

1686

English scientist Edmund Halley created one of the first scatter-like graphs. He plotted air pressure against altitude (height above sea level) to study the atmosphere.

1833

English polymath John Herschel used scatter-style plots to show connections between pairs of measurements in astronomy, bringing the technique into wider scientific use.

1886

Francis Galton studied the heights of parents and their children. He plotted hundreds of data points and noticed a clear upward trend — taller parents generally had taller children. This work launched the field of correlation.

1901

Karl Pearson, a student of Galton, developed a mathematical formula to measure exactly how strong a relationship is between two quantities. His work turned scatter plots into a powerful tool used in science, medicine, and business.

Today, scatter plots are everywhere — from tracking athletic performance to predicting the weather. They help us move from "I think these two things are related" to "Here's the data that shows how."

Section 2

Core Definitions & Ideas

Before we build a scatter plot, let's nail down some key vocabulary. These terms will come up again and again, so understanding them now will make everything else click.

Bivariate Data

Bivariate means "two variables." Bivariate data is a set of paired measurements — each data point has two numbers. For example, each student's height and shoe size together form one data pair.

Scatter Plot

A scatter plot is a graph that places one dot for each data pair. The horizontal axis (x-axis) represents one variable, and the vertical axis (y-axis) represents the other. The dots' positions reveal patterns.

Association (Correlation)

When two variables tend to change together, we say there is an association. If one goes up and the other goes up, that's a positive association. If one goes up and the other goes down, that's negative.

Clusters & Outliers

Sometimes dots bunch together in groups called clusters. A point that sits far away from the rest is called an outlier. Both give you important information about the data.

✦ KEY TAKEAWAY

Think of a scatter plot like a detective's bulletin board. Each pushpin (dot) holds one clue (a data pair). When you step back and look at all the pushpins together, you start to see connections you'd never notice by reading a list of numbers. That's the power of visualizing bivariate data.

Section 3

Visual Guide: Anatomy of a Scatter Plot

Let's look at a real scatter plot. Imagine we collected data on hours spent studying and test scores for 12 students. Each dot below represents one student.

Study Time vs. Test Score — dots trend upward showing a positive association.

Notice how the dots move from the lower left to the upper right. That tells us as study hours increase, test scores tend to increase too. The dashed yellow line is called a trend line — it shows the overall direction. This is a positive association.

Every scatter plot has four key parts: a labeled x-axis (horizontal), a labeled y-axis (vertical), data points (the dots), and a title. Always label your axes so anyone reading your graph knows what each dot represents.

Section 4

How to Build a Scatter Plot Step by Step

Making a scatter plot isn't hard, but you need to follow the steps in order. Here's your game plan.

Step-by-Step Process

Step 1 — Collect paired data. You need two measurements for every item. For example, for each student you might record hours of sleep and number of missed questions.

Step 2 — Choose your axes. Decide which variable goes on the x-axis and which goes on the y-axis. If one variable might cause the other, put the cause on x. (Hours of sleep might affect missed questions, so sleep goes on x.)

Step 3 — Set your scales. Look at the range of each variable and pick a scale that fits. Use even intervals (like counting by 1s, 5s, or 10s). Make sure your scale covers from below your smallest value to above your largest.

Step 4 — Plot each pair. For each data pair (x, y), find the x-value on the horizontal axis, then move straight up to the y-value and place a dot.

Step 5 — Label and title. Write clear labels on both axes and give your graph a descriptive title.

Plotting a Point

(x, y) → go right to x, then up to y

Example: (3, 75) means go right to 3 on the x-axis, then up to 75 on the y-axis, and place a dot there.

When you're choosing scales, a useful formula for figuring out your interval size is:

Scale Interval

interval ≈ (largest value − smallest value) ÷ number of grid lines

Then round up to a friendly number like 5, 10, or 25.

For example, if your y-values range from 50 to 100 and you want about 5 grid lines: (100 − 50) ÷ 5 = 10. So you'd count by 10s on that axis — 50, 60, 70, 80, 90, 100. Easy!

✦ KEY TAKEAWAY

Building a scatter plot is like setting up a treasure map. The x-axis and y-axis are your compass directions, each dot marks where "treasure" (a data pair) is buried, and when you see all the dots at once, the map reveals the big picture — the pattern hiding in the data.

Section 5

Types of Patterns in Scatter Plots

Once you have your scatter plot, the next step is interpreting it — figuring out what the pattern of dots tells you. There are several types of patterns you should know.

Four types of scatter plot patterns: positive linear, negative linear, nonlinear (curved), and no association.

Here's a quick guide to each pattern:

Pattern	What the Dots Look Like	What It Means
Positive Linear	Dots go up from left to right, close to a straight line	As x increases, y increases
Negative Linear	Dots go down from left to right, close to a straight line	As x increases, y decreases
Nonlinear	Dots follow a curve (U-shape, arch, etc.)	There's a relationship, but it's not a straight line
No Association	Dots are scattered randomly with no pattern	The two variables don't seem related

You should also describe how strong or weak an association is. If the dots are packed tightly around the trend line, the association is strong. If they're loosely spread out but still show a general direction, it's weak.

Finally, watch for outliers — points that don't fit the overall pattern. An outlier might be a data error, or it might be a genuinely unusual case worth investigating.

Section 6

Worked Example

Let's work through a complete problem from start to finish. Here's a data set showing the temperature (°F) on different days and the number of bottles of water sold at a snack stand.

Temperature (°F)	Bottles Sold
60	15
65	22
70	28
75	35
80	40
85	48
90	55
95	60

Temperature vs. Water Bottles Sold

Step 1 — Set Up the AxesTemperature is likely the cause (hotter days cause people to buy water), so put temperature on the x-axis. Bottles sold goes on the y-axis. Temperature ranges from 60 to 95, so we can label the x-axis from 55 to 100, counting by 5s. Bottles sold ranges from 15 to 60, so we label the y-axis from 10 to 65, counting by 5s or 10s.

Step 2 — Plot Each PointTake the first pair: (60, 15). Go right to 60, then up to 15, and place a dot. Repeat for every pair: (65, 22), (70, 28), (75, 35), (80, 40), (85, 48), (90, 55), (95, 60).

Step 3 — Look at the PatternThe dots climb steadily from lower left to upper right. This shows a positive linear association. The dots are fairly close together, so the association is strong.

Step 4 — Check for OutliersAll dots follow the trend closely. There are no outliers in this data set.

Step 5 — Interpret in ContextAs the temperature increases, the number of bottles of water sold also increases. This makes sense — hotter weather makes people thirstier! The relationship is strong and positive, meaning temperature is a good predictor of water sales at this snack stand.

Section 7

Strengths & Limitations of Scatter Plots

Scatter plots are incredibly useful, but like any tool, they have both strengths and limitations. Let's be honest about both.

Strengths ✓	Limitations ✗
Show patterns at a glance — you can instantly see positive, negative, or no association	Can be hard to read with a very large number of data points (dots overlap)
Reveal outliers that might be hidden in a table of numbers	Only show two variables at a time — real life often involves many factors
Help you make predictions by drawing a trend line	Association does NOT mean causation — just because two things are related doesn't mean one causes the other
Work for any size of data set	Choosing bad scales can make a pattern look stronger or weaker than it really is

✦ KEY TAKEAWAY

Here's a big one: correlation is not causation. Imagine you notice that as ice cream sales go up, so do swimming pool accidents. Does eating ice cream cause accidents? Of course not! Both happen more in summer because of hot weather. A scatter plot can show you that two things are associated, but it can't tell you why — you need to think critically about what else might be going on.

Section 8

What Comes Next?

Now that you can build and interpret scatter plots, you're standing at the doorway to some really powerful ideas in statistics. Here's a preview of where this leads.

What You Know Now	What's Coming Next	Why It Matters
Eyeballing the direction and strength of an association	Line of best fit — a straight line drawn through the data to model the trend	Lets you write an equation and make precise predictions
Describing patterns as positive, negative, or none	Correlation coefficient (r) — a number from −1 to +1 that measures strength and direction	Gives you an exact measurement instead of just "strong" or "weak"
Noticing linear vs. nonlinear patterns	Residual analysis — studying the gaps between dots and the trend line	Helps you decide if a straight line is really the right model

You'll explore the line of best fit very soon in 8th grade. That skill builds directly on everything you've learned here. When you draw that line, you're summarizing the scatter plot with a single equation — and that equation can predict values you haven't even measured yet. Pretty cool!

In high school, you'll learn about the correlation coefficient, usually written as r. When r is close to +1, the positive association is super strong. When it's close to −1, the negative association is super strong. When it's near 0, there's no real association. Think of it as a "scorecard" for your scatter plot.

Section 9

Practice Problems

PROBLEM 1 — CONCEPTUAL

What does the word bivariate mean, and why do scatter plots need bivariate data?

PROBLEM 2 — BASIC

A scatter plot shows hours of video games played per week on the x-axis and grade point average (GPA) on the y-axis. The dots go from the upper left down to the lower right. What type of association does this scatter plot show?

PROBLEM 3 — INTERMEDIATE

Here are data pairs for the number of practice sessions and free-throw percentage for 6 basketball players: (2, 40), (4, 50), (5, 58), (7, 65), (8, 72), (10, 80). If you plotted these on a scatter plot, describe the association you'd see. Is it positive or negative? Strong or weak? Linear or nonlinear?

PROBLEM 4 — APPLIED

A park ranger records the daily high temperature (°F) and the number of visitors to a nature trail over 7 days: (55, 120), (62, 145), (68, 180), (74, 200), (80, 210), (88, 190), (95, 100). Plot these (or imagine them). Describe the overall pattern. Are there any points that break the trend? What might explain the pattern?

PROBLEM 5 — CHALLENGE

A student creates a scatter plot comparing shoe size and math test score for students in grades K through 12. The plot shows a strong positive association — bigger shoe sizes tend to have higher test scores. The student concludes, "Having bigger feet makes you better at math." Explain the flaw in this reasoning, and propose a better explanation for the pattern.

Summary

Putting It All Together

A scatter plot is a graph that displays bivariate data — two measurements for every item — as dots on a coordinate plane. You build one by choosing your axes, setting a scale, and plotting each (x, y) pair as a point. Once the dots are in place, you look for patterns of association. A positive association means both variables increase together; a negative association means one goes up while the other goes down. If the dots follow a straight-line trend, the association is linear; if they curve, it's nonlinear. The tightness of the dots tells you whether the association is strong or weak.

Always watch for outliers — data points that don't fit the pattern — and remember the golden rule: association does not prove causation. A scatter plot reveals relationships, but a lurking variable might be the real reason two things seem connected. These skills will serve as the foundation for drawing lines of best fit and computing correlation coefficients as you move deeper into statistics.