Not sure which lesson to do first? Find your level and get a plan →

Math · Statistics & Probability · Grade 6-8

Scatter Plot

A scatter plot graphs paired numerical data as points.


Quick Answer

A scatter plot graphs paired numerical data as points. Use it when each person or thing has two measurements, such as a student's hours of sleep and test score or a car's age and price, and you want to see how they relate. Each dot is one case, placed at its (x,y) pair. The overall cloud shows whether the variables rise together, move in opposite directions, curve or show no pattern, and it makes outliers and clusters easy to spot.

Why This Matters

A scatter plot shows relationships that tables hide: in 30 rows of height and arm-span numbers, the near-straight line only appears once the points are plotted. It is also the check to make before drawing a trend line or computing a correlation, because a curve, two separate clusters or one wild point can make those summaries misleading.

Intuitive Explanation

For each of 10 students, write two numbers: hours of video games last week and hours of sleep per night. Put games on the horizontal axis and sleep on the vertical axis. A student with 4 hours of games and 9 hours of sleep becomes the dot at (4,9); one with 15 hours of games and 6.5 hours of sleep becomes the dot at (15,6.5). Ten students make ten dots, and each dot keeps both of one student's numbers together.

Then read the cloud as a whole. Dots drifting down to the right mean more gaming goes with less sleep, a negative association; dots drifting up mean a positive one. A bend, like an arch, means the relationship is curved. A dot far from the rest, such as a student with 20 hours of games and 10 hours of sleep, is an outlier worth checking: it may be a typing mistake or a real exception.

A line graph also puts dots on two axes, but it joins them in time order because the same thing is measured again and again, like one city's temperature each hour. Scatter-plot dots are separate cases, so joining them would invent a path that does not exist. And if one variable is a category, like favorite sport, there is no number to place on an axis, so a bar graph fits that data instead.

Think of it like this. A scatter plot works like pins on a map. Each pin marks one place by two numbers, how far east and how far north. A scatter plot pins each person by their two measurements, and the pattern of pins shows how those measurements relate, just as pins crowded along a coast show where people live.

Core idea. A scatter plot shows how two numerical variables move together.


When to Use

Use a scatter plot when every case has two numerical values and the question asks about a relationship, a trend or an outlier: height and arm span, study time and score, age and price. It is the first step before describing a correlation or fitting a line. For one numerical variable on its own, use a histogram, dot plot or box plot; for categories, a bar graph; for one quantity changing over time, a line graph.

Tip. Ask: Does each dot need both an x-value and a y-value?

How to Recognize It

Before drawing a scatter plot, check what each dot would stand for and whether both of its coordinates are numbers measured on the same case.

  1. Are the cases separate individuals, like 30 different students, rather than one thing checked at many times?

    If yes, a scatter plot fits and the dots stay unconnected. If it is one quantity tracked over time, such as a plant's height each week, use a line graph, where joining the points in time order makes sense.

  2. Is the question about how one measurement relates to another, rather than how a single measurement is spread out?

    If yes, plot the pairs. If you only have one variable, such as 50 students' heights, a histogram or dot plot shows its shape; a scatter plot needs a second variable for the other axis.

  3. Do you want to spot cases that break the trend, like a new car with a very low price?

    If yes, a scatter plot shows them as dots far from the cloud, which a table or a single summary number would hide. If you only need the strength of the link as a number, the correlation coefficient r gives it.

  4. Is one of the variables a category, such as favorite sport?

    If yes, a scatter plot does not fit, because a category has no position on a number line. Compare the groups with side-by-side box plots or a bar graph of averages instead.

Scatter Plot compared with Line graph and Histogram

Scatter plots, line graphs and histograms all put numbers along a horizontal axis, so they can look alike. Ask what one mark stands for: one case's pair of measurements (scatter plot), one moment in a sequence (line graph), or the count of values in a range (histogram).

Scatter Plot

Meaning
Use this when each observation has two numerical variables and the question asks about their relationship.
Key test
Does each dot need both an x-value and a y-value?
Example
A data set records study hours and test score for each student. What graph fits?

Line graph

Meaning
Connects ordered values, often over time.
Key test
Use when continuity or sequence matters.
Example
Temperature over days

Histogram

Meaning
Shows distribution of one numerical variable.
Key test
Use for one-variable frequency.
Example
Heights of students

Formula & Notation

How to read it: Each point is (xi,yi) where x is the explanatory variable and y is the response variable

Worked Examples

Example 1 — Study and score

Easy

A data set records study hours and test score for each student. What graph fits?
Show the steps and the answer
  1. Count the variables for each student: 2, study hours and test score.

    Each student gives a pair of numbers, not a single value.

  2. Check that both are numerical: hours (such as 2.5) and scores (such as 84) lie on number scales.

    Each variable can go on a numbered axis.

  3. Plot each student as one dot: 3 hours and a score of 78 sits at (3,78).

    Study hours go on the x-axis because they may help explain the score on the y-axis.

  4. Leave the dots unconnected.

    The students are separate cases, not one thing measured over time.

Answer: Scatter plot

Takeaway: Two numbers per case, one dot per case: that is the data a scatter plot is built for.

Example 2 — Favorite subject counts

Standard

A survey counts how many students prefer each subject. Should this be a scatter plot?
Show the steps and the answer
  1. Identify the variable: favorite subject, such as math, art or history.

    It is a category; there is no number to place on an axis for 'art'.

  2. Count the values per student: just 1.

    A scatter plot needs a pair of numbers for each case, and each student here gives one answer.

  3. Choose a bar graph: one bar per subject, with heights equal to the counts, such as math 8, art 12, history 5.

    Comparing counts of categories is exactly what bar graphs do.

Answer: No

Takeaway: Categories and single variables do not make a scatter plot; it needs two numerical values for every case.

Example 3 — Find the mistake: connecting the dots

Application

Six students record (hours of exercise per week, resting heart rate in beats per minute): (2, 78), (5, 70), (1, 82), (7, 64), (4, 72), (3, 75). Tess plots them, joins the dots in the order listed, and writes: "The line goes up and down, so there is no pattern." Find the mistake and correct it.
Show the steps and the answer
  1. Spot the error: Tess joined the dots in the order the data happened to be written.

    Each dot is a separate student, and a connecting line suggests one thing moving from point to point.

  2. Remove the lines and read the dots from left to right: (1,82), (2,78), (3,75), (4,72), (5,70), (7,64).

    Reading by the x-value shows how heart rate changes as exercise increases.

  3. Describe the pattern: every step to the right goes with a lower heart rate.

    That is a clear negative association, the opposite of 'no pattern'.

Answer: Unconnected, the dots show a clear negative association: more exercise goes with a lower resting heart rate.

Takeaway: Do not join scatter-plot dots; read the overall cloud from left to right to see the pattern.

Where You'll See It

  • Home size against sale price

    An estate agent plots each sold home's floor area against its price. Most dots rise along a band, so bigger homes usually cost more. A dot far below the band, such as a large home priced like a small one, stands out at once as a fixer-upper or a bargain.

  • Predicting Old Faithful eruptions

    Plot each eruption of Yellowstone's Old Faithful geyser as (eruption length, wait until the next eruption) and the dots form two clusters: eruptions of about 2 minutes are followed by waits near 55 minutes, and eruptions of about 4 minutes by waits near 80. Rangers use that pattern to predict the next eruption.

Common Mistakes

  1. The slip: Connecting scatter-plot dots automatically

    Instead: dots show cases, not necessarily a path.

  2. The slip: Using categories as if they were paired measurements

    Instead: scatter plots need numerical x and y values.

  3. The slip: Claiming causation from a visible trend

    Instead: association alone does not prove cause.


Mini Practice

Try these on your own. Tap Reveal when you want to check.

  1. On a scatter plot of age (years) against height (cm), there is a dot at (10,140). What does this dot represent?

    Hint: Each dot is one person with two measurements.

  2. Plot the points (hours studied, score): (1,60),(2,65),(3,72),(4,80),(5,84). Describe the trend.

    Hint: Look at how the score changes as hours increase.

  3. Car age (years) and price (thousands of dollars): (1,28),(3,22),(5,17),(7,12),(9,8). Describe the association and estimate the price of a 4-year-old car.

    Hint: Find the two plotted ages on either side of 4 years.

  4. The points (1,3),(2,5),(3,7),(4,18),(5,11),(6,13) are plotted. Which point does not fit the pattern?

    Hint: Look at how much y grows each time x goes up by 1.

  5. You record your town's temperature every hour for one day. Should you use a scatter plot or a line graph?

    Hint: Is this one quantity changing over time, or two measurements of many cases?

  6. Raj plots height against arm span for 8 classmates, then joins the dots from left to right and says the line shows how one student grew. Correct his thinking.

    Hint: Ask what each dot stands for.

Want the full set?

50 practice questions for this concept — free to try, every one with a complete worked solution showing the why, not just the answer.

Frequently Asked Questions

Which variable should go on the horizontal axis?

Put the variable you think might explain or predict the other on the horizontal axis, and the one that responds on the vertical axis. Hours of practice go across and points scored go up, because practice might affect scoring, not the other way round. If neither variable comes first, as with a person's arm span and height, either choice is fine: the picture is the same, only flipped.

How many points do I need before a trend means anything?

More than you might think. With only four or five points, even random numbers often seem to line up, just as a few coin tosses can come up heads every time. With 20 or 30 points, a trend that still shows is much harder to get by luck. If you only have a handful of cases, describe the trend as a hint to check with more data, not as a finding.

What can a scatter plot show that a single correlation number hides?

Quite a lot. In 1973 the statistician Francis Anscombe made four small data sets with the same means and the same correlation, about 0.82. Plotted, one is a straight band, one is a smooth curve, one is a perfect line with a single outlier, and one is a vertical stack with one far-off point. The number was identical; the stories were completely different. That is why you plot first and calculate second.

Learning Path

← Before

Coordinate Plane
Scatter Plot

You are here

You need the Coordinate Plane, to place each pair as a point (x,y). A scatter plot adds a picture of how two variables relate across many cases. Next, Correlation puts a number on the direction and strength of the pattern, and Inference for Regression asks whether a trend seen in a sample holds for a whole population.

See Also