SAT Math · Problem-Solving and Data Analysis

Mean and median

How to find each one, and why an outlier pulls the mean but barely moves the median.

Updated

An outlier at 20 drags the mean to 6. The median stays at 4.5, inside the cluster.02468101214161820median 4.5mean 68 values

mean=sumcount\text{mean} = \dfrac{\text{sum}}{\text{count}}

An outlier at 20 drags the mean to 6. The median stays at 4.5, inside the cluster.

What are the mean and the median?

mean=sumcount\text{mean} = \dfrac{\text{sum}}{\text{count}}median=middle of the sorted list\text{median} = \text{middle of the sorted list}

Both are one number standing in for the middle of a data set. The mean shares the total out equally across every value. The median is the value that splits the sorted list into two equal halves.

Seven values: median 4 (the fourth value), mean 5 (35 shared seven ways).012345678910median 4mean 57 values
Seven values: median 4 (the fourth value), mean 5 (35 shared seven ways).

The mean is where the data balances

below: (5−1)+(5−4)=5\text{below: } (5-1)+(5-4) = 5above: (6−5)+(9−5)=5\text{above: } (6-5)+(9-5) = 5

Put the dots on a seesaw. It balances at the mean, because the distances below the mean add up to exactly the distances above it. Every value pushes on the mean, and the farther away it sits, the harder it pushes.

1, 4, 6 and 9 balance on a fulcrum at 5.012345678910mean 54 values
1, 4, 6 and 9 balance on a fulcrum at 5.

An outlier drags the mean, not the median

mean 4→6\text{mean } 4 \to 6median 4→4.5\text{median } 4 \to 4.5

Start with 2, 3, 3, 4, 5, 5, 6: mean 4, median 4. Add a 20 and the seesaw must shift far right to balance, so the mean jumps to 6. The median only counts positions, so it barely moves.

Adding 20 drags the mean to 6; the median only reaches 4.5.02468101214161820median 4.5mean 68 values
Adding 20 drags the mean to 6; the median only reaches 4.5.

Why outliers move the mean

Answer each question in your head, then scroll for the answer.

An outlier at 20 drags the mean to 6. The median stays at 4.5, inside the cluster.02468101214161820median 4.5mean 68 values

Think it through

  1. Question 1

    Data: 2, 3, 3, 4, 5, 5, 6, 6. Change the last 6 to 20. What happens to the sum?

  2. Answer

    It grows by 14, so the mean grows by 14÷8=1.7514 \div 8 = 1.75.

  3. Question 2

    Does the median change?

  4. Answer

    No. The two middle values are still 4 and 5, so it stays 4.5.

  5. Question 3

    So which is better for incomes or home prices?

  6. Answer

    The median, because a few huge values pull the mean up.

Try it yourself

Work it out first, then check your answer. SAT-style questions, not official College Board questions.

A dot plot of 4, 7, 8, 10, and 11.0123456789101112median 8mean 85 values

Question 1

Find the mean and median of 4, 7, 8, 10, 11.

A dot plot of six values, with the median between 6 and 9.01234567891011121314median 7.56 values

Question 2

What is the median of 9, 3, 12, 5, 13, 6?

A dot plot of five test scores with a mean of 87.707580859095100mean 875 values

Question 3

The mean of 5 test scores is 87. What must a sixth score be for the mean of all 6 to be 88?

A dot plot of seven values with an outlier at 45.101520253035404550median 15mean 19.37 values

Question 4

The data set 12, 14, 15, 15, 16, 18, 45 has an outlier at 45. If 45 is removed, which decreases more, the mean or the median?

Stuck on a question like this?

Talk it through with Aristotle. It asks guiding questions instead of giving the answer. Intro is free with three sessions.

Try Aristotle free

The definitions

Mean

mean=sum of valuesn\text{mean} = \dfrac{\text{sum of values}}{n}

nn is the number of values. Every value affects the mean, including outliers.

Median

middle value, in order\text{middle value, in order}

With an even count, average the two middle values.

Sum from the mean

sum=mean×n\text{sum} = \text{mean} \times n

The key step for questions that add or remove a value and ask for the new mean.

Range

range=max⁡−min⁡\text{range} = \max - \min

A simple measure of spread. Standard deviation measures how far values typically sit from the mean.

Which way the mean moves
Shape of the dataMean vs. medianBetter center
symmetricabout equaleither
skewed right (high outlier)mean>median\text{mean} > \text{median}median
skewed left (low outlier)mean<median\text{mean} < \text{median}median

Try it

Drag the outlier

Drag the last value and watch which measure follows it.

A dot plot of eight values. The last value is 6.051015202530median 4.5mean 4.38 values
6

Data2, 3, 3, 4, 5, 5, 6, 6

Mean34 ÷ 8 = 4.25

Median(4 + 5) ÷ 2 = 4.5

Drag the last value out to 30. The mean chases the outlier; the median barely moves.

Common mistakes

Not sorting before finding the median

Put the values in order before you pick the middle.

Picking one middle value from an even count

With six values there is no single middle. Average the third and fourth values.

Averaging the means of two groups

For groups of different sizes, add the sums and divide by the total count.

Reading a frequency table as a list

“3 appears 4 times” adds 3 four times and counts as four values.

How it shows up on the SAT

Problem-Solving and Data Analysis is about 15% of SAT Math. Expect dot plots and tables, a missing value from a mean, and which measure an outlier changes.

What Aristotle tracks in one-variable data: distributions and measures of center and spread

5 SAT Math skills, including:

  • Interpreting data distributions from tables, histograms, dot plots, and box plots

  • Calculating and comparing mean, median, and range

  • Comparing distributions with different means and the same standard deviation

  • Comparing distributions with different standard deviations

Questions students ask

A few things you might wonder before getting started.

What is the difference between the mean and the median?

The mean is the arithmetic average: add all the values and divide by how many there are. The median is the middle value when the data is in order. The mean is sensitive to outliers, and the median is not.

How do you find the median with an even number of values?

How does an outlier affect the mean and median?

When is the median better than the mean?

How do you find a missing value from the mean?

How can Aristotle help with SAT data analysis?

A learner having an idea, surrounded by maps and equations

Practice it out loud

Talk it through with Aristotle

Work a question like these out loud. Aristotle guides you one step at a time, so you can do the next one on your own.