SAT one-variable data: mean, median and spread
One-variable data questions ask where a data set is centered and how spread out it is. Read the display carefully first; the arithmetic is usually short.
These are the concepts this skill rests on, the same pages as in the lesson.
Measures of Center
The Mean Is the Balance Point
The mean of a data set is the sum of the values divided by the number of values. For 2, 3, 5 and 10, the sum is 20 and there are 4 values, so the mean is .
Think of the mean as the point where the data balance. The values below 5 sit units under it, and the value above sits 5 units over it, so the two sides cancel out. That's why one far-off value, like the 10, pulls the mean toward itself.
The Median Is the Middle
The median is the middle value once the data are in order. Always sort first. The list 7, 2, 9, 4, 6 has 9 printed in the middle, but in order it's 2, 4, 6, 7, 9, so the median is 6.
With values, the median is the th value in order. When is even, that position lands between two values, and the median is their mean. For 3, 5, 8, 12 it's , which isn't one of the values at all.
One List, Four Answers
One list can give four different answers, so the measure the question names decides which one you want. For 1, 2, 2, 3, 5, 6, 9:
| Measure | What it is | Here |
|---|---|---|
| mean | sum ÷ count | |
| median | the middle value in order | 3 |
| mode | the value that occurs most often | 2 |
| range | greatest value minus least value |
Circle the measure word before you compute anything. On the SAT, "average" means the mean.
A Frequency Table Is a Compressed List
A frequency table lists each value once, with a count of how many times it shows up. Each value counts as many times as its frequency, so the mean is a weighted mean.
| Value | Frequency |
|---|---|
| 1 | 4 |
| 2 | 5 |
| 3 | 1 |
- Multiply each value by its frequency and add: .
- Divide by the total frequency, 10, not by the 3 rows: the mean is 1.7.
A table of percents instead of counts works the same way. The percents are the weights, and they add up to 100. With 50%, 30% and 20% for the values 1, 2 and 3, the mean is .
Count In to Find the Median
To find the median from a frequency table, work out the middle position first. Then add up the frequencies until you reach it.
| Value | Frequency |
|---|---|
| 2 | 5 |
| 3 | 8 |
| 4 | 6 |
| 5 | 4 |
| 6 | 2 |
There are values, so the median is the 13th. The running counts go 5, 13, …, so the 6th through 13th values are all 3, and the median is 3. With an even total such as 20, you'd average the 10th and 11th values.
The middle row of the table, 4, isn't the median. The rows hold different numbers of values.
Reading Data Displays
Dot Plots and Bar Graphs
A dot plot and a bar graph show the same thing as a frequency table. The horizontal axis gives the values, and the height of a stack or bar tells you how many times each one shows up.
Both pictures show the same ten values: two 1s, four 2s, three 3s and one 4. Three values equal 3, and values are greater than 2. Everything you do with a frequency table works here too: weight by the heights for a mean, and count in for a median.
Two misreadings are common. If the question asks how many, the answer is a height, not a value on the horizontal axis. And watch the wording: "more than 2" leaves out the 2s, while "at least 2" includes them.
Histograms Group Values Into Bins
A histogram is a bar graph of bins: each bar counts the values that fall in an interval. The question tells you which edge a bar includes. Here each bar includes its left edge but not its right, so the second bar counts values from 10 up to, but not including, 20.
A histogram shows bins, not the values themselves, so you can't get an exact mean or median from it. You can tell where they could be, though:
- The median's bin. There are 10 values, so the median is the mean of the 5th and 6th. The first bar holds the 1st and 2nd values, and the next holds the 3rd through 6th. Both middle values are in the bin from 10 to 20, so the median is too.
- Limits on the mean. Say the values are whole numbers. For the smallest possible mean, put every value at the bottom of its bin: . For the largest, put every value at the top, 9, 19, 29 and 39: . So the mean could be 15, but it can't be 25.
A Box Plot Shows Five Numbers
A box plot boils a data set down to five numbers: the minimum, the first quartile , the median, the third quartile and the maximum. A quartile is a quarter: 25%. The quartile cutoffs split the ordered data into four roughly equal groups.
Think of the five marks as the 0th, 25th, 50th, 75th, and 100th percentiles, in order. , the 25th percentile, is a cutoff with about 25% of the values below it; the median has about 50% below it, and about 75%. The minimum and maximum mark the two ends. With repeated values or a small data set, the shares strictly below each mark need not be exact.
Read each mark where its line meets the scale. Here the dots above the plot are the 12 values it summarizes.
- The box runs from to . It holds the middle half of the data, and its width is the interquartile range: .
- The whiskers reach out to the minimum and the maximum, so the range is .
- The dashed lines split the dots into four groups of 3. Each part of the plot holds about a quarter of the values, however long or short it's drawn. The right whisker is the longest part, yet it holds no more values than the short left half of the box.
A box plot does not show the mean, how many values there are, or any value other than those five. Without the dots, you couldn't tell there were 12.
Measures of Spread and Comparison
Range and Interquartile Range
The range is the greatest value minus the least. The interquartile range (IQR) is , the width of the middle half. To find the quartiles, split the ordered data at the median, then find the median of each half.
With an odd number of values, leave the median out of both halves. Here 10 is the median of nine values. The lower half, 2, 5, 6, 8, has median , and the upper half, 11, 13, 14, 30, has median . So the IQR is 8, and the range is .
That one far value, 30, sets the range all by itself. The IQR ignores the extremes, so it would still be 8 whether the greatest value were 15 or 300.
Standard Deviation Measures Clustering
The standard deviation (SD) measures how far the values typically sit from their mean. Values bunched close to the mean give a small SD, and values far from it give a large one. On the SAT you compare standard deviations; you never have to compute one.
Both sets have mean 5. Every value in Set A is within 1 of 5, while every value in Set B is 2 or 4 away. So Set B has the larger standard deviation.
To compare two sets, ask whose values typically sit closer to their own mean. If you slide a whole set along the scale by adding the same number to every value, its standard deviation doesn't change.
A Wider Range Is Not Always More Spread
The range looks only at the two extremes. The standard deviation looks at every value's distance from the mean. So the set with the wider range can have the smaller standard deviation.
Both sets have mean 5. Set P reaches farther, from 0 to 10, but six of its eight values sit right at the mean, and only two are far from it. Every value in Set Q is 3 from the mean. A standard deviation counts far values extra, so a few far values can outweigh many close ones, but here they don't. Work it out and Set P's standard deviation is 2.5 while Set Q's is 3. Set Q has the larger standard deviation, even though its range, 6, is smaller.
How many values a set has isn't a measure of spread either: more dots doesn't mean more spread.
Move P's two end values with the slider above. At distance 5 from the mean, P's range is 10 and its SD is 2.5. Move them to distance 6: P's range is now 12, twice Q's, but their SDs only tie at 3. Move them farther and P finally has the larger SD. The readouts do the calculating; your job is to compare the dots.
Comparing Two Data Sets
To compare two data sets, take one claim at a time. Name the measure it's about, then decide it from what you're given.
- Center: Set B's median, 20, is greater than Set A's, 16.
- Spread: both IQRs are 10, but Set A's range, , is greater than Set B's, 18.
- Can't tell: which set has the greater mean, or more values. A box plot doesn't show either one.
Compare center and spread separately, because a set can have the greater center and the smaller spread. When a question asks what must be true, it wants a claim the data actually prove. Throw out any claim about a measure the display or table doesn't show.
Shape and Change
Symmetric and Skewed
The shape of a distribution is how its values pile up. A symmetric distribution is a mirror image of itself around its center, so its mean and median are equal. A skewed distribution has a long, thin tail on one side, and it's named for the side the tail is on.
A long tail usually pulls the mean toward it, while the median stays with the crowd. So skewed right usually puts the mean above the median, and skewed left puts it below. For 1, 2, 2, 3, 12, the median is 2 but the mean is 4.
That's a rule of thumb, not a law. It can fail when the other side has a short but heavy tail, or when a big pile sits just past the median. For 1, 2, 3, 3, 3, 3, 4, 4, 4, 4, the tail points left, yet the median is 3 and the mean, 3.1, is above it, because the four 4s sit just past the middle. When you're given the counts, compute both measures before you compare them.
What One Extreme Value Does
Adding or removing one value moves the mean, unless that value equals the mean. The farther the value is from the mean, the more the mean moves. The median can only move as far as the middle position shifts, which is half a place. So the new median lands between the old median and the value next to it.
Adding 20 to 3, 4, 4, 5, 6 drags the mean from 4.4 all the way to 7, but nudges the median only from 4 to 4.5. The range jumps from 3 to 17, while the IQR stays 2.
Don't just assume the median stays put. Count to it. If the two middle positions hold the same value before and after, the median doesn't move. If they hold two different values and the new value lands outside them, the median moves from their midpoint onto one of them: half the gap.
Now move one value rather than adding another. The widget keeps four values at 3, 4, 4, and 5. Move the fifth from 6 to 10: the mean rises from 4.4 to 5.2, but the median stays 4. Move it down to 0: the mean drops to 3.2 and the median is still 4. For this list, the two fixed 4s keep the middle value in place.
Changing Every Value
When every value changes the same way, each measure changes in a way you can predict.
| Change to every value | Mean and median | Range, IQR and SD |
|---|---|---|
| add a number | increase by | unchanged |
| multiply by a positive number | multiplied by | multiplied by |
Adding 5 slides the whole picture 5 to the right. The mean goes from 3.6 to 8.6, and the range stays 4, because the gaps between the values don't change. Multiplying by 2 doubles every gap, so the mean goes to 7.2 and the range to 8. Converting units, such as kilograms to grams, is multiplying.
Working With Totals
Mean Times Count Is the Total
Flip the definition of the mean around, and a mean and a count give you a total:
The values 3, 7, 5 and 9 have mean 6, so their total is . In the picture, the parts sticking up above the mean line exactly fill the gaps below it.
That's how you find a missing value. For five values to have a mean of 7, they must add up to . If four of them are 3, 7, 5 and 9, the fifth is . Remember to count the new value: the target total uses 5, not 4. A removed value works the same way: the total before minus the total after is the value that left.
Combining Two Groups
To find the mean of two groups together, add their totals and divide by the combined count. Don't average the two means unless the groups are the same size.
Six values with mean 10 total 60, and two values with mean 30 also total 60. All eight total 120, so their mean is , not . The combined mean always lands between the two group means, and when the groups are different sizes, it lands closer to the bigger group's mean.
When you don't know a count, call it a letter and write each total with it. For values with mean 10 and values with mean 25, the mean of all of them is . The letter cancels out.
Keep going
Build the basics first
Foundations has free readiness checks and lessons in the English and math the SAT expects.
Check Your Skills, the strategy for each kind of question, worked examples and practice from Easy to Hard are in the Quaero SAT Course.
More Problem-Solving and Data Analysis
Ratios, rates and units
Set up ratios and proportions, work with rates, convert units by cancelling them, and recognize proportional relationships.
Read the concepts →SAT essentials · MathPercentages
Learn what a percent means, find shares of a total, and apply one percent change or several in a row.
Read the concepts →SAT essentials · MathTwo-variable data
Read two-variable data, predict with a line of best fit, understand residuals, interpret slope and intercept in context, and choose a model.
Read the concepts →SAT essentials · MathProbability
Find a probability as a part over a whole, combine events, read two-way tables, and handle conditional probability and independent events.
Read the concepts →SAT essentials · MathSample statistics and margin of error
Tell a sample from a population, build an interval from an estimate and its margin of error, see why sample size matters, and scale results up.
Read the concepts →SAT essentials · MathEvaluating statistical claims
Tell observational studies from experiments, and see what random selection and random assignment each allow a study to conclude.
Read the concepts →