Digital SAT Math · Problem-Solving & Data Analysis

One-variable data (center & spread)

A whole column of the test hands you one variable drawn as a picture — a dot plot, a histogram, a box plot, a frequency table — and asks a question you answer by counting, not by calculating. Almost nothing here needs a formula. What it needs is the discipline to find the middle observation instead of the middle of the axis, and to weight a table by its frequencies instead of averaging the labels.

On the test

Recognition cues: dot plot, histogram, frequency, box plot, median, mean, range, standard deviation, must be true, greater spread, after the value was corrected.

Pattern recognition

Name the display before you read the question. The display decides the method, and there are only four of them.

1. List — values printed in a sentence. Sort first, always. The median is the middle observation, not the middle of the printed order.
2. Dot plot — one dot per observation stacked over its value. The dots are the frequencies; count them.
3. Frequency table — a value column and a count column. Every statistic must be weighted by the count column.
4. Box plot — the five-number summary and nothing else. It shows min, Q1, median, Q3, max. It does not show the mean, the number of observations, or any individual value.

Method

1. Name the display and the total. Write down $n$ — the number of observations, not the number of bars, dots-per-stack, or table rows. On a dot plot, $n$ is the total number of dots; on a frequency table, $n$ is the sum of the count column.
2. Median by cumulative count. The middle observation sits at position $\frac{n+1}{2}$ when $n$ is odd, and is the average of positions $\frac{n}{2}$ and $\frac{n}{2}+1$ when $n$ is even. Run a running total up the value column until you pass that position, then read the value you landed on.
3. Mean by weighting. From a frequency table or a dot plot,

The denominator is $n$, never the number of distinct values. 4. Spread by looking. Range is $\text{max} - \text{min}$. Standard deviation is not the range — it measures typical distance from the mean, so judge it by clustering: the set whose values hug the center has the smaller standard deviation. You will never need to compute one. 5. Answer the asked statistic. Median, mean, range, interquartile range and standard deviation are five different answers to five different questions, and the choice list usually contains at least three of them.

Practice

Answer before you open the explanation. Two items are student-produced response (type the number, no choices), matching the real test, and five are built on a display — a dot plot, a histogram, a box plot or a frequency table — because that is how the test actually presents this skill. Every wrong choice below is a specific error with a name.

Common mistakes

1. Median read off the middle of the axis — the midpoint of the number line under a dot plot, or the middle bar of a histogram, has nothing to do with where the observations pile up. Count.
2. Median taken from the printed order — a list in a stem is almost never sorted, and the value sitting fourth in the sentence is not the fourth-smallest.
3. Position reported instead of value — with 25 observations the median is at position 13; the answer is the value that observation 13 holds, not 13.
4. Mean computed without the frequencies — averaging the value column of a frequency table treats a row representing 9 crews and a row representing 2 crews as equals.
5. The wrong column averaged — dividing the total number of observations by the number of table rows produces a number with no meaning, and it is always in the choice list.
6. Range treated as standard deviation — the range sees only the two endpoints, so two sets can share a range and differ sharply in spread. Judge standard deviation by how tightly values cluster around the mean.
7. Assuming an outlier drags the median as far as the mean — a far-away value moves a balance point a long way and a middle position hardly at all.
8. Adding a constant to every value assumed to change the spread — shifting the whole set moves the mean and the median by that constant and leaves the standard deviation exactly where it was.
9. Reading a mean off a box plot — the line inside the box is the median. A box plot shows five numbers, none of which is the mean and none of which is $n$.
10. Answering a different statistic — the median when the stem said mean, the interquartile range when it said range, or the modal value’s frequency when it asked for the value.

FAQ

Do I ever have to compute a standard deviation? No. The Digital SAT asks you to compare or interpret spread, never to calculate it, and no standard-deviation formula appears on the reference sheet. If a question seems to demand the number, you have misread it — re-read for the word greater, smaller or unchanged.

What is the difference between range and standard deviation? Range is a subtraction between the two most extreme values; standard deviation summarises how far a typical value sits from the mean. One outlier can inflate the range without much affecting how clustered the rest of the data is, which is why the two can disagree, and why the test enjoys building an item on exactly that disagreement.

A box plot is on screen and the question asks for the mean. Now what? It cannot be answered. A box plot reports min, Q1, median, Q3 and max; individual values and the count are invisible, so the mean is not recoverable. On a “must be true” item, any statement about the mean of a box-plotted set is a distractor unless the stem supplies the values separately.

How do I type these answers into a grid-in? Type the number the last line asks for and nothing else — no units, no percent sign. A weighted mean often lands on a decimal such as 3.95, and the grid accepts it; it also accepts an equivalent fraction. If your answer is a count of observations, it will be a whole number, which is a useful sanity check on the arithmetic.

Does the mode show up? Rarely, and never as the point of a question. The tallest stack on a dot plot is worth noticing because it tells you where the data is dense, but College Board’s skill is named for center and spread. Do not report the height of the tallest stack when a stem asks for a value.

Editorial team · last reviewed 2026-08-12