SAT guideStudy & preparation

SAT® Problem-Solving and Data Analysis: Rates, Percentages and Data

Published 7 min read3 official sources

Short answer

College Board lists 5–7 Problem-Solving and Data Analysis questions across the SAT Math section, covering quantitative relationships, data and statistical reasoning. Solve them by identifying the requested quantity and units, choosing the correct base or denominator, and checking what the result means. A correct calculation still needs an interpretation supported by the data and study design.

Which quantities and units should you identify first?

College Board lists 5–7 Problem-Solving and Data Analysis questions across the SAT Math section. To solve them, identify the quantity requested, keep its units visible, and check that your denominator and conclusion match the question.

The section count was checked November 4, 2026. It describes the whole Math section, not each module or a fixed score weight. The domain appears in both Math modules.

College Board's detailed skill descriptions, checked November 11, 2026, cover ratios, rates and units; percentages; one-variable distributions and summary statistics; two-variable models; probability; sample inference and margin of error; and claims based on study design. The worked examples below focus on five useful setups within that scope. All situations and data are fictional, original teaching examples, not official SAT questions.

Before calculating, write a short answer label: dollars per notebook, meters per second, percent change, mean, or probability within a named group. Then underline the information that belongs in that calculation. This is an editorial practice routine, not an additional test requirement.

When the main obstacle is forming or solving a linear equation, use the SAT Algebra guide. Here, start with what each number represents.

How do ratios, rates and percentages change the setup?

Example 1: Find a unit rate, then check the units

Suppose 12 identical notebooks cost $30, with a constant price per notebook and no additional charges. What would 7 notebooks cost?

Unit price = $30 ÷ 12 notebooks = $2.50 per notebook.

Cost of 7 = 7 notebooks × $2.50 per notebook = $17.50.

The ratio 12 ÷ 30 gives 0.4 notebooks per dollar. That is a valid rate, but it is not dollars per notebook. Multiplying 7 notebooks by that reversed rate does not produce the requested cost. Read the units before choosing which quantity goes on top.

For a separate conversion, express 72 kilometers per hour in meters per second:

72 km/hour × 1,000 meters/km × 1 hour/3,600 seconds = 20 meters/second.

Kilometers and hours cancel, leaving the requested units. Multiplying by 1,000 alone would give meters per hour; the time unit would still need conversion. After any conversion, check the unit on your final answer rather than stopping when the arithmetic is finished.

Example 2: Identify the base of each percentage

A quantity starts at 100, increases by 20%, then decreases by 20%. Does it return to 100?

  1. The increase is 0.20 × 100 = 20, so the new quantity is 120.
  2. The decrease uses 120 as its base: 0.20 × 120 = 24.
  3. The final quantity is 120 − 24 = 96.

The two percentages do not cancel because their bases differ. Subtracting 20 in the second step would incorrectly reuse the original base.

Percentage points answer another question. If a proportion rises from 40% to 50%, it rises by 10 percentage points. Its relative percentage increase is:

(50 − 40) ÷ 40 × 100% = 25%.

The difference is 10 points; the original 40 is the base for relative change. Before calculating, decide which of those quantities the question requests. Labeling both answers during practice makes the distinction easier to check.

When comparing future preparation options for the skills you need, see how prep for the SAT® exam, planned for launch, covers Problem-Solving and Data Analysis.

What can a data table tell you about center and probability?

Example 3: Choose the requested summary statistic

Consider the ordered data 2, 2, 3, 5, 8.

Requested quantityCalculation or selectionAnswer
Mean(2 + 2 + 3 + 5 + 8) ÷ 54
MedianThe middle observation in the ordered list3
RangeLargest value minus smallest: 8 − 26

There are five observations, even though only four different values appear. Dividing the total 20 by four would ignore that the value 2 occurs twice. For the median, use the middle observation after ordering the values; do not average the largest and smallest values.

These statistics describe the supplied data. They do not explain why the values differ or establish a claim about a wider population. First check which statistic is requested, then count every observation it uses.

Example 4: Choose the correct probability denominator

A fictional group of 60 people gives these responses:

GroupYesNoTotal
A181230
B121830
Total303060

Question A: If one person is chosen at random from Group A, what is the probability that the person answered Yes?

Only Group A is eligible for selection. The answer is 18 ÷ 30 = 0.60, or 60%. This is the probability of Yes given Group A.

Question B: If one person is chosen at random from all 60 people, what is the probability that the person belongs to Group A and answered Yes?

Now all 60 people are eligible. The answer is 18 ÷ 60 = 0.30, or 30%.

Both questions use the same numerator, 18. Their denominators differ because the selection groups differ. Using 60 for Question A would answer Question B instead. Circle the words that restrict the group before taking a fraction from the table.

If your review reveals repeated interpretation errors, look for explanations that address the setup; prep for the SAT® exam, planned for launch, includes an explanation for every option.

How do sample results differ from a causal claim?

Example 5: Separate a population estimate from a cause

Suppose a random sample of 100 residents from one town includes 62 who favor a proposed park. The sample proportion is 62 ÷ 100 = 62%.

That proportion estimates support among the town's residents, with sampling uncertainty. It does not establish that exactly 62% of all town residents favor the proposal. No margin of error is supplied in this example, so do not invent a numerical interval.

It also does not show that building the park would cause a health improvement. The survey measured opinions about a proposal, not the effects of building it.

College Board's study-design skills distinguish two uses of randomness:

  • Random sampling concerns who enters the sample and supports generalizing from it to the population sampled.
  • Random assignment concerns which treatment group participants enter and provides evidence for a causal comparison.

Random selection of residents is not random assignment to treatments. Likewise, an observed association alone does not establish causation. A survey answered only by volunteers does not, by itself, justify treating its percentage as representative of the entire town. Random assignment provides causal evidence; it does not remove all uncertainty from an experiment.

For an inference question, check how people were selected, whether treatments were assigned, and what was actually measured. Then choose a conclusion that stays within those limits.

How do you choose the next skill to practice?

Use the error in your setup to choose a narrow practice task. The following is an editorial review plan:

If your mistake was…Practice nextCheck before answering
Reversing a rate or missing a conversionRatios, rates and unitsDo the remaining units match the requested quantity?
Reusing an old percentage basePercentagesWhat quantity is this percentage of?
Confusing mean, median and rangeOne-variable dataWhich statistic is requested, and how many observations are there?
Dividing by the whole table automaticallyConditional probabilityWhich group is eligible under the stated condition?
Turning a survey or association into a causal claimStudy design and inferenceWhat conclusion does the selection and assignment method support?

College Board's Student Question Bank, checked November 4, 2026, supports filters for assessment, section, domain, skill and difficulty. Select SAT, Math and Problem-Solving and Data Analysis, then narrow to the skill that caused your mistake. The SAT practice-question selection guide explains how to organize targeted work.

Attempt fresh questions, explain the denominator or base in writing, and review the first step that went wrong. A narrow drill result is not a scaled SAT score. If your difficulty instead concerns nonlinear equations or equivalent expressions, move to the SAT Advanced Math guide.

For your next session, pick one row from the table and practice that skill before mixing it with others. If you are planning additional preparation, join the launch list for prep for the SAT® exam rather than counting on it for a fixed date.

Trust the source

Official sources

Exam policies can change. Use these primary sources for the most current details.