SAT Math: Problem-Solving & Data Analysis
Problem-Solving and Data Analysis makes up roughly 15% of Digital SAT Math questions and focuses on quantitative reasoning applied to real-world data: ratios and rates, percentages, statistical measures, scatterplots, and probability.
Unlike Algebra and Advanced Math, this domain rarely requires solving multi-step equations; instead it rewards careful reading of data presented in tables, graphs, and word problems, plus fluency with a handful of core statistical concepts.
Last updated · Reviewed by the MyScorePrep content team
Ratios, rates, and proportions
Ratio problems ask you to compare quantities or scale a relationship up or down. Set up a proportion by keeping the same units aligned in the same position in each fraction, then cross-multiply to solve. Rate problems (speed, price per unit, work per hour) follow the same logic: identify the rate as a ratio of two quantities and multiply or divide as the scenario requires.
Worked example
A recipe uses 3 cups of flour for every 2 cups of sugar. How many cups of flour are needed for 10 cups of sugar?
- Set up a proportion: 3 flour / 2 sugar = x flour / 10 sugar
- Cross-multiply: 2x = 3 × 10 = 30
- Divide both sides by 2: x = 15
Answer: 15 cups of flour
Percentages and percent change
A percentage is a ratio out of 100, so 'x percent of y' translates to (x/100) × y. Percent change is calculated as (new value − original value) / original value, expressed as a percentage; a positive result is a percent increase and a negative result is a percent decrease. Watch for problems that apply successive percent changes (for instance, a price increased by 20% and then decreased by 10%), since these do not simply combine additively — you must apply each change to the updated value in sequence.
Worked example
A $80 jacket is marked up 25%, then later discounted 20% off the marked-up price. What is the final price?
- Apply the markup: 80 × 1.25 = 100
- Apply the discount to the new price: 100 × (1 − 0.20) = 100 × 0.80 = 80
Answer: $80 (the two percent changes cancel out numerically here, even though 25% and 20% are different percentages)
Unit conversion
Unit conversion questions give a conversion factor (such as 1 mile = 5,280 feet) and ask you to convert a quantity or rate between units. Set up the conversion as a fraction equal to 1 (for example, 5,280 feet / 1 mile) and multiply so that the unwanted unit cancels, leaving the desired unit. This dimensional-analysis approach is especially reliable for multi-step conversions involving rates, like converting miles per hour to feet per second.
One- and two-variable data
One-variable data questions involve a single list of values and ask about measures of center and spread: mean (the sum divided by the count), median (the middle value when sorted), and standard deviation (a measure of how spread out the values are around the mean — a larger standard deviation means the data is more spread out, not necessarily larger on average). Two-variable data questions involve paired values, often shown in a table or scatterplot, and ask about the relationship between the two variables.
For standard deviation questions, the SAT rarely asks you to calculate the exact value; instead it asks you to compare which of two data sets has a larger or smaller standard deviation by reasoning about how tightly the values cluster around the mean.
Scatterplots and lines of best fit
A line (or curve) of best fit summarizes the general trend in a scatterplot and lets you estimate values, interpret slope as a rate in context, and identify outliers — points that fall unusually far from the trend. When asked to use a line of best fit to make a prediction, plug the given x-value into the line's equation just as you would with any linear function, and treat the resulting y-value as an estimate rather than an exact data point.
Probability and conditional probability
Basic probability is (number of favorable outcomes) / (total number of outcomes). Conditional probability questions restrict the total outcomes to a specific subgroup — for example, 'given that a selected student is a senior, what is the probability the student plays a sport' restricts the denominator to seniors only, not the entire group. Two-way frequency tables are the most common format for these questions: locate the correct row or column for the given condition, then divide the specific cell by that row or column's total.
Worked example
In a class of 30 students, 18 are seniors and 12 are juniors. Of the seniors, 10 play a sport. If a senior is chosen at random, what is the probability that student plays a sport?
- The condition restricts the sample space to seniors only: 18 students.
- Favorable outcomes are seniors who play a sport: 10.
- Probability = 10/18 = 5/9.
Answer: 5/9
Sample statistics, margin of error, and study design
Questions on sampling ask you to evaluate whether a study's conclusions can be generalized. A key idea: a sample statistic (like a sample mean or proportion) is only a reliable estimate of a population parameter if the sample was selected randomly and is reasonably large; results from a non-random or small sample should not be generalized to a broader population. Margin of error describes the range within which the true population value likely falls; a larger sample size generally produces a smaller margin of error, meaning a more precise estimate.
For questions about establishing cause and effect, remember that only a randomized controlled experiment (with random assignment to treatment and control groups) can support a causal claim; an observational study, no matter how large, can only support a claim of association.
Common mistakes to avoid
- Applying successive percent changes by simply adding or subtracting the percentages instead of applying each change to the updated value in order.
- Computing conditional probability using the full total instead of restricting the denominator to the given condition's subgroup.
- Confusing mean and median, especially in skewed data sets where the two values differ noticeably.
- Assuming a strong correlation in a scatterplot implies causation, when only a randomized experiment can support a causal claim.
- Mixing up which value increases or decreases standard deviation when data points are added or removed.
- Setting up unit conversions with the conversion factor upside down, so units fail to cancel correctly.
Frequently asked questions
Start free practice
Practice SAT Problem-Solving and Data Analysis questions with instant feedback.
Keep going
MyScorePrep is an independent SAT preparation platform and is not affiliated with or endorsed by College Board.