GRE Data Interpretation

Data Interpretation (DI) is less about difficult arithmetic than careful reading. A chart may contain every number you need, but a missed unit, category, or time period can make a perfectly calculated result wrong. Before reaching for a formula, translate the question into a sentence: which group, measure, and comparison does it ask for? Then find those data in the display.

Read the display before the question details

Start with the title, labels, units, time period, and notes. Determine whether values are dollars, thousands of dollars, percentages, people, or index points. Check whether the chart reports annual totals, monthly averages, or a percentage of a separate group. These labels determine how values can be combined.

Inspect the scale on each axis. A bar chart may begin above zero, which makes visual differences look larger. A line graph may use uneven intervals or plot several series with different units. A circle graph represents parts of a whole, so percentages should sum to 100% apart from rounding. Never infer an exact value from bar height when a labeled table or axis gives the intended reading.

Read footnotes and symbols. A star may identify an estimate, a blank cell may mean no data rather than zero, and parentheses may indicate a negative value. If the chart says ‘in thousands,’ the displayed value 24 means 24,000 in the stated unit. Keep that multiplier attached until the comparison is complete.

Translate the prompt into a calculation

Most DI calculations use familiar arithmetic: totals, differences, ratios, averages, percentages, or changes over time. Write the requested quantity before substituting numbers. If a question asks for the percentage increase from 40 to 50, the denominator is the original value, 40. If it asks what percent of the final value the increase represents, the denominator is 50. Those are different questions.

Keep the units in the setup. If the numerator is in millions and the denominator is in thousands, convert before dividing or explicitly account for the factor of 1,000. Ratios may cancel units when both quantities use the same unit, but differences do not. A difference of 12 million is not the same as a difference of 12 thousand.

Original practice table

A regional library system reports the following number of program visits. Values are in hundreds of visits.

Question 1: total visits

How many total visits were reported across all programs in 2025?

1. 6,400
2. 6,800
3. 64,000
4. 680

Answer: A. Add the displayed values: 24 + 27 + 13 = 64. The table reports hundreds, so the total is 64 × 100 = 6,400 visits. A result of 64 ignores the scale; 64,000 adds an extra factor of ten.

Question 2: percentage change

What was the percent increase in technology-help visits from 2023 to 2025?

1. 25%
2. 37.5%
3. 62.5%
4. 162.5%

Answer: C. The displayed count rises from 8 to 13, an increase of 5. Divide by the original 8: 5/8 = 0.625, or 62.5%. The hundreds unit cancels because both values share it. 13/8 gives the final value as a percentage of the starting value, 162.5%, not the percent increase.

Question 3: compare changes

Between 2023 and 2025, which program category had the greatest percentage increase?

1. Children's programs
2. Adult programs
3. Technology help
4. The increases were equal

Answer: C. Children's programs rose by 6 from 18, so the increase is 6/18 = 33 1/3%. Adult programs rose by 3 from 24 to 27, so the increase is 3/24 = 12.5%. Technology help rose by 5 from 8 to 13, or 62.5%. The largest absolute increase is not always the largest percentage increase; each change uses its own starting value.

Totals, averages, and weighted averages

To compute an average, add the values for the relevant observations and divide by the number of observations. Confirm which observations the question includes. If the question asks for the average of three years, do not include a fourth year shown in the table. If a category is missing for one year, do not silently treat the blank as zero unless the display says zero.

A weighted average is needed when groups have different sizes. Suppose a program has a 70% success rate among 10 participants and an 80% rate among 90 participants. The combined rate is not 75%, the simple average of 70% and 80%. It is (7 + 72) successful participants divided by 100 total, or 79%. The group sizes determine the weights.

When data give both totals and percentages, use whichever representation makes the requested comparison easiest. If two departments have 30% and 40% completion rates but very different enrollment, compare rates for performance and calculate actual completions when asked for counts. A higher rate does not necessarily mean more completed cases.

Rates versus totals

A total is an amount accumulated over a period. A rate is an amount per unit, such as sales per employee or incidents per 1,000 users. Do not compare a total to a rate without a common basis. A department with more incidents may simply have more employees. If the chart supplies population, compute the rate requested rather than concluding from the raw count.

Likewise, a percentage share does not show absolute size. A small category can have a large share of a small total and still contribute fewer units than a modest share of a much larger total. Multiply the share by the total when the question asks for the underlying amount.

Reading line and bar charts

A line chart emphasizes change over ordered intervals. Compare the correct points and note whether the intervals are equal. A steeper line suggests a larger change per interval only when the scales and interval lengths are comparable. A bar chart emphasizes category comparisons. Read its axis values rather than estimating from visual height.

When two series share a graph, confirm which line or color corresponds to each group. If one series uses a secondary axis, its vertical position may not be directly comparable to the other series. Ask whether the question compares values, growth rates, or trends. Two series can finish at similar levels after very different changes.

Percent change and successive changes

Successive percentage changes apply to changing bases. A 20% increase followed by a 20% decrease does not return to the starting amount. Starting at 100, the increase produces 120. A 20% decrease from 120 removes 24, leaving 96. For a chart spanning several years, calculate each change from the value at the beginning of that interval, not from the original year unless asked.

For approximate questions, estimate before calculating exactly. If the choices are 10%, 50%, 100%, and 300%, rough division may be enough. Estimation can also catch decimal errors. A change of 5 from an original 8 must exceed 50%, so an answer near 6% is implausible.

Common traps

• Using the final value rather than the original value as the denominator for percent increase.
• Forgetting that a table is measured in hundreds, thousands, or millions.
• Comparing raw counts when the question asks for rates per person or per unit.
• Treating a blank entry as zero without a note that says so.
• Averaging percentages without weighting groups by their sizes.
• Reading a truncated graph axis as if it began at zero.
• Reporting the change as a percentage when the question asks for percentage points.

Percentage points measure the arithmetic difference between two percentages. A rate rising from 20% to 25% rises by 5 percentage points, or by 25% relative to its original rate. The question's wording tells you which measure to report.

A second worked chart

A company reports that online orders rose from 1,200 to 1,500 while total orders rose from 4,000 to 5,000. Online orders increased by 300, or 25% of 1,200. Total orders increased by 1,000, also 25% of 4,000. Online share remained 30% in both years because 1,200/4,000 = 0.30 and 1,500/5,000 = 0.30. The count rose while the share did not.

If the question asked how much of the total increase came from online orders, divide the online increase of 300 by the overall increase of 1,000. The answer is 30%. That is different from online's 30% share of total orders, even though the numerical result happens to match in this example. Always build the requested ratio from the nouns in the question.

A repeatable DI checklist

1. Read the title, units, scale, dates, and notes.
2. Circle or mentally identify the groups named in the question.
3. Write the requested measure: count, difference, ratio, average, rate, or percent change.
4. Use the denominator that matches the wording and carry units through the calculation.
5. Estimate whether the result is plausible and compare it with the answer choices.

The current GRE includes quantitative items in two sections with 27 questions total, but it does not promise a fixed number of DI items. Data Interpretation can appear in several visual formats and may be combined with other quantitative reasoning. Practice reading and calculating from the information presented rather than trying to predict a specific chart count.

How to practice

Work untimed at first and write units on every intermediate value. Then practice under section timing. For each error, label it as display-reading, setup, arithmetic, or interpretation. A display-reading error needs slower attention to labels. A setup error needs translation practice. An arithmetic error may call for estimation. An interpretation error means you calculated a quantity other than the one asked.

ETS offers free preparation resources, including a math review. Use them to revisit concepts, then practice with fresh tables and graphs. The goal is not to memorize a particular chart; it is to use titles, units, and comparisons accurately when the data change.