GMAT (Graduate Management Admission Test) Data Insights: Data Analysis & Integrated Reasoning Flashcards

24 sample cards from the Data Insights: Data Analysis & Integrated Reasoning deck

Sampled from the end of the deck, so these are different cards from the ones shown on the syllabus page.

1. What is the format of a Two-Part Analysis question and its answer grid?

   A prompt poses a problem with two related components. The answer is a table with two columns (one per component) and a shared list of options in rows; you select exactly one option for each column.

2. For a quantitative Two-Part Analysis problem, what is a reliable solving strategy?

   Translate the conditions into equations/expressions, then test the answer options against the constraints for each column. Because options are shared, you can plug candidate values in to satisfy both parts simultaneously.

3. For a verbal/logical Two-Part Analysis problem (e.g., identify a premise and a conclusion), how do you approach it?

   Determine the logical role each column asks for (e.g., supports vs. weakens, cause vs. effect), then evaluate each option against that role independently, selecting the best fit for each column.

4. In Two-Part Analysis, what does it mean that the two answer columns can be interdependent?

   The choices may be linked—for example, the two selections must be consistent with each other (such as a quantity and its complement, or two values whose sum is fixed). You must pick a pair that jointly satisfies all conditions.

5. Can the same answer option be selected for both columns in a Two-Part Analysis question?

   Sometimes, but only if the problem permits it. Often the two columns require different selections; read the prompt to determine whether repeating an option is logically valid.

6. What defines a tradeoff/optimization scenario in Data Insights, and what is the goal?

   A scenario presents options with competing attributes (e.g., cost vs. speed) and constraints. The goal is to find the option that best maximizes or minimizes a target while satisfying all constraints—the optimal choice, not just a feasible one.

7. In an optimization problem with a budget or capacity constraint, what is the standard solving method?

   List each option's relevant metrics, eliminate those that violate the constraint, then among the feasible options pick the one that best optimizes the objective (lowest cost, highest value, best ratio).

8. What operations does the GMAT on-screen calculator support, and what is its main limitation?

   It is a basic four-function calculator (add, subtract, multiply, divide, plus square root, percent, and memory). It does not handle complex expressions, so order of operations must be managed manually by the user.

9. When is it better to estimate rather than use the on-screen calculator?

   When the answer choices are far apart or the question only needs a rough magnitude or comparison. Estimation is faster; reserve the calculator for precise values or when small differences between options matter.

10. What is the formula relating rate, quantity, and time, and how is it applied in data problems?

    Quantity = rate × time (so rate = quantity/time). For example, units produced = production rate × hours; use it to convert between totals and per-unit rates in tables.

11. How do you compute a unit rate from table data, and why is it useful for comparisons?

    Unit rate = total amount / number of units (e.g., price per item = total price / quantity). It normalizes different rows to a common basis so you can compare options that have different totals fairly.

12. How do you correctly combine two successive percentage changes (e.g., +20% then −10%)?

    Multiply the factors, don't add: 1.20 × 0.90 = 1.08, a net +8% change. Successive percents compound on the running value, so adding the percents gives the wrong answer.

13. What is the difference between a percentage and a percentage point in data interpretation?

    A percentage point is the arithmetic difference between two percentages (e.g., 40% to 50% is a 10-point rise), while percent change measures relative change (40% to 50% is a 25% increase). Distinguish them to avoid misreading claims.

14. State the key principle distinguishing correlation from causation in data analysis.

    Correlation means two variables move together; causation means one variable produces a change in the other. Correlation alone does not prove causation—a relationship in the data can arise from coincidence, reverse causation, or a third (confounding) variable.

15. What is a confounding (lurking) variable, and why does it matter for causal claims from data?

    A confounding variable is an unmeasured third factor that influences both variables of interest, creating an apparent correlation between them. It matters because it can fully explain a correlation without any direct causal link, so causal conclusions from observational data are unsafe.

16. What is reverse causation, and how does it threaten a causal interpretation of correlated data?

    Reverse causation is when the assumed effect is actually the cause (Y causes X rather than X causing Y). It threatens conclusions because the same correlation is consistent with the causal arrow pointing the opposite way.

17. What is a sanity check (reasonableness check) on a computed result, and give an example?

    It is verifying that an answer is plausible in magnitude/direction before committing. Example: if a part exceeds the whole, a percent exceeds 100% impossibly, or a sum of slice values overshoots the pie total, the result is wrong and should be recomputed.

18. How can rounding help you estimate quickly while keeping a result trustworthy?

    Round inputs to convenient values (e.g., 198 → 200, 4.9% → 5%) to compute mentally, then note the direction of rounding so you know whether your estimate is slightly high or low relative to the exact answer.

19. When a table question asks how many records meet two conditions simultaneously, what is the efficient method?

    Sort by the more restrictive condition first to group qualifying rows, then within that group count the rows that also satisfy the second condition. This avoids scanning the entire table for both criteria at once.

20. What is the relationship between mean and median, and what does it reveal about a data set's skew?

    If mean > median, the data is right-skewed (high outliers pull the mean up); if mean < median, it is left-skewed; if they are roughly equal, the distribution is roughly symmetric. The median resists outliers more than the mean.

21. On a graph with two different y-axes (dual-axis chart), what must you be careful about when reading values?

    Each plotted series corresponds to a specific axis (often left vs. right). Match each line/bar to its own axis and scale before reading values, since the two axes can have different units and ranges.

22. How do you read a stacked (segmented) bar chart to find a single segment's value?

    A segment's value is the difference between its top and bottom boundary on the axis, not the top boundary alone. Subtract the cumulative height below the segment from the cumulative height at its top.

23. What is the difference between a frequency (count) and a relative frequency (proportion) when reading data displays?

    A frequency is the raw count of items in a category; a relative frequency is that count divided by the total, expressed as a fraction or percent. Relative frequencies across all categories sum to 1 (100%).

24. For a weighted average across groups in a table, what formula must you use instead of a simple average?

    Weighted average = Σ(value × weight) / Σ(weights). You cannot simply average group averages unless the groups are equal in size; each group's average must be weighted by its count.

What this deck covers

Data Insights: Data Analysis & Integrated Reasoning flashcards FAQ