Relvora Prep
Concept library
One short lesson per GMAT Data Insights skill: how to read the chart or table, the method for each statement type, a worked example, and the traps. Every practice question links straight to the part of the lesson it tests.
Reading the display
- Read ValuesEvery chart question starts with a reading. Fix the category on the horizontal axis, the series in the legend, and read the mark against the gridlines; in a table, fix the row by its name and the column by its header. Compare values, never bar lengths.7 sections · 5 traps · 14 min
- Filter & SortSorting answers 'what position', filtering answers 'among those that…'. Sort by clicking the column header, count positions from the end the statement names, and always strike out the rows a condition removes before you compare.7 sections · 4 traps · 14 min
- Count ConditionsCounting is a checklist, not a glance. Decide what happens at the threshold ('more than' excludes it), apply every condition to every row, and for derived conditions compute the value for each row first.7 sections · 4 traps · 14 min
Computing from data
- Percent ChangePercent change = (new − old) ÷ old: the base is the earlier value. Percentage points are a plain difference of two percents. Successive changes multiply.8 sections · 5 traps · 16 min
- RatiosA ratio keeps its order (A to B puts A first) and a rate divides one quantity by another for the same row or dot. Compute rates before comparing: totals mislead.7 sections · 4 traps · 14 min
- Median & MeanMean = sum ÷ count. Median = middle of the sorted values (halfway between the two middle ones for an even count). Removing a value below the mean raises the mean; the median moves only if the middle changes.7 sections · 3 traps · 13 min
- Weighted AverageA combined rate is total (rate × count) ÷ total count. It lands nearer the bigger group's rate. The plain average of the rates is right only when the groups are the same size.6 sections · 2 traps · 12 min
- Part to WholeA segment's value is its length (end minus start), its share is that length over the whole bar, and a change in share is a difference in percentage points.7 sections · 4 traps · 15 min
Relationships and inference
- TrendsCompare each point with the one before it, divide total change by the number of intervals (not points), and project from the last value at the average rate.8 sections · 4 traps · 17 min
- CorrelationJudge a relationship from the whole cloud of dots. Count regions with two imaginary lines. A conditional fraction's denominator is the group the condition picks out. A tendency is not a rule.7 sections · 5 traps · 15 min
- DeterminableList the numbers a quantity needs. If every one is in the table (or follows from columns that are), it can be determined. New information helps only if it supplies the one missing number.6 sections · 3 traps · 13 min
Data Sufficiency
- DS: number propertiesData Sufficiency asks whether the statements are enough to answer, not what the answer is. Judge statement (1) alone, then (2) alone, then both together. A statement is sufficient only if every case that fits it gives the same answer, and a definite 'no' counts.7 sections · 3 traps · 4 min
- DS: equations and valuesA statement is sufficient when it leaves exactly one value for what the question asks. Count the solutions each statement allows: a linear equation in x gives one, a square gives two, and one equation in two unknowns gives endless pairs, unless it fixes the very combination asked for.7 sections · 5 traps · 11 min
- DS: inequalitiesTurn every statement into the values it allows, on a number line or as signs. A statement is sufficient when all of those values answer the question the same way, and a definite 'no' counts. Before you trust one, test a negative number, 0, a fraction between 0 and 1 and the endpoint itself.7 sections · 4 traps · 15 min
- DS: word problemsTurn the story into unknowns and equations, then ask whether the facts pin down the one quantity asked. Two unknowns usually need two independent facts, but a ratio never fixes a count, a percent needs a base, and whole-number counts can make one equation enough.7 sections · 5 traps · 13 min
- DS: statisticsKnow what each measure needs before you read the statements. A mean needs the total and the count (or a fact tying them together), a median needs the middle position, a range needs both ends, and an overlap needs three of the four regions.7 sections · 5 traps · 13 min
Two-Part Analysis
- Two-Part: equationsA Two-Part Analysis question gives one list of options and two columns. Each column asks for a different value, both must be right, and the same option can answer both. Solve for each value on its own, then find it in the list.7 sections · 8 traps · 8 min
- Two-Part: word problemsA Two-Part word problem tells one story and asks for two quantities, one per column. Name both unknowns, solve for each one on its own, and check the pair against every fact and limit in the story before you look at the list.7 sections · 9 traps · 14 min
Multi-Source Reasoning
- Multi-Source: rules and dataIn Multi-Source Reasoning the information is split across tabs. One tab often states rules (prices, limits, eligibility) and another holds the data; a third tab can change a rule for one group. Read every tab first, then apply each rule to exactly the rows a statement names.7 sections · 9 traps · 9 min
- Multi-Source: combining sourcesMulti-Source Reasoning splits the facts across tabs, and every statement needs at least two of them. Read every tab first and note its keys, units and rules; then, for each statement, find the tabs it needs, translate and convert, and only then compute or compare.7 sections · 8 traps · 12 min