Data Interpretation (DI) is the quiet workhorse of every judiciary prelims aptitude section and the CLAT-PG quantitative paper. It rarely tests a new formula; instead it tests whether you can read a table, a pie chart or a caselet under time pressure, pull out exactly the figures you need, and convert them into a percentage, ratio or average without arithmetic panic. A candidate who is fluent in percentages, ratio and proportion but slow at DI will still lose easy marks, because DI bundles four or five sub-questions onto one data set and rewards calculation discipline over cleverness. This chapter builds DI from the ground up: the data formats you will meet, the read-before-you-calculate habit, approximation and elimination techniques, and the trap questions examiners reuse year after year. Treat every dataset the way a judge treats a paper-book exhibit—orient first, locate the relevant entry, and only then compute.
What Data Interpretation Actually Tests
Data Interpretation is not a topic in the way time and work is a topic. It is a skill layer sitting on top of the arithmetic you already know. A DI set gives you raw data—a table of district-wise case filings, a pie chart of a law firm's revenue, a bar graph of bar-council enrolments across five years—and then asks four to five questions that each require you to (i) locate the correct cell or slice, (ii) decide which arithmetic operation answers the question, and (iii) execute that operation quickly and accurately.
The examiner is testing three distinct competencies at once. The first is comprehension: can you read the legend, the units, and the footnotes correctly? Many wrong answers come not from bad arithmetic but from reading "in thousands" as units, or missing that one column is a percentage while another is an absolute number. The second is selection: a DI set deliberately surrounds the relevant figure with distractors, so the skill is filtering. The third is computation speed—the underlying maths is usually just percentage, ratio, average or growth, all of which you should already command from arithmetic foundations.
Because the data is shared across a question cluster, DI is the highest return-on-investment region of the aptitude paper. One careful orientation pays off across five questions. Conversely, one misread of the table caption can poison the entire cluster. This is why disciplined DI practice—always reading the full data frame before touching question one—is the single most valuable habit in the section.
The Five Data Formats You Must Recognise
Almost every DI set in Indian judiciary and CLAT-PG papers falls into one of five formats. Recognising the format on sight tells you in advance what kind of arithmetic is coming.
1. Tables. The purest format: rows and columns of raw numbers. Tables carry the most information and the least visual help, so they reward methodical reading. Watch the row/column headers and any unit note. Tables are favoured when the examiner wants multi-step calculation—totals, averages across rows, ratios between cells.
2. Bar graphs. Bars encode magnitude by height (or length). Simple bar graphs compare one variable across categories; grouped (clustered) bars compare two or more series side by side; stacked bars show components summing to a whole. The trap in stacked bars is reading a segment's top edge as its value when you must instead subtract the segment below it.
3. Line graphs. Best for trend over time. Slope is the story—steepness signals rate of change, and an intersection of two lines is a favourite question ("in which year were filings equal?"). Read intersections and turning points carefully.
4. Pie charts. A circle of 360° representing 100%, sliced into shares. Pie questions blur the line with percentage and ratio work: a slice may be given in degrees, in percent, or as an absolute value, and you must convert. Remember 360° = 100%, so 1% = 3.6° and 90° = 25%.
5. Caselets. Pure prose with the data buried in sentences—no chart at all. The skill is to extract the numbers into your own mini-table on the rough sheet before answering. Caselets are the most time-consuming format and the one where rough-work discipline matters most.
Mixed or combination sets (a table plus a pie chart referring to the same population) are increasingly common; they simply layer two formats and test whether you can carry a figure from one frame to the other.
The Read-Before-You-Calculate Discipline
The most common DI error is calculating before fully reading the data frame. Before you look at a single question, spend twenty to thirty seconds reading the entire set the way a litigator reads a document before cross-examination. Check four things every time.
The title. What does the dataset describe—cases filed, cases disposed, pendency, revenue? "Filed" and "disposed" are different populations, and a question asking about pendency may require filed minus disposed, not a figure printed anywhere.
The units. Are the numbers in absolutes, hundreds, thousands, lakhs, crores, or percentages? A figure of "45" under a heading "(in thousands)" is 45,000. Mixing unit scales across columns is a deliberate trap.
The legend. In grouped or multi-line charts, which colour or pattern maps to which series? Swapping two series is the fastest way to get every question in the cluster wrong.
The footnotes. Small-print notes such as "figures for 2024 are provisional" or "the Others category includes appeals" frequently hold the key to one tricky question. Examiners reward the candidate who reads the fine print.
Only after this orientation should you read question one. This sequencing mirrors sound advocacy: you never argue from an exhibit you have not first read in full. The thirty seconds invested in orientation are recovered many times over because you stop re-reading the frame for every sub-question.
Working With Tables
Tables are the backbone format. Suppose a table gives, for five High Court registries, the number of writ petitions filed and disposed in a year. A typical cluster will ask for: the registry with the highest disposal rate; the combined pendency of two registries; the ratio of filings between two registries; and the average disposal across all five.
Disposal rate is disposed divided by filed, expressed as a percentage—a direct percentage operation. When asked for the highest rate, you often need not compute every value precisely; estimate each ratio and compare, computing exactly only the two closest contenders.
Pendency is filed minus disposed for the period (ignoring opening backlog unless the table supplies it). The trap is a question that smuggles in "cumulative pendency" when the table only gives one year—if the data is not there, the answer is "cannot be determined," and that option is sometimes correct.
Ratios between two cells should be reduced to lowest terms; the answer options are almost always in lowest terms, so reducing aligns you with the key and exposes arithmetic slips. Averages across a row or column are the total divided by the count—add carefully, because a single addition error propagates. For tables with many rows, add in pairs and keep a running subtotal on the rough sheet rather than trying to total a long column in one pass.
Bar Graphs: Simple, Grouped and Stacked
Bar graphs trade precision for visual speed. With a simple bar graph—say, advocates enrolled per year across five years—reading values is direct, and questions usually ask for the year of maximum/minimum, the difference between two years, or percentage growth. Percentage growth from year A to year B is (B − A) / A × 100; never divide by B. This is the same growth logic that underlies compound interest, where each period grows on the previous base.
Grouped bar graphs place two or more bars per category—for instance, filed and disposed side by side for each year. The legend is critical here. A common question asks for the year with the largest gap between the two series, which you answer by scanning the visual gap, not by computing every pair. Only compute the top two candidates.
Stacked bar graphs are where most candidates lose marks. Each bar is a column of segments summing to a total. The value of an upper segment is the difference between its top edge and the edge below it—not the height of its top edge read off the axis. If a stacked bar shows civil cases from 0 to 30 and criminal cases stacked from 30 to 50, the criminal value is 20, not 50. Stacked charts often also give the total directly (the top of the stack), which is convenient for percentage-of-total questions: the share of criminal cases is 20/50 = 40%.
Line Graphs and Trend Reading
Line graphs are built for time series—pendency over a decade, monthly disposals, year-on-year enrolment. The reader's eye should follow slope. A steep upward segment is rapid growth; a flat segment is stagnation; a downward segment is decline. The examiner exploits this by asking which interval saw the greatest rate of change—the steepest segment—or by asking for the year of a peak or trough, which is just the highest or lowest point.
The signature line-graph question involves two intersecting lines: "in which year were cases filed equal to cases disposed?" The answer is the intersection point—no calculation needed, just careful reading. A related trap asks for the interval where one series first overtook another; that is the year just after the crossing, so read the direction of the lines.
Percentage-change questions on line graphs use the same (new − old)/old × 100 formula as bar graphs. When a question asks for average over the period, total the plotted points and divide by their count, exactly as with a table. The visual nature of line graphs tempts candidates to eyeball values that fall between gridlines; when a question's options are close, read the value conservatively and, if two options straddle your reading, recompute using the nearest gridline as an anchor.
Pie Charts and the Degrees-Percent Conversion
A pie chart shows how a single whole divides into parts. The whole is 360° and equals 100% of the total quantity, so the two conversions you must do reflexively are 1% = 3.6° and degrees ÷ 3.6 = percent. Anchor values worth memorising: 90° = 25%, 180° = 50%, 36° = 10%, 18° = 5%.
Pie questions arrive in three flavours. When slices are given in percent and you are told the total absolute value, each slice's absolute value is its percent times the total—straightforward percentage work. When slices are given in degrees, convert to percent first (divide by 3.6), then to absolute. When you are given one slice's absolute value, you can back-solve the total and then any other slice.
The classic two-pie question shows, say, a firm's revenue by practice area in two different years with different totals. The trap is comparing percentages directly: a practice area can be a larger percentage in year one yet a smaller absolute figure if the year-two total is much bigger. Always convert to absolute values before comparing magnitudes across two pies with different totals. Reduce ratio answers to lowest terms to match the options, the same discipline used throughout ratio and proportion problems.
Caselets: Turning Prose Into a Table
A caselet hides the data inside sentences. "In a District Court complex, 40% of the 1,200 advocates practise on the civil side; of the rest, three-fifths handle criminal matters and the remainder appear in family court." There is no chart; you must build one. The technique is to read the prose once, then on your rough sheet draw the table or tree the data implies before attempting any question.
For the example: total = 1,200; civil = 40% = 480; rest = 720; criminal = three-fifths of 720 = 432; family = 720 − 432 = 288. Now any sub-question—ratio of criminal to family (432:288 = 3:2), percentage in family court (288/1200 = 24%)—is a quick lookup against your extracted table rather than a re-reading of the paragraph.
Caselets reward fraction fluency because the data often arrives as "two-thirds," "one-fifth," "the remainder." Convert each to a number against the running base immediately, and track what "the rest" refers to—it is the trickiest pronoun in DI, because "three-fifths of the rest" means three-fifths of 720, not of 1,200. The same proportional reasoning powers ratio and proportion and partnership problems, so a candidate strong there will find caselets the easiest DI format despite their forbidding prose.
Approximation and Estimation
Speed in DI comes less from fast multiplication than from knowing when not to compute exactly. Many DI questions, especially comparison and "approximately equal to" questions, can be answered by estimation. If a question asks which of five registries had the highest disposal rate, you do not need five exact percentages; you need to identify the leader, which a rounded estimate usually settles, with exact computation reserved for the two closest.
The core technique is rounding to convenient numbers: 4,870 becomes roughly 4,900 or even 5,000 depending on how close the options are. The art is calibrating roundness to the spread of the options—if options differ by 10%, round freely; if they differ by 1%, round tightly or compute exactly. Fraction approximation is the other key tool: 1/3 ≈ 33.3%, 1/6 ≈ 16.7%, 1/8 = 12.5%, 1/9 ≈ 11.1%, 1/11 ≈ 9.1%, 1/12 ≈ 8.3%. Memorising these unit-fraction percentages converts "what percent is 3,000 of 36,000" into an instant "one-twelfth, about 8.3%" rather than a long division.
For percentage growth, the approximate-then-adjust method works well: to grow 4,200 by 15%, take 10% (420) plus half of that (210) = 630, so the new value is about 4,830. These mental scaffolds, drilled until automatic, are what separate a candidate who finishes the DI section from one who runs out of time on the last cluster.
Common Traps and How Examiners Set Them
DI traps are remarkably consistent across exams. Knowing the catalogue lets you defuse them on sight.
Percentage versus percentage point. A rise from 20% to 25% is a five percentage-point rise but a 25% relative increase. Questions exploit the confusion between the two; read whether the question asks "by how many percentage points" or "by what percent."
Unit mismatch. One column in thousands, another in absolutes; one pie in degrees, another in percent. Always normalise units before combining figures.
Comparing across different totals. The two-pie trap: a larger share need not be a larger absolute when the bases differ. Convert to absolutes before comparing.
"Cannot be determined." Some questions genuinely lack the data—asking for opening backlog when only one year's filings and disposals are given. Resist the urge to invent; if the figure is not derivable, that option is the answer.
Average of percentages. You cannot average percentages computed on different bases and treat the result as the overall percentage; the correct overall percentage uses the combined totals. This mirrors the weighted-average logic that appears in ratio and alligation problems.
Cumulative versus incremental. A line graph of "total cases registered to date" is cumulative; the cases added in a year is the difference between consecutive points, not the point itself.
Multi-Step and Linked Questions
The hardest DI sub-questions chain operations: "By what percentage did the disposal rate of Registry A exceed the average disposal rate of all five registries?" This is three steps—compute A's disposal rate, compute the average disposal rate across all five, then compute the percentage by which the first exceeds the second. Each step uses ordinary arithmetic, but the sequencing is the test.
The discipline for multi-step questions is to write each intermediate result on the rough sheet and label it, so that a later sub-question reusing the same intermediate (the average disposal rate, say) need not be recomputed. DI clusters are designed so that work done for question two often feeds question four; the efficient candidate banks intermediate results rather than starting each sub-question cold.
Linked questions also test whether you carry units correctly through a chain. If you converted a pie slice from degrees to percent to absolute for an early question, keep that absolute value labelled; a later "what is the ratio of slice X to slice Y" question reuses it. Treat your rough sheet as a working ledger, the way a judgment records findings of fact once and then relies on them—not re-litigating each figure for every issue.
Time Strategy in the Exam Hall
DI rewards triage. Within the aptitude section, scan the DI sets and grade them: a clean table or simple bar set is high-yield and should be done first; a dense caselet with fractions of fractions, or a combination set spanning a table and two pies, is time-expensive and best deferred. Securing the easy clusters first banks marks before the clock pressure peaks.
Within a set, attempt sub-questions in order of effort, not order of printing. A "which year had the maximum" question is usually a free mark—take it before a three-step percentage-excess question on the same data. Because the orientation cost is already paid once you have read the frame, every additional sub-question you solve on that frame is pure efficiency.
Set a personal ceiling—if a single sub-question has consumed more than about ninety seconds and you are not converging, mark it for review and move on; the opportunity cost of one stubborn DI question is two easy ones elsewhere. This budgeting mindset is identical to the pacing logic in time, distance and speed problems, where chasing one hard scenario can sink an entire section. Finish the paper, then return to flagged questions with whatever time remains.
Why DI Matters for Judiciary and CLAT-PG
Several state judicial service preliminary examinations and the CLAT-PG paper carry a general-aptitude or quantitative component in which DI is a reliable scorer. The data sets in these papers frequently use a legal flavour—case filings, disposal rates, pendency across courts, bar enrolments, legal-aid statistics—but the underlying arithmetic is identical to any DI question. The legal dressing is cosmetic; do not be intimidated by court-themed data.
The deeper reason DI belongs in a judicial aptitude paper is that adjudication itself is data interpretation. A judge reading a charge-sheet, a balance sheet in a winding-up petition, or pendency statistics in a docket-management review is doing precisely what a DI question asks: orienting to a data frame, locating the relevant figures, and drawing a defensible inference. The skills drilled here—reading the fine print, distinguishing absolute from relative change, refusing to assert what the data does not support—are the same evidentiary instincts that good judging demands.
To consolidate, pair this chapter with the broader arithmetic toolkit: revisit percentage, ratio and proportion for the engine of most DI computation, arithmetic foundations for calculation speed, and return to the full aptitude and reasoning hub to map DI against the rest of the syllabus. Mastery here is mostly a function of disciplined practice: do a few sets daily, time them, and review every error to classify it as a comprehension, selection or computation mistake.
Frequently asked questions
What is the single most common mistake candidates make in Data Interpretation?
Calculating before fully reading the data frame. Most wrong DI answers come not from arithmetic errors but from misreading the units, the legend or a footnote—treating "in thousands" as units, swapping two series in a grouped chart, or comparing percentages across pies with different totals. Spend twenty to thirty seconds orienting to the title, units, legend and footnotes before touching question one.
How do I convert between degrees and percentages in a pie chart?
The whole pie is 360° and equals 100% of the total, so 1% = 3.6° and degrees ÷ 3.6 = percent. Memorise anchors: 90° = 25%, 180° = 50%, 36° = 10%, 18° = 5%. If a slice is given in degrees and you need an absolute value, convert the degrees to a percent first, then apply that percent to the stated total.
When should I estimate rather than compute exactly in DI?
Estimate whenever the question asks for a comparison, an "approximately equal" value, or the maximum/minimum among several options that are well spread apart. Round to convenient numbers, identify the leading candidate, and compute exactly only for the two closest contenders. Compute precisely only when the answer options differ by a small margin (around 1%), where rounding could push you to the wrong option.
What is the difference between a percentage increase and a percentage-point increase?
A move from 20% to 25% is a five percentage-point increase but a 25% relative increase, because the rise of 5 is one-quarter of the original 20. DI questions deliberately exploit this confusion. Read whether the question asks "by how many percentage points" (subtract) or "by what percent" (divide the change by the original base, then multiply by 100).
Why can't I just average two percentages to get the overall percentage?
Because percentages computed on different bases cannot be simply averaged. If 80% of 100 cases and 20% of 900 cases were disposed, the overall disposal rate is not (80+20)/2 = 50%; it is total disposed divided by total cases = (80+180)/1000 = 26%. The correct overall figure uses combined totals—the same weighted-average logic found in ratio, proportion and alligation problems.
How is Data Interpretation relevant to a judicial career, not just the exam?
Adjudication is itself a form of data interpretation. A judge reading pendency statistics in docket management, a balance sheet in a company petition, or figures in a charge-sheet performs exactly what a DI question demands—orient to a data frame, locate the relevant figures, and draw a defensible inference without asserting what the data does not support. The exam skill mirrors the evidentiary discipline the bench requires.