JIPMAT common mistakes
JIPMAT · Understand the traps and how to avoid them.
Reading a ratio question (say A:B = 3:5, B:C = 4:7) and immediately cross-multiplying to get A:C, producing 12:35 instead of linking through the LCM of the B terms to get 12:20:35.
Candidates treat two separate ratios as a single proportional chain and skip the step of making the shared middle term equal. The arithmetic looks plausible, so the error survives a quick re-check.
What to do instead
Before computing, write the shared quantity in both ratios, take its LCM, scale both ratios, then merge into a three-part ratio. Practise five such chains until the LCM step is automatic, and always verify by substituting a concrete number for the middle term.
Solving time-speed-distance or pipes-and-cistern problems by picking a single unit rate and multiplying, without checking whether the rates combine additively or whether one entity works against another.
These problems are memorised as formulas rather than modelled as rates. Under exam pressure the candidate reaches for the familiar 'product over sum' template even when the setup is a leak, a relative-speed case, or a two-stage journey.
What to do instead
Start every such problem by writing the rate of each actor with explicit units (e.g. 1/12 tank per hour) and a sign for direction. Build the combined rate as a signed sum, then invert once at the end. Do ten mixed problems where some actors subtract.
Losing marks on simplification and approximation items by expanding large products fully instead of using algebraic identities, then making a sign slip in the expansion.
The instinct is to grind out arithmetic because it feels safe, but long multi-step expansions create many opportunities for a single sign or carry error, and they consume time needed elsewhere.
What to do instead
Drill the standard identities (difference of squares, (a±b)^2, sum and difference of cubes) and scan each expression for a matching pattern before computing. When a numeric approximation is offered, round to two significant figures and compare to the options rather than computing exactly.
Leaving probability and permutation/combination answers as an unreduced fraction or raw count that does not match any option, then guessing rather than re-examining whether order mattered.
The decision between nPr and nCr is made from keyword memory ('arrangement' versus 'selection') rather than from asking whether swapping two chosen items produces a different outcome. When the option set is in reduced-fraction form, the mismatch is blamed on arithmetic.
What to do instead
Before writing any formula, state in one line whether position matters and whether repetition is allowed. Then compute, reduce the fraction, and check the answer lies between 0 and 1. Keep a written decision checklist for the four standard cases.
In data interpretation, computing every value from the table or chart before reading the question, then running out of time with three of five questions unanswered.
Candidates fear missing data and default to full computation. DI sets are designed so that most questions can be answered by comparison, estimation, or reading a single row, so exhaustive calculation is wasted effort.
What to do instead
Read all questions for a set first, then compute only the cells each question needs. For percentage-change and ratio questions, compare using rough magnitudes and eliminate options before doing exact division. Timebox each set to about four minutes.
Misreading units or base years in DI - comparing a figure in thousands with one in lakhs, or computing growth from the wrong base year - and selecting an option that is off by a factor of ten or a year.
The chart legend and axis labels are skimmed because the numbers look familiar. Exams deliberately mix units across a table or place the base year in a footnote.
What to do instead
Before answering, annotate the table header with a single common unit and circle the base period. Sanity-check every answer against the order of magnitude of the inputs; a percentage change should rarely exceed a few hundred percent.
In logical reasoning seating and ranking puzzles, drawing a fresh diagram for each clue instead of maintaining one evolving arrangement, then contradicting an earlier clue.
Candidates diagram reactively, one clue at a time, and lose track of which possibilities have already been eliminated. Under time pressure they commit to the first arrangement that fits most clues.
What to do instead
Set up one master grid or linear layout, write each clue as a constraint beside it, and mark eliminated positions rather than erasing. For puzzles with more than six entities, test the two or three most constrained positions first and branch only when forced.
In verbal ability, choosing an option because it repeats a word from the passage rather than because it restates the idea, especially on inference and 'author's tone' questions.
Keyword matching feels efficient and is reinforced by surface similarity. Inference items are built precisely to punish lexical overlap and reward paraphrase.
What to do instead
For each option, locate the sentence that would have to be true for it to hold, then check whether the passage actually asserts it. Reject options that merely repeat vocabulary, and prefer the choice that is supported but not directly quoted.
Answering para-jumble and sentence-completion items by 'what sounds right' without identifying the opening sentence or the grammatical connector that links two clauses.
Spoken-language intuition is unreliable for formal written English, and jumbles hinge on structural signals - pronouns, definite articles, and discourse markers - that are easy to overlook.
What to do instead
For jumbles, find the sentence with no unresolved pronoun or definite reference; that is usually the opener. Then chain via connectors (however, therefore, this) and verify the full order reads coherently. For completions, predict the missing word's function before scanning options.
Spending uniform time across all three sections and all questions, so that a single hard puzzle in reasoning consumes the minutes that three easy verbal questions would have earned.
There is no per-question marking scheme in the candidate's head, so effort is allocated by order of appearance rather than by expected return. Hard questions are also more engaging, which prolongs the attempt.
What to do instead
Adopt a two-pass strategy: on the first pass answer only questions solvable within about 60 seconds, flag the rest, and return with remaining time. Set a hard cap per question in each section and practise with a visible timer until the cap is habitual.