UPSSSC ASO common mistakes
UPSSSC ASO · Understand the traps and how to avoid them.
Treating लिंग/वचन and case markers as optional while answering grammar items: choosing the unmarked masculine singular form in fill-in-the-blank sentences even when the sentence's verb or postposition forces feminine or plural agreement.
Candidates memorise vocabulary lists but never drill agreement paradigms, and Hindi grammar in school is often taught as definitions rather than as sentence-level transformation practice. Under time pressure they match the meaning of the word rather than its grammatical fit.
What to do instead
Build a one-page agreement table (masculine/feminine × singular/plural, plus the आ/ई/ए pattern for adjectives) and practise 20 transformation sentences daily where the verb ending is the only clue. Before marking an option, locate the agreeing verb or postposition in the sentence and check the noun against it.
Answering 'correct the sentence' and मुहावरा/लोकोक्ति items by sound familiarity rather than by literal meaning, picking an idiom that 'sounds right' but whose sense does not match the sentence context.
Idioms are memorised as isolated phrases with a one-word gloss, so candidates cannot distinguish near-neighbours such as अंधे की लाठी vs अंधा क्या चाहे दो आँखें. Also, the printed option often repeats a word from the stem, creating a false familiarity pull.
What to do instead
Maintain an idiom notebook in three columns: idiom, literal sense, one original sentence of your own. Study them in contrasting pairs (similar meaning vs opposite meaning) and revise weekly by covering the meaning column and recalling it. In the exam, insert each candidate idiom into the sentence and check whether the resulting sentence makes logical sense.
Losing marks on spelling and शब्द-शुद्धि questions by choosing the phonetic spelling (e.g., writing परिक्षा for परीक्षा) or by ignoring the distinction between similar-looking pairs such as अंश/अंस and कृतज्ञ/कृतघ्न.
Spoken Hindi neutralises short/long vowel distinctions and many conjunct consonants, so candidates write what they hear. Schooling rarely includes systematic dictation, and printed options are deliberately close to the common misspelling.
What to do instead
Keep a 100-word शब्द-शुद्धि list of the errors you personally make, updated from every mock test. Do a five-minute dictation of 15 words from that list each day, writing them aloud syllable by syllable. For similar pairs, learn the meaning of each member, since meaning is the only reliable discriminator.
Applying a memorised grammar rule to the wrong target in error-spotting questions, e.g., marking a part as wrong because it contains a plural noun when the actual error is a subject-verb agreement or article error elsewhere in the sentence.
Candidates scan for known error categories instead of reading the whole sentence for meaning and structure. Because one part is 'flagged' early, confirmation bias makes them stop searching, and they never check the remaining parts.
What to do instead
Read the full sentence once for meaning before looking at the labelled parts. Then check in a fixed order: verb agreement, tense consistency, article/determiner, preposition collocation, pronoun reference, parallelism. Only mark a part after the whole checklist has been applied, and practise with questions where the error is deliberately placed in the last part.
Choosing an English word by Hindi-to-English translation rather than by collocation, e.g., selecting 'do a mistake', 'give an exam', or 'discuss about the matter' in fill-in-the-blank and sentence-improvement items.
Candidates translate mentally from Hindi, where the verb is सही but its English equivalent takes a different collocate. Vocabulary is learned as word-meaning pairs, not as phrases, so collocational restrictions are invisible.
What to do instead
Learn vocabulary in chunks of three to five words (make a mistake, take an exam, discuss the matter, cope with, consist of). Keep a collocation column in your word list and add only phrases, never bare words. In the exam, test each option by silently completing the phrase and rejecting options that need an extra preposition.
Spending disproportionate time on one reading-comprehension passage and then answering its inference questions from general knowledge instead of from the passage text.
The passage looks familiar in topic, so candidates answer from prior belief. Time lost on the first passage also forces rushed guessing on later questions, and inference items feel ambiguous when the passage has not been read closely.
What to do instead
Adopt a fixed budget: skim the questions first, read the passage once at moderate speed, then answer. For every inference or tone question, underline the exact line in the passage that supports the choice and reject any option that is true in the world but not stated. Practise with a timer and abandon any single question after 45 seconds.
Answering static-GK questions from outdated memorised facts (former capital names, abolished schemes, old appointment holders) and current-affairs questions from headlines without checking the date of the event.
Coaching material and older notes circulate for years, and candidates revise a single compiled PDF without tracking when each fact was last true. Current affairs are read as news, not as exam-relevant facts with dates, places and offices attached.
What to do instead
Date every fact in your notes (e.g., 'appointed March 2024') and mark superseded entries as struck through rather than deleting them. For current affairs, maintain a monthly one-page sheet covering appointments, schemes, sports, awards and indices with the month written at the top. Before answering, check whether the question asks for the current or a past holder.
Confusing similarly named Indian institutions and constitutional provisions, e.g., mixing up the roles of the Finance Commission and the NITI Aayog, or Article 32 with Article 226.
Static polity is learned as a list of names and numbers without the underlying function, so items that differ only in mandate or issuing authority collapse into one memory.
What to do instead
For each institution or article, write a one-line 'who, what, to whom' card: who constitutes it, what it does, whom it reports to or binds. Group look-alike items side by side in a comparison table (32 vs 226, Finance Commission vs NITI Aayog, CAG vs Election Commission) and revise the table, not the individual entries.
Solving seating-arrangement and blood-relation puzzles by holding the arrangement in working memory and drawing nothing, then losing track after the third clue and guessing.
Diagrams feel slower than mental simulation for easy puzzles, so candidates skip them, and the habit fails exactly when the puzzle has four or more variables. Under time pressure, memory overload produces confident wrong answers rather than acknowledged uncertainty.
What to do instead
Always draw: a line for linear arrangements, a circle with fixed positions for circular ones, and a family tree with gender and generation markers for blood relations. Enter each clue as a symbol on the diagram and re-check every clue against the finished diagram before answering. Time yourself to prove that diagramming is faster overall.
Treating 'no conclusion follows' and 'either-or' options in syllogism and inequality questions as filler, and marking only the first conclusion that seems true.
Candidates evaluate conclusions one at a time and stop at the first true one, ignoring that the answer may require checking both conclusions and the either-or possibility. The possibility case ('some X being Y is a possibility') is also misread as a certainty claim.
What to do instead
Use the standard two-step routine: draw the Venn diagram for the premises, then test each conclusion for certainty; only if both are individually uncertain should you test whether they form a complementary pair. Learn the complementarity rules (some/no, all/some-not) explicitly and practise 10 either-or questions in a row until the pattern is automatic.
In percentage, profit-loss and discount questions, applying successive changes additively (treating a 20% increase followed by a 20% decrease as no change) instead of multiplicatively.
The base changes after each operation, but candidates anchor on the first number and add percentages as if they were absolute quantities. This is a conceptual gap, not a calculation slip, so it repeats across every successive-change question.
What to do instead
Convert every percentage change into a multiplier (increase 20% → ×1.2, decrease 20% → ×0.8) and multiply the chain. Memorise the standard result that a% up followed by a% down gives a net decrease of a²/100 per cent. Practise by verifying with a concrete base of 100 so the error becomes visible.
Leaving ratio-and-proportion and time-and-work problems unsolved or guessed because the candidate sets up equations with two or three unknowns instead of using unitary or LCM methods.
School algebra training pushes towards formal equation solving, which is slow under exam conditions, so candidates either run out of time or make substitution errors. They never learn the shortcut methods that make these questions 30-second items.
What to do instead
Learn the LCM method for time and work (assume total work = LCM of the given days) and the unitary method for ratios (find the value of one part first). Practise 15 questions of each type using only the shortcut, then compare with the equation method once to confirm the answer.
In data-interpretation and statistics items, computing a mean, median or percentage from the wrong base - dividing by the number of categories instead of total frequency, or taking a simple average of already-averaged group means.
Tables present both raw frequencies and derived columns, and candidates grab whichever numbers are visually prominent. The weighted nature of a combined mean is understood as a formula to memorise rather than as a consequence of group sizes.
What to do instead
Before any calculation, write down explicitly 'numerator = …, denominator = …' and check that the denominator is a total frequency or total count, not a count of rows. For combined-group problems, always reconstruct the total sum (mean × frequency) for each group before combining. Do a sanity check: the combined mean must lie between the two group means.
Mixing up which measure of dispersion or central tendency is appropriate - reporting standard deviation when the question asks for variance, or choosing mean for heavily skewed data where the median is intended.
Formulas are memorised in isolation without the property that distinguishes each measure, so the names become interchangeable labels. The variance/standard-deviation distinction is especially lost because one is the square of the other.
What to do instead
Build a property table for each statistic: units (original units vs squared units), sensitivity to outliers, suitability for open-ended classes, and additivity. Attach a one-line cue to each (variance = squared units; median = resistant to outliers; mode = categorical data). Revise the table before every mock and re-read the question's wording for 'variance', 'standard deviation' and 'coefficient of variation'.