One mock gives you plenty of attempts and too many avoidable errors. In the next, confidence collapses and the attempt count goes with it. This six-week loop steadies the decisions behind both outcomes: copy the live rules, diagnose, repair concepts, practise selection, add time pressure, then analyse matched simulations. Every count, time and threshold in it is a training model you build for yourself, not an official pH Test pattern.
1. Copy the live rules, then run a 30-question baseline
Copy a rule card only from current eLitmus instructions or the candidate portal: live section names | item counts | time limits | navigation | scoring or penalty | permitted tools. Where the live card and any practice plan disagree, the live card wins.
Accuracy comes first because the pH Test reports a percentile, a rank against the rest of the cohort, so a wrong answer costs position and not only marks: eLitmus pH Test: The Percentile System, Negative Marking, and the Hard DI Section works that through. eLitmus pH Test Pattern: Sections, Percentile Logic and Attempt Quality names what Quantitative Aptitude, Verbal Ability and Problem Solving each measure.
Run a separate 30-question baseline, 10 per lane. Use a personal 45-minute cap: 18 quantitative, 14 problem solving, 10 verbal and 3 review. Log attempted, correct, wrong, skipped and lane time.
Lane | Attempted | Correct | Wrong | Skipped | Attempted accuracy | Coverage |
|---|---|---|---|---|---|---|
Quantitative | 9 | 6 | 3 | 1 | 6/9 x 100 = 66.7% | 9/10 x 100 = 90% |
Problem Solving | 8 | 7 | 1 | 2 | 7/8 x 100 = 87.5% | 8/10 x 100 = 80% |
Verbal | 10 | 8 | 2 | 0 | 8/10 x 100 = 80% | 10/10 x 100 = 100% |
Overall: 27 attempts, 21 correct, 21/27 x 100 = 77.8% accuracy and 27/30 x 100 = 90% coverage. Repair quantitative first. High coverage means attempts alone are insufficient.
2. Turn every loss into one cause and one scheduled repair
Tag each loss once: C missing concept, P process or calculation slip, R reading or interpretation error, and D bad decision under time. No route or an unsupported guess is D.
The baseline has 6 wrong + 3 skipped = 9 losses. Tags are C=3, P=2, R=1, D=3; 3+2+1+3=9. Split early practice between concept repair and decision control.
Item | Lane | Chosen | Correct | Tag | Proof | Scheduled repair |
|---|---|---|---|---|---|---|
Q7 | Quantitative | 52.5 km/h | 51.43 km/h | C | Averaged two speeds directly | Total distance / total time, then two changed-value variants on Day 1 and Day 3 |
Q19 | Problem Solving | Ravi in seat 3 | Meera in seat 3 | D | Guessed at 2:10 with four of eight clues placed | Stop at 90 seconds while the diagram is still incomplete, then two clue drills on Day 2 |
Q23 | Quantitative | 0.72 days | 7.2 days | P | Decimal slip dividing 36 by 5 | Recompute with the units written out, then two variants on Day 4 |
Require a primary tag, proof line and next action. A chapter name is not a diagnosis. Close only after solving the original unaided and a changed-value variant; reading the key does not count.
3. Weeks 1 and 2: repair concepts before compressing time
Weekly allocation: quantitative 2 h 30 m, problem solving 1 h 30 m, verbal 1 h, error-log review 1 h. Total: 2 h 30 m + 1 h 30 m + 1 h + 1 h = 6 h. Quantitative leads because its baseline accuracy was the lowest of the three lanes at 66.7%.
Each 45-minute block is 10 recall, 25 solving and 10 checking and logging.
For Q7, travel 180 km at 60 km/h and return 180 km at 45 km/h. Times are 180/60 = 3 h and 180/45 = 4 h; totals are 360 km and 7 h. Average speed is 360/7 = 51.43 km/h. Reject (60+45)/2 = 52.5 because times differ.
Variant: 120 km at 40 km/h and 120 km at 60 km/h take 3 h and 2 h; average 240/5 = 48 km/h.
The sequence 4, 9, 19, 39, 79 follows x2+1, so next is 79x2+1 = 159. The rule must fit every transition.
Where the gaps run wider than the error log, Aptitude Courses for Exams and Placements covers the quantitative and reasoning ground topic by topic.
4. Week 3: practise question selection with an evidence rule
After scanning, label an item A, attack now when method and first step are clear; P, park when plausible but lengthy; or L, leave for this pass when no route appears. Revisit only when live navigation rules permit it.
Apply the labels in a five-item drill:
Q1, percentage change: method clear at 0:12, A, finish 1:05, correct.
Q2, arrangement: diagram clear at 0:18, P, revisit, finish 2:40, correct.
Q3, unfamiliar series: no rule fits three transitions by 0:35, L.
Q4, reading inference: evidence found at 0:20, A, finish 1:10, correct.
Q5, work rate: at 0:25,
1/12 + 1/18 = 3/36 + 2/36 = 5/36, A. Combined time is36/5 = 7.2 days, correct.
Selection is not timid skipping. It moves time from unsupported choices to checkable routes.
5. Weeks 4 and 5: climb a timer ladder through an accuracy gate
For comparable self-made sets, Rung A is 12 questions in 18 minutes, B is 12 in 15, and C is 12 in 12. Promote after three consecutive sets, each with at least 10 correct and at most 1 wrong.
At Rung A, record:
Set | Correct | Wrong | Skipped |
|---|---|---|---|
1 | 10 | 1 | 1 |
2 | 11 | 0 | 1 |
3 | 10 | 1 | 1 |
Totals are 10+11+10 = 31 correct and 1+0+1 = 2 wrong, hence 33 attempts and 31/33 x 100 = 93.9% accuracy. All pass, so try Rung B.
Two Rung B sets with 2 or more wrong each trigger two passing Rung A sets. Do not let speed amplify unstable methods.
6. Compare matched simulations by clean answers, not attempt count
Keep the same self-made 30-question mix and personal time cap. Do not convert the result into an exam score; compare accuracy and clean answers across simulations instead.
Simulation A has 29 attempted, 21 correct, 8 wrong and 1 skipped: 21/29 x 100 = 72.4% accuracy. After selection practice, B has 25 attempted, 23 correct, 2 wrong and 5 skipped: 23/25 x 100 = 92%. Four fewer attempts produce 2 more correct answers.
Simulation C has 27 attempted, 25 correct, 2 wrong and 3 skipped: 25/27 x 100 = 92.6%. B and C both sit at 92% or better, so the selection rule has held. Keep the mix fixed and let the timer ladder carry the rest.
Before the next simulation, review for 40 minutes: 10 recounting, 15 resolving wrong items, 10 inspecting parked or skipped items, and 5 scheduling reattempts. Check: 10+15+10+5 = 40. Starting another mock first leaves the loop incomplete.
7. Week 6: consolidate the repeatable plan
Schedule one matched simulation 1 h, review 40 m, two weakest-lane blocks 90 m, two mixed selection sets 60 m, and error-log reattempts 50 m. Total: 60+40+90+60+50 = 300 minutes = 5 hours. Run the simulation early enough to repair it. Add no topic unless the live rule card shows an omission.
KnowledgeGate's Mathematics and Aptitude, Reasoning, and Verbal Ability (English) banks hold about 20,000 practice questions between them, which is more than enough to build a fresh 30-question mix for every week of the plan.
The short version
Copy live rules, measure accuracy and coverage, repair causes, select on evidence, compress time through a gate, and review every simulation. IT Recruitment Exam Preparation gathers the placement-test track the pH Test sits in, and Mera Placement Hoga: Complete Placement Preparation is the route if you would rather follow a built schedule than assemble one.
