Eighteen Days, One Attempt: Allocating Study Time Against the Blueprint
Priya has booked the OK POST examination for 18 days from today and has one attempt left before her authorization window closes. The candidate handbook's blueprint assigns weights of 10% to Domain 1, 15% to Domain 2, 25% to Domain 3, 20% to Domain 4, 20% to Domain 5 and 10% to Domain 6. Across four 30-question practice sets her per-domain accuracy is D1 88%, D2 70%, D3 52%, D4 58%, D5 82% and D6 90%. She works 45 hours a week and has roughly 40 study hours available before test day, mostly in 45-minute blocks on a commuter train where she cannot run the software that Domain 4 questions describe. Her employer will fund one reschedule if it is requested at least 7 days before the appointment. She has read each domain once and has never sat a timed full-length practice test.
1. Recommend how Priya should split the 40 hours. Show the calculation and the assumptions behind it.
Read the worked answer
Method. Score still available is better estimated by weight x (100 - accuracy) than by raw accuracy, because a weak low-weight domain is worth less than a middling high-weight one.
- D1: 10 x 12 = 120, share 4%
- D2: 15 x 30 = 450, share 15%
- D3: 25 x 48 = 1,200, share 39%
- D4: 20 x 42 = 840, share 27%
- D5: 20 x 18 = 360, share 12%
- D6: 10 x 10 = 100, share 3%
- Total 3,070
Split. Reserve 6 hours for two timed full-length tests plus review; divide the other 34 hours in the shares above: D3 13.5, D4 9, D2 5, D5 4, D1 1.5, D6 1.
Assumptions and corrections. Thirty questions across six domains is about five items each, so 52% versus 58% is roughly one question - treat D3 and D4 as one 22.5-hour block and re-measure rather than tuning to noise. The method also assumes the practice sets are blueprint-representative; if they were drawn from memory or a single source, reweight toward D3 and D4 regardless. Domain 4 needs a machine, so its hours go to evenings at a desk; train blocks carry D3 and D2 reading and recall. D1 and D6 keep a small spaced-retrieval dose, not zero.
2. On day 10 she re-tests at 66% on Domain 3 and 61% on Domain 4, with 8 study hours left. Should she reschedule? Set out the decision rule before answering.
Read the worked answer
Rule first, verdict second. A defensible rule is: sit only if a timed full-length score is at or above 72% and both D3 and D4 are above 70% on the most recent attempt; otherwise reschedule. Fixing the rule before seeing the numbers is what stops her from moving the target.
Applying it. D3 moved 52% to 66%, about 14 points, which is real progress. D4 moved 58% to 61%, roughly one question on a five-item practice set - statistically flat. Eight hours at that observed rate will not lift D4 to 70%. Against the rule, she fails on two counts.
Weighing the cost. Rescheduling does not consume her remaining attempt, it defers it; the cost is a delay and the employer's single funded reschedule. The genuine risk is a rescheduled date falling outside her authorization window, which she must confirm before accepting the new slot. Sitting now risks burning the last attempt on evidence that already says she is short.
Condition. If a full-length mock on day 12 comes in at 75% or better with D4 above 68%, the rule is met and she should sit. Absent that, reschedule and keep the 8 hours for D4.
3. A colleague advises skipping Domain 6 entirely because it is only 10%. Evaluate that advice.
Read the worked answer
The arithmetic alone does not settle it. At 10% weight, a 10-item contribution, dropping from 90% to 55% costs about 3.5 points of total score. That matters if her expected total sits near the pass mark and is irrelevant if she is 15 points clear - so the advice depends on her margin, which she cannot currently estimate because she has never sat a full-length test.
Two further arguments run against skipping. First, risk asymmetry: with one attempt and a history of weak D3 and D4 performance, she is more likely to be near the cut than far above it, and points lost to decay are the cheapest points she owns - one maintenance hour protects them. Second, safety and risk reasoning is not an isolated island. Scenario items in Domains 3 and 5 routinely hinge on hazard, residual risk and stop-work judgement, so decayed Domain 6 knowledge reappears as errors elsewhere and is effectively double-counted.
The correct response to a low-weight domain is a small spaced-retrieval dose - a short review every third day - not deletion and not extra hours. Priya should reduce Domain 6 to about one hour and reallocate nothing back to it unless a mock shows D6 accuracy below 75%.