Risk, Probability & Statistical ThinkingIndependent failures, expected loss and the gambler's fallacy — probability for managers who decide.
Thirteen quantitative items and five scenarios: independent and conditional probability, expected value, reading a distribution in plain words, and the judgement to apply statistical thinking when a rare loss would exceed the whole budget. Twenty-six minutes; scored on Numerical Reasoning and Decision Quality.
Statistical thinking applied to management decisions, not dice
One item gives a factory two independent suppliers, each with a small chance of a major failure this year, and asks for the chance that at least one fails — the tempting answers are the sum and the product. Another offers a launch now with a stated chance of a large profit against a certain smaller profit later, and asks which wins on expected value and by how much. A third describes a consultancy that has lost four straight bids with a stable win rate and asks the chance of winning the next. The scenarios then ask what to do when the expected loss says do not insure and the actual loss would exceed your unit's annual budget.
All scenarios, companies and datasets are invented and described in text; the items are original constructs with no published assessment content reproduced. The 13 quantitative keys are verified by computation, and the distractors are the standard errors — adding probabilities that are not exclusive, expecting a streak to correct itself, treating expected value as the only criterion. The five situational items use a graded key: the response that separates expected value from ruin risk earns full credit, a defensible but incomplete response earns partial credit, and deference to the dashboard earns nothing.
The report scores Numerical Reasoning and Decision Quality from their own tagged items and bands each with a coaching interpretation, so a manager who computes correctly but decides as if variance did not exist sees that named. All 18 items are reviewed with the working or the graded reasoning — the complement rule, the expected-value arithmetic, why a stable rate has no memory, why a survivable premium against an unsurvivable loss is not waste. The growth plan is built from the weaker competency.
What you walk away with
Numerical Reasoning covers independent and conditional probability, expected value and distributions; Decision Quality covers what you do with the number when the tail could ruin you. Each is banded with a coaching interpretation, so the report names which half is weaker.
Situational options are graded, so the overall 0–100 score and its band reflect how consistently you separated expected value from ruin risk and a stable rate from a streak, rather than how often you avoided the worst option.
Every item returns with your choice, the key and the computation or graded reasoning — the complement rule for at-least-one failures, why four losses do not raise the next win, when a premium above expected loss is rational. It doubles as a refresher in applied probability.
Strengths and development areas are drawn from your own responses, and the plan targets the weaker competency with habits — computing the complement, asking about the worst case, separating variance from trend — usable in the next steering meeting.
Inside your report
Illustrative sample — your report is generated from your own responses.
Every competency this test declares, scored from the items tagged to it — so the shape is yours, not a template.
Assembled from your lowest competencies, in order — not a generic checklist.
Band edges at 40, 60 and 80. Each band carries its own coaching interpretation for that competency.
Built for
- Managers and decision-makers who sign off on launches, insurance, supplier risk or forecasts
- Hiring teams assessing candidates for risk, planning, procurement and general-management roles
- Analysts moving into roles where they must explain probability to non-technical leaders
- Executives who want to know whether their intuition about streaks and averages is reliable
Four lost bids do not make the fifth more likely — and neither does wishing.
Eighteen items in 26 minutes, every probability and every graded call explained.
Take free · full report ₹150 (incl. GST)
Frequently asked questions
Two linked abilities: computing probability and expected value correctly — independent and conditional events, at-least-one failures, expected loss, reading a distribution in plain words — scored as Numerical Reasoning across 13 items; and applying that thinking to real management decisions, such as insuring against a rare loss that would exceed the budget, scored as Decision Quality across five scenarios. It suits managers, risk and planning roles, and the teams hiring for them.
The 13 quantitative items have single keys verified by computation, and the distractors are the intuitive errors — adding probabilities that are not exclusive, expecting a losing streak to correct itself, treating expected value as the only criterion — so intuition mostly lands on a wrong option. The five situational items are graded: separating expected value from ruin risk earns full credit, partial reasoning earns partial credit, and deferring to the dashboard earns nothing.
Twenty-six minutes for 18 items in one sitting, on any device. The report is generated immediately: an overall 0–100 score with a band, banded bars for both competencies with a coaching interpretation each, a radar, a personalised narrative, strengths and development areas, a growth plan and a review of all 18 items with the computation or graded reasoning behind each key. Everything is included; there is nothing further to unlock.
Hiring teams set a band threshold for risk, procurement, planning and general-management roles and use a scenario from the review in interview — asking a candidate to defend not insuring a plant is a good conversation. Management teams sit it together as a baseline before a forecasting or risk-appetite exercise and retake later. There is no pass mark; the bands and the item review carry the decision.
Yes. Probability does not change with geography, and the scenarios use invented companies with no local regulation; where a currency appears in an item it is incidental to the arithmetic. The report is written in plain international English. The test is priced in INR and USD, with the live price for each shown on this page, so international managers and global employers pay in whichever currency suits them.
Numerical, verbal, logical, abstract, spatial, data and workplace-judgment banks — each an original item bank with every rule stated in the explanation, scored per competency and reported with a banded profile and a growth plan. Sit one, or build a battery from several.
See the aptitude catalogue →Methodology: All scenarios, companies and datasets are invented and described in text. Items are original constructs; no published assessment content is reproduced or paraphrased.