West Bengal Police Sub‑Inspector (SI) – Mock Tests
Foundation & Exam Strategy
(Advanced, MIT‑level treatment for the West Bengal Police Sub‑Inspector Examination)
Goal: Build a rigorous, data‑driven mock‑testing program that maximises reliability, validity, and test‑taker learning while reflecting the real exam environment.
Table of Contents
| # | Section | Key Topics |
|---|---|---|
| 1 | Introduction & Exam Overview | Exam format, content domains, marking scheme |
| 2 | Foundations of Test Design | Validity, reliability, fairness, test‑construction principles |
| 3 | Classical Test Theory (CTT) | Scoring, standard error, Cronbach’s α |
| 4 | Item Response Theory (IRT) | 1PL/2PL/3PL models, discrimination, difficulty, guessing |
| 5 | Item‑Level Analysis | Difficulty (p‑value), discrimination (point‑biserial), distractor analysis |
| 6 | Designing a Mock‑Test Item Bank | Content balance, parameter distribution, distractor construction |
| 7 | Administration & Logistics | Timing, environment, test‑taker instructions |
| 8 | Post‑Test Analysis | Score distribution, item statistics, learning curves |
| 9 | Edge Cases & Mitigation | Test‑anxiety, irregular patterns, low‑confidence items |
| 10 | Strategy for Exam Day | Time management, risk‑management, question‑selection heuristics |
| 11 | Case Study | Sample mock‑test, item‑analysis, feedback loop |
| 12 | Challenging Problem | IRT‑based test design & reliability calculation |
1. Introduction & Exam Overview
| Feature | Details |
|---|---|
| Exam Body | West Bengal Police – Sub‑Inspector (SI) |
| Format | Paper‑based + Computer‑Based Test (CBT) (optional) |
| Sections | 1. General Knowledge (GK) |
| Marking | +1 for correct, –0.25 for wrong, 0 for unattempted |
| Time | 120 minutes (Paper‑based) / 150 minutes (CBT) |
| Total Marks | 200 per paper |
| Passing Criterion | 40 % in each paper (i.e., 80 marks) |
| Attempts | Max 2 papers (optional third as reserve) |
Why Mock Tests?
- Benchmarking – Establish baseline performance.
- Item‑Level Feedback – Identify strengths/weaknesses.
- Time‑Management Training – Replicate exam timings.
- Confidence Building – Reduce test anxiety.
- Data‑Driven Study – Use psychometric analysis to refine study plan.
2. Foundations of Test Design
2.1 Validity
- Content Validity: Item bank reflects the syllabus and relative weight of each topic.
- Construct Validity: Items align with the underlying trait (knowledge, reasoning).
- Criterion‑Related Validity: Correlation of mock‑test scores with actual exam scores (historical data).
2.2 Reliability
- Internal Consistency – Cronbach’s α ≥ 0.80 for the mock‑test.
- Test‑Retest Reliability – Correlation (r ≥ 0.70) between two administrations.
2.3 Fairness & Bias
- Item Bias – Conduct differential item functioning (DIF) analysis to ensure no demographic bias.
- Difficulty Distribution – Target 30–40 % items at medium difficulty (p ≈ 0.5).
2.4 Test‑Construction Principles
- Item Pool: ≥ 500 items, evenly distributed across domains.
- Item Parameters: Difficulty (b), discrimination (a), guessing (c).
- Pilot Testing: Administer to a representative cohort; collect item statistics.
- Revision Cycle: Remove low‑discrimination items, adjust distractors.
3. Classical Test Theory (CTT)
3.1 Scoring Model
Let X be a test score, θ the true ability.
( X = θ + ε ) where ε is error.
3.2 Reliability
( r = \frac{σ_θ^2}{σ_X^2} )
Cronbach’s α approximated as:
( α = \frac{k \bar{c}}{1 + (k-1)\bar{c}} )
where k = number of items, (\bar{c}) = average inter‑item correlation.