Mock tests

Mizoram TET – Foundation & Exam Strategy

Topic: Mock Tests

Audience: Aspiring TET candidates, coaching staff, test‑designers
Level: MIT‑grade, first‑principles & rigorous reasoning
Goal: Master the theory, design, analysis, and strategy around mock tests to maximise exam performance and test‑development quality.

Table of Contents

1. What is a Mock Test?

Key insight: A mock test is both a practice tool and a research instrument. Treat it as a miniature version of the real exam.

2. Psychometric Foundations

2.1 Classical Test Theory (CTT)

• Formulation
  ( X = T + E )
  Where ( X ) = observed score, ( T ) = true score, ( E ) = error.
• Key metrics
  • Reliability ( r = \frac{\sigma_T^2}{\sigma_X^2} ) (Cronbach’s alpha for multiple items).
  • Item difficulty ( p = \frac{\text{number of correct responses}}{\text{total candidates}} ).
  • Item discrimination ( r_{iT} = \frac{\text{cov}(i, T)}{\sigma_i \sigma_T} ).
• Assumptions
  1. Unidimensionality
  2. Linear relationship between true and observed scores

Limitation for mock tests: CTT treats items as independent from examinee ability; not ideal for adaptive testing.

2.2 Item Response Theory (IRT)

• Model: 3‑parameter logistic (3PL)
  ( P(\theta) = c + \frac{1-c}{1+e^{-a(\theta-b)}} )
  • ( \theta ) = examinee ability
  • ( a ) = discrimination parameter
  • ( b ) = difficulty parameter
  • ( c ) = guessing parameter
• Advantages
  • Ability‑specific item information
  • Item parameters invariant across samples
  • Foundation for Computer‑Adaptive Testing
• Key metrics
  • Test information ( I(\theta) = \sum_i I_i(\theta) ) where ( I_i(\theta) = \frac{a_i^2(1-P_i(\theta))(P_i(\theta)-c_i)^2}{(P_i(\theta)-c_i)^2} ).
  • Standard error ( SE(\theta) = 1/\sqrt{I(\theta)} ).
• Application to mock tests
  • Calibration: Estimate ( a,b,c ) using pilot data.
  • Diagnostic reports: Provide candidate‑specific item‑level performance.

2.3 Reliability & Validity of Mock Tests

• Reliability: Must be high enough to differentiate high‑performers from low‑performers. Target ( \alpha > 0.85 ) for high‑stakes mock tests.
• Validity:
  • Content validity: Blueprint adherence.
  • Construct validity: Correlation with external measures (e.g., prior classroom performance).
  • Criterion validity: Correlation with the actual TET score (if available).

Takeaway: Mock tests should be psychometrically comparable to the real TET to be useful diagnostics.

3. Test Construction Principles

3.1 Item Bank Development

3.2 Sampling & Scaling

• Random Sampling: Ensure each mock test has a representative mix of items across the blueprint.
• Scaling: Convert raw scores to scaled scores to match TET scoring (e.g., 0–100).
  [