Not so NEET: Why normalised scores in computer tests might not work as in JEE, Pg17

Key Highlights:

• The National Eligibility-cum-Entrance Test (NEET) is considering shifting from a single-session pen-and-paper (PnP) exam to a multi-session computer-based testing (CBT) format.
• This shift aims to mitigate issues like paper leaks and transportation vulnerabilities associated with the traditional NEET-PnP.
• Conducting NEET-CBT for approximately 24 lakh students may require around 20 sessions with different question papers, necessitating score normalization.
• The National Testing Agency (NTA) plans to use the same normalization methodology for NEET-CBT as in JEE (Main).

Detailed Insights:

• The primary challenge in multi-session CBT is fairly comparing candidates who attempt different question papers across sessions due to varying difficulty levels.
• Normalization statistically adjusts raw marks to compensate for session-wise variations, ensuring candidates are not unfairly advantaged or disadvantaged.
• A candidate's performance is assessed relative to others in the same session using percentile scoring, indicating the percentage of candidates scoring equal to or below them.
• Concerns arise regarding the precision, transparency, fairness, and verifiability of normalized percentile scores, as no normalization method is entirely error-free.
• In NEET, even minor normalization adjustments may significantly alter ranks and admissions, especially for candidates near critical cutoffs, potentially creating distrust.
• Unlike global digital exams like SAT and GRE, NEET and JEE are not adaptive tests and serve as single-criterion rank-based admission tests.
• To ensure fairness, question-difficulty balancing, session equivalence testing, disclosure mechanisms, and independent technical oversight are essential for NEET-CBT.
• Disclosing both raw and normalized scores can improve transparency, but potential ranking distortions may still generate distrust and controversy.

Key Concepts Involved:

• Normalization: A statistical process used to adjust scores, ensuring fair comparison when different test forms are used.
• Percentile Score: Indicates the percentage of candidates scoring equal to or below a particular candidate in the same session.
• Optical Mark Recognition (OMR): The process of electronically extracting intended data from marked fields on forms like answer sheets.