For a better exam season next time

The 2025 admissions period ended with many paradoxes: the benchmark score of 29 - 30 was widespread, some students with extra points still failed, and the virtual search and filtering system was malfunctioning.

Candidates taking the 2025 high school graduation exam in Ho Chi Minh City. This year, the admission scores of many universities have increased to a record high in the method of considering high school graduation exam scores - Photo: THANH HIEP

At that time, Resolution 71 of the Politburo required "building a project to innovate university admissions in the direction of properly assessing learners' abilities, ensuring unified control of input standards of majors and training institutions" - a clear guiding framework for the upcoming breakthrough.

In that framework, the recent turmoil shows the loopholes in the 30-point scale conversion, bonus point mechanism and technology operation capacity.

From a governance perspective, it is worth noting that listening capacity and consultation mechanisms have not been operating effectively, causing some decisions to not be timely based on scientific evidence and empirical data.

The result is "forcing" all methods to a 30-point scale despite measurement differences; announcing "percentiles" but lacking data transparency; talking about "competency assessment" but relying on "soulless" bonus points; risk management and accountability are both vague.

The three knots of the "tangle" are evident.

First, the technical understanding of "fairness" imposes different measures into the same mold, distorting the signal and confusing the standard: with the same capacity, one school passes, another fails.

Second, the priority paradox: tightening regional priorities but opening up certificate points, unintentionally shifting the advantage to the group with conditions, while disadvantaged students lose their small "shield".

Third, poor technology operations: a national event that happens only once a year should have strong infrastructure and backup plans; when technology can change the outcome of a person's life, it is no longer a system failure, but a management failure.

International experience reminds us of a simple thing: stable rules of the game and autonomy with accountability. South Korea keeps CSAT stable but allows schools to add criteria; Japan separates test organization from state management, schools design their own criteria to ensure stratification; the US uses transparent standard measures, schools can use or not but must explain with data. The common denominator is predictable rules, open data and clear accountability.

Short-term lesson: It is necessary to professionalize the question-making process to ensure the "dual goals" of both graduation and enrollment: standardizing the structure - matrix, building a question bank according to a standard scale, and independent pre-testing and post-testing.

From that foundation, the following steps should take place seamlessly: stabilize regulations within a minimum five-year cycle to end the "change of laws in midstream" situation; if multiple methods are maintained, there must be a national reference framework based on empirical research, publish models - data - errors, require piloting before expanding, and update according to a non-shocking roadmap.

At the school level, increase transparency by publicizing the proportion and effectiveness of each method, along with admission data by channel for social monitoring; in terms of technology, order admission platforms according to independent technical standards and thoroughly implement the principles: mandatory load testing before exam season to expose errors, track the system and open a complaint channel; where there is a failure, trace who is responsible. At the same time, stop thinking about "leverage" through bonus points.

Absolutely do not add or convert foreign language certificates unreasonably into admission scores; after admission, use valid certificates to exempt foreign language courses, place students in classes and/or recognize equivalent credits, without changing the admission score.

In the long term, admissions must become a standard signal for high school teaching and learning: not creating skewed learning motivation but measuring the right competencies that predict success in university (quantitative thinking, academic reading and writing, scientific and IT knowledge, foreign languages).

Moving from discrete points to standardized, verifiable evidence of competence; from administrative adjustments to evidence-based design with predictive value research across multiple courses; from single testing to standardized evidence sets (national exams that ensure minimum thresholds combined with domain-specific readiness assessments with clear thresholds, no mechanical additions), with support mechanisms to ensure opportunities for students in disadvantaged areas. And all innovations must go through pilot testing, data disclosure, and only be applied when proven not to push the system toward skewed learning.

A better competition season does not come from more terminology or more virtual filters, but from stable rules, scientific basis, transparent data and clear accountability.

When policy makers know how to listen and dare to experiment - measure - correct mistakes, when schools are given autonomy and accountability, and when candidates' interests are put first, the next exam season will have less sighs and worries and more smiles.

Back to topic

TS HOANG NGOC VINH

Source: https://tuoitre.vn/de-mua-thi-sau-tot-hon-20250829084313848.htm