By Datuk Omar Mustapha

The Numbers We Do Not See
Let us begin with what can be said plainly.
In England, roughly seven in ten students pass GCSE Mathematics and English at age 16. In Malaysia, SPM pass rates tell a similar story on the surface. Approximately three in four pass Mathematics. English somewhat higher. On headline terms, Malaysia does not appear weaker.
That is where most comparisons stop. It is also where they go wrong.
The real difference between the two systems is not the number who pass. It is the number of things each system is willing to tell you about those who do not.

The Numbers We Argue About
Recently, a set of figures purporting to show SPM 2025 results by ethnicity began circulating online. The Ministry of Education responded swiftly. The figures were declared false.
That response was necessary. False data must be corrected.
But it was not sufficient.
SPM pass rates are a narrow construct. A student does not obtain the certificate if they fail Bahasa Malaysia or History, regardless of performance elsewhere. The aggregate reflects those requirements as much as it reflects attainment across subjects.
The table may have been wrong on the numbers. It was not wrong in what it was trying to do.
It was attempting to answer a question the system does not answer directly.
How are different communities performing in Mathematics and English.
When that question is not answered with official data, it does not disappear. It is answered instead by speculation.
And speculation, once it enters the system, is very difficult to displace.
To understand what an honest system looks like, it is worth examining one.

What England Knows
In England, the Department for Education publishes annually not just pass rates, but who is passing and who is not.
For the 2022 to 2023 cohort, 65.1 percent of pupils achieved a pass in both English and Mathematics. Chinese students achieved 88.6 percent, the highest of all ethnic groups. Indian students were the second highest performing group nationally. White British students performed at 63.6 per cent, just below the national average.
This is not commentary. It is state-issued data, published as a matter of routine.
It can be tracked year by year. It can be disaggregated by income, region, gender, or school type. It can be tested against intervention.
England does not debate whether there is a gap. It debates what to do about it.
Malaysia cannot yet do the same.

What Malaysia Cannot Say With Precision
Malaysia can say this. Around one in four students fails Mathematics. Tens of thousands fail English. National averages are improving.
What it cannot do, in any routine and defensible way, is describe the distribution.
What proportion of Malay students passes Mathematics each year. How that compares with Chinese or Indian students. Whether the gap is narrowing or widening.
There is no continuous public series.
There are fragments.
The Malaysia Education Blueprint, published in 2013, cited performance in the 1119 Cambridge English paper, the external benchmark sitting alongside SPM, at credit level: based on the 2010 SPM examination. Bumiputera students at 23 percent. Chinese at 42 percent. Indian at 35 percent.
That is a serious gap.
The more revealing fact is the date.
More than a decade has passed. The examinations have continued. The data has been collected.
The system knows.
The public does not.

The Difference Is Not Data. It Is Disclosure.
England did not arrive at this level of visibility by accident. The Freedom of Information Act 2000 changed the operating environment of the state. Once information can be requested, it becomes easier to publish than to withhold. Over time, this produces standardised datasets, continuous time series, and independent scrutiny. Precision becomes normal.
Malaysia operates differently. Data is released episodically, selectively, often through safer proxies such as income or geography. Ethnicity, where it matters most institutionally, is handled cautiously. Not because the data does not exist. But because of what the data might imply.
This is the wrong calculation.
When a system answers a question only by denial, it leaves the field open for the next, more convincing version of the same claim. The Ministry was right to correct the false table. But correction does not answer the underlying question. It merely postpones it.
The alternative is not complicated. It is simply harder. The question must be answered directly, with official data, not selectively, not once, but routinely.

From Disclosure to Intervention
Publication is not the end point. It is the precondition for something more consequential.
Policy is not made from averages. It is made from distribution.
If one group sits at 88 percent and another at 50, the question changes immediately. It is no longer whether the system is improving overall. It becomes why the movement is uneven.
That is where intervention begins.
Without disaggregated data in the public domain, a ministry can only spend visibly. With it, it can target. Language. Teachers. Early foundations. The environment into which the child is placed before the school day begins.
The Ministry of Education almost certainly holds this data. The picture exists. The limitation is not analytical. It is political. No party in government, and no party waiting for government, has yet judged it safe to publish what the state already knows. That is not a technical failure. It is a choice, made repeatedly, across administrations, to manage political risk at the cost of educational precision.

The Structure Beneath the Numbers
To understand why that political choice has been so consistently avoided, it is necessary to look at what the distribution actually shows.
Malays, who form the numerical majority of the student population, show more uneven attainment in Mathematics and English, with a heavier lower tail. Chinese students, in aggregate, are more consistently strong in Mathematics. Indian students tend to perform more strongly in English.
If this pattern holds, the consequences do not remain within each community.
They accumulate.
High-productivity sectors draw disproportionately from Chinese and Indian Malaysians. Malays carry the weight of uneven foundational capability.
The system continues to function.
Over time, it functions with less depth.
In a region where countries such as China are raising baseline capability across entire populations, that narrowing of depth is not a marginal concern. It is a structural one. And it is a structural concern that neither Chinese nor Indian Malaysians can afford to treat as someone else’s problem.
The temptation, for Chinese and Indian Malaysians reading this data, is a form of quiet relief. Their communities perform well in Mathematics. Their children are in the pipeline. The gap, on this reading, is a Malay difficulty.
That reading is not merely ungenerous. It is structurally illiterate.
A high-productivity economy that draws disproportionately from Chinese and Indian Malaysians, while Malays remain concentrated in the lower rungs of the capability distribution, is not a stable arrangement. It does not produce inter-ethnic harmony. It produces the conditions for exactly the politics that Chinese and Indian communities fear most: redistributive pressure, institutional resentment, and the periodic mobilisation of Malay economic anxiety as a substitute for genuine policy.
Chinese and Indian Malaysians have lived through enough of that history to know its shape.
The social contract in a multiethnic society does not hold because of goodwill. It holds because the gains from the system are distributed broadly enough that no major community feels systematically excluded from them. When that condition fails, the response is rarely liberal. It is communal.
A Malay community that is educationally underprepared in the subjects that determine economic participation is therefore not a Malay problem with Malay consequences.
It is a structural risk to the stability of arrangements that Chinese and Indian Malaysians equally depend upon.
Chinese and Indian Malaysians do not need to care about Malay attainment out of generosity.
They need to care about it out of an accurate reading of their own interests.

The Question That Remains
That reading requires honesty about where the failure sits.
If Malays continue to show uneven attainment in Mathematics and English, the question is no longer whether there is a gap.
It is whether the failure sits within the student, or within the system.
The response depends entirely on that answer.
If the system is delivering and the outcomes remain uneven, policy will not correct what is not being taken up.
If the system itself is not delivering, effort will not compensate for what is not taught.
The distinction cannot be avoided.
Because if Malays do not consistently meet the baseline required by a modern economy, the consequence does not remain within the community.
It sits with the country.

What the Next Generation Is Inheriting
That national consequence is not distant. It is already arriving, and it is arriving through a generation the system has not prepared honestly for what it is about to face.
The generation now sitting SPM is the first that will enter a labour market shaped by artificial intelligence at scale.
This matters for the data argument in a specific way.
The jobs that historically absorbed students who left school with weak Mathematics and English, clerical work, administrative processing, low-complexity service roles, are precisely the jobs that automation is eliminating fastest. But before that elimination completes itself, something else has happened.
The gig economy arrived first.
It created a vast intermediate layer of delivery riders, ride-hailing drivers, and logistics runners that absorbed the pressure quietly. Young people with insufficient foundational capability found a market for their labour. They are employed. They are earning. They do not appear in any unemployment statistic. And so the system looks like it is functioning, because the market found a use for people it was not equipped to develop.
A generation of Grab riders requires neither Mathematics nor English. They have been reduced, without drama and without anyone declaring it a crisis, to the automatons of the mass consumption economy.
That is not a safety net. It is a pressure valve. And it has allowed the state to avoid confronting what the education system is actually producing, because the consequences are dispersed across hundreds of thousands of individual arrangements that feel like choices but are in fact the residue of systemic failure.
The ledge, however, is temporary.
Autonomous vehicles, drone delivery, automated logistics. The roles absorbing underprepared graduates today are not secure. They are next. The generation that fell through the educational net is now building its livelihood on the next category scheduled for disruption.
The sequence is therefore this.
The school system fails to build foundational capability evenly. The gig economy absorbs the gap invisibly. Automation removes even that absorption. And the state, throughout, declined to measure what was happening honestly enough to intervene.
A country that will not look at what its schools are producing cannot see what its economy is about to lose.
That is not a governance failure in the abstract.
It is a specific generational betrayal, conducted quietly, across administrations, in the name of political caution.

The Pandora’s Box Argument
The reluctance is understandable.
Disaggregated ethnic data, in Malaysia’s political context, is rarely received as clinical. It is weaponised. The false table that circulated online is itself evidence of the risk.
But that risk is already present.
Opacity has not prevented politicisation. It has ensured that it occurs without anchor.
Each year the data remains unpublished, the vacuum fills. With rumour. With grievance. With a form of certainty that no ministerial correction can fully displace.
The cost of opacity is not neutrality. It is accumulated institutional mistrust.
The question is not whether Malaysia can afford to open this data.
It is whether Malaysia can continue to afford not to.

The Political Question
Transparency of this kind does not emerge organically. It is chosen.
The United Kingdom chose it, and then locked it in through law.
Some Malaysian states have enacted Freedom of Information legislation. The federation has not.
Pakatan Harapan argued for transparency in opposition. It now governs.
The question is no longer theoretical.
It is whether Malaysia is prepared to see, and to allow others to see, the system as it is.
A government that cannot tell you which communities are falling behind cannot seriously claim to be governing for everyone.
And a system that defeats false data with silence rather than truth has not defended the integrity of its numbers.
It has deferred the point at which it must answer for them.

Footnote: Two Eras, One Standard That Moved

There is a historical dimension to this argument that deserves its own acknowledgment.

The SPM was not an invention of the Malaysian state. It was the direct structural descendent of the British GCE O-Level, introduced in 1964 as Malaysia’s national equivalent of that qualification. For the first fourteen years of its existence, the examinations were conducted by the University of Cambridge Local Examinations Syndicate. Cambridge handed administrative control to Malaysian agencies in 1978, but retained an advisory role on standards. The 1119 English paper continued for some years thereafter to carry dual grading by both the Malaysian Examinations Syndicate and Cambridge, meaning that a student’s results slip showed two grades simultaneously: one domestic, one externally verified.

That external verification has since been discontinued. The 1119 paper no longer carries a Cambridge grade. The SPM English paper is now assessed entirely within the domestic system.

What has happened across the same period on the British side is also relevant. The GCE O-Level was abolished in the United Kingdom in 1988 and replaced by the GCSE. That replacement examination underwent a further deliberate reform in 2015, increasing content rigour, removing coursework components, and raising the threshold for a standard pass. Most UK universities now require an SPM Credit, Grade C6 or above, to consider it comparable to a GCSE Grade 4.

The trajectory of the two systems has therefore not been parallel. One has been periodically toughened by external pressure and institutional accountability. The other has adjusted its domestic floor downward in the interest of broader certification. Neither fact is contested. What remains unexamined is what that divergence, accumulated quietly across four decades, has cost the students who passed through it.