The ins and outs of Scottish schooling are, for obvious reasons, not at the top of many people’s priority list right now, but one particular claim about the nation’s education system still managed to make headlines in recent days: apparently, Scotland is one of only three countries that refuses to publish exam data. 

This is a very significant and serious claim, and one that various politicians jumped on without a moment’s hesitation. 

There’s just one, small problem: it simply is not true. 

In actual fact, we publish a lot of exam data in this country. Most of it is provided directly on the SQA website, allowing anyone to access, download and analyse recent and historic statistics about examinations and qualifications. 

Read more: What is the attainment gap that politicians love to bring up so much?

This is why I can tell you all sorts of things about 2023 exam results in Scotland: 67% of Higher Biology students in private schools achieved an A grade; five of the ten girls who completed a Scottish Baccalaureate in Expressive Arts were awarded a Distinction; the average score in the National 5 English close reading paper was 15.7 out of a possible 30. 

If you’re more interested in historic data, that’s fine too – did you know, for example, that between 2009 and 2013 the pass rate for Higher Administration increased from 69% to 81%, or that just 1% of pupils who got a C for National 5 Maths in 2018 managed an A at Higher the following year? 

It is definitely, demonstrably, and – if we’re honest – very obviously incorrect to claim that Scotland does not publish exam data, so where did the story come from? 

The answer is the latest Education at a Glance report by the Organisation for Economic Co-operation and Development (OECD), which is the same body that organises and publishes the infamous PISA international school league tables. 

The Herald:

The report looks at a number of different features of education systems to allow for comparisons to be made between countries. It is an enormous and extremely technical document that is probably all but impenetrable to general readers, but if just do a search for the word ‘Scotland’ you’ll eventually end up on page 408 where you will find the following paragraph: 

“Most countries disseminate the results of national/central assessments aggregated at country/state level through press releases to the media and make them available to the general public through a public website. However, in the Slovak Republic, aggregated results are only available upon request and Romania and Scotland (United Kingdom) do not share any results.” 

This is the source of the false claim that Scotland doesn’t publish exam data, but the paragraph in question is actually referring to an entirely separate type of assessment. 

This isn’t about the exams that young people sit before they leave school, but rather the ways in which we measure things like literacy and numeracy at primary and lower secondary levels.  

The Herald:

In Scotland we use teacher-generated data called Achievement of Curriculum for Excellence Levels (ACEL) to do this, with allegedly diagnostic standardised testing helping to inform decisions about the progress of each individual pupil. The OECD has, in fairness, been quite clear about this, and just a couple of pages earlier had even explicitly mentioned the Scottish National Standardised Assessments. 

In fact, it is the failure to publish results of those tests that makes Scotland stand out in the OECD’s report but, even then, there’s a very good reason for it: the data would be worthless at best and utterly poisonous at worst. 

As the OECD itself points out, our standardised tests are not taken at the same time by all students across Scotland. This means that some pupils might go through them in August while others wait until the following April, raising obvious implications for the comparability of the data. Put simply, there’s no guarantee that two test scores are really comparing like with like, and that’s before we even deal with the fact that lots of young people don’t sit the tests, which are not compulsory and are not used by all schools. 

There is, then, a perfectly clear explanation for the apparent anomaly that the OECD identified, but that still doesn’t mean that all is well. The truth is that Scotland’s system for tracking literacy and numeracy, and checking up on the overall performance of the system, is not fit for purpose. 

Teachers make judgements about their students every single day, and their professional expertise in this area should be trusted, but that does not preclude the need for a regular, objective snapshot of the overall school system. 

Read more: We know 2+2=4, but is numeracy in schools more than just counting?

That is why, back in 2021, the OECD explicitly advised the Scottish Government to “re-develop a sample-based evaluation system to collect robust and reliable data.” They were referring to the Scottish Survey of Literacy and Numeracy (SSLN), which provided us with excellent data about what was happening in our schools. It used a sample-based method to ensure that there was no way to coach pupils for the assessments, and the results were never tied to individual pupils, teachers or even schools, which meant that there was no reason to try to game the system. All of that made the data extremely reliable, and indeed it was the results of an SSLN survey that sparked the whole ‘judge me on my record’ era in Scottish education.  

And then the SNP decided to replace it with something far less accurate, useful and robust, because it suited them politically. That is the real scandal. 

So no, it’s not correct to say that Scotland doesn’t share exam data; and no, we shouldn’t start publishing the scores from the government’s standardised tests. 

But we do need to solve our problem with poor education data, and the inevitable impact that it has on decision-making, political debate, and the public’s ability to hold leaders to account. 

That’s why we should take the OECD’s advice and reintroduce, then expand, and once again publish, the Scottish Survey of Literacy and Numeracy.