2020 research paper proposing that national testing in England and USA "have shown little or no improvement over the years".

The writers, David Bolden and Peter Tymms, use daya from PISA, TIMSS and PIRLS to show that - other than a dramatic improvement between 1995-2000 - educational standards have remained "remarkably stable over time".

The key issue raised is that, despite decades of continuous reforms, there has been a lack of change in educational standards.

Conclusion

From the conclusion:

This paper has made it clear that educational standards, by which we mean levels of academic attainment, of affluent countries, are incredibly stable over a decade and more. Further, major efforts at improvement have noticeably failed. From this, one lesson is clear; we need to alter the way we view large-scale educational change. That is, we should think of national educational change as a process of small incremental improvements, which may accumulate over long periods of time (decades). We should expect these increments to be difficult to both attain and maintain. We should be sceptical of quick answers, which we can see, with hindsight, are often superficial and glib. If this could be widely accepted, it should deal with the first two explanations of the lack of change listed above: speed and frequency of reform.

The writers propose:

• restructure mechanisms of national policymaking: government should set general direction but leave educational decisions to a "non-political body";
• a permanent advisory research-based unit that can be called upon to provide evdence-based advice on on-going issue;
• a large-scale, multi-disciplinary overarching theoretical structure: "the most serious potential explanation for the general failure of educational reforms is our general lack of deep understanding of educational systems".