02 March 2004

I congratulate Mike Tomlinson and the Working Group on a far-sighted and ambitious blueprint for reform, which opens up a long-term agenda of profound importance to the future of education.

I applaud the clear statement in the report that implementation of a reform of this magnitude will take a decade.

In his media appearances, Mike Tomlinson has talked about evolution not revolution. This is a critical point.

Equally important is the point that the evolution must be managed, not left to serendipity. The task before us now is to develop the sequence of reforms which need to be put in place to the year 2014, and to decide the year in which each of those reforms will come on line, so that the first full year of the operation of the four diplomas will be the seamless outcome of an effectively managed process.

In my remarks today, which focus on assessment, but are not limited to it, I start from the assumption:

• that the units of which the diplomas are comprised will all carry a credit value;
• that the units will be of various sizes which will be reflected in the amount of credit;
• that some units must be taken before others, and
• that the diploma will be awarded on completion of a total number of points built up in core and main learning according to certain rules of combination.

That seems to me to be a necessity once QCA, as the contracted builder, gets down to the work of actually constructing the qualification according to the specifications of Tomlinson and Company, Architects for the Future, who have been commissioned by the owner - the Secretary of State on behalf of the people of this nation.

It also recognises that the14-19 Reform, the Skills Strategy and the Higher Education White Paper are integrated elements within a single, massive and complex reform process, which is designed to maximise the human capital of this country.

The 14-19 element must establish pathways that lead to further and higher education and training. It must articulate with the other elements of the total reform.

The task before us then is to move deliberately forward with a program of managed evolution on a specified timeline extending over the next ten years, informed by the clear sense of direction given by the report.

With regard to assessment, on which I have been asked to dwell, I intend to focus on:

• the changes which could be made in assessment;
• the sequence in which such changes should be made;
• and the year in which each of the changes might best be made.

What are the changes we should make in assessment?

First, we must reduce the overall summative assessment load in secondary education. The issue should no longer be debated. The jury is in and the verdict is delivered.

The owner and the architect are unambiguous on this point, and QCA as the builder, and the awarding bodies as our subcontractors, are agreed that it can be done.

And it should be done as early as possible in the construction period.

Almost all GCE qualifications have six teaching and learning modules associated directly with six discrete assessment units - three AS and three A2 units.

One of the A2 assessment units is designed to assess synoptically the links between the whole learning programme in each: there is no requirement for there to be a one-to-one link between teaching modules and assessment units.

The simplest way of reducing the assessment load is to leave the number and content of teaching and learning modules unchanged, and to reduce the number of assessment units by developing new and fewer assessment units that assess more than one teaching module.

We could aim, say, to reduce the six assessment units to four.

This would reduce assessment by one-third, but provided there is sufficiently reliable sampling of content and skills, I am strongly of the belief that the assessment would be as valid and reliable as at present.

I don't believe our system of assessment is more valid by virtue of its load than systems in many other countries, which don't so frequently pull up the plant to look at the roots.

Such a reduction would require only minor changes to the qualification criteria, although more substantial change would be required to the subject criteria.

It is possible that this could be achieved for a September 2006 start. The timings are tight and carry some risk, and would require the awarding bodies to have high quality specifications and guidance materials available to schools and colleges in good time.

But it can be done, if we:

• start redevelopment work immediately following the Government response to final report, say September 2004
• have the awarding bodies submit revised specifications in March 2005
• accredit new specifications and publish new guidance materials in September 2005
• begin first teaching under the new assessment regime in September 2006, for AS awards in August 2007 and A level awards in August 2008.

That is a tight timeline, but it is not a rushed one if it is managed well.

The lesson we surely have learned from 2002 is not to be slower, but to manage better. The Earth will not stop spinning on its axis: this country is behind its competitors, and its education and training reforms need strategic and agile management, not expansive and relaxed contemplation.

Reducing the assessment load in GCSE examinations is less easy to bring about in the short-term, because they are not a universally unitised qualification.

However, the pilot programs set up by QCA to investigate different ways of reducing the external assessment load, and to explore different ways of developing coursework tasks that encompass more than one subject, should provide valuable evidence to support the development of the personal challenge and extended projects recommended by the Interim Report.

Second, the decoupling of AS and A2.

I think Mike Tomlinson should go down in history for, amongst his many fine other contributions to this country, giving currency to the term decoupling. It conjures up images of uncoupling, of separation, of satellites leaving their docking stations, of divorce, of train carriages being split off into different branch lines.

It is a wonderful term, and decoupling is in fact inevitable.

And that is not just because of the potential gains, which are

• a probable reduction in the resitting of AS units to achieve extra UMS marks to improve the overall A level
• it would make it easier for students who move institutions at the end of their AS courses to transfer to an institution using different A level syllabuses, and
• it would reduce the assessment load for some, although my hunch is that most still would want to take it.

The inevitability arises from the fact that we can't move from the A level qualifications to an Advance Diploma qualification without decoupling.

The question is not whether we decouple, but when.

And the answer to that depends on how much time we need to develop the set of units within the diploma which carry forward the existing A level standard.

Because until we do, decoupling would, at a stroke, mean that there would no longer be an A level standard. That is why the interim report is so cautious on this point. There would be an AS standard and an A2 standard.

There is no separate A2 standard at present – there is an A level standard that arises from the combination of AS and A2 units. If the A2 were deemed to be the new A level, then the immediate effect would be to depress the overall levels of A level grades by removing the contribution made by marks achieved on the AS units.

We have no choice but to undertake the necessary decoupling, yet it is an imperative that we also maintain the A level standard and carry it forward into the diploma system.

This means we first have to develop new A level units and qualifications.

One aspect of the current A level standard is the volume of learning needed within a qualification.

In a linear subject like French or pure mathematics, students need to understand the material in the AS course to be able to tackle the A2 course successfully.

In some other subjects - perhaps history is an example - the A2 is more free standing, and students might be able to do well on the A2 course without having covered the AS material at all.

So, our objective should be to develop modified advanced level qualifications with distinct and separate A1 and A2 components which will maintain the current A level standard.

It is clear that this work will need to be done by design, according to the architect’s specifications, and across all advanced level qualifications.

We can't decouple before that work is complete. Once it is done, the decoupled qualifications would then carry forward into 2014, when they will cease to be stand-alone qualifications and will become credit-carrying units within the Advanced Diploma qualification.

As the builder, I await the architect's instructions, which will reflect the wishes of the owner – the Government representing the people of this nation. The nature and timing of decoupling is therefore best left until the final report of the Working Group, which will be the blueprint from which we will work.

Third, fit-for-purpose assessment.

The interim report states that the assessment system should:

• suit the subject, level and type of learning;
• be valid, reliable and have wider currency;
• help learners extend, deepen or consolidate their learning;
• provide access to a range of assessment experiences;
• provide formative feedback;
• limit the assessment burden on students, teachers and awarding bodies;
• embrace the potential benefits of e-assessment, and
• make appropriate use of teacher judgement.

It proposes that assessment within the diploma should '...provide formative feedback on progress' and that '...effective integration of teacher-led assessment into the diploma framework is central to a more responsive and effective assessment system.'

In other words, it envisages the harnessing of formative assessment for summative purposes.

There will of course need to be a range of quality assurance procedures in place, almost certainly as some form of external moderation, if formative assessment evidence is to be sufficiently reliable for summative use.

But unless those quality assurance procedures are suitably non-intrusive it likely that the value of the formative assessment will be reduced.

The enormous 'win' that the report envisages, and needs exploring carefully, is whether or not it is possible to integrate true formative assessment into the diploma assessment system.

The interim report points to the 'academic drift' that has occurred in the assessment of vocational learning and the move away from practical and continuous assessment towards external written examinations.

One reason commonly given for the drift is that there is a strong historical and cultural belief that rigour and reliability can only be assured through the use of carefully controlled forms of external assessment. These cultural beliefs will only be shifted if it can be demonstrated that alternative forms of assessment are equally valid and sufficiently reliable.

In my view there is a far more compelling reason: written examinations are far less resource intensive than on-the-job assessment of the core of vocational performance.

And that core is the observation and reporting of the coordination of hand, mind and eye - the essential attributes of what we call vocational education and training - and inspection of the qualities of the finished product.

Coordination of eye, mind and hand - in working with a particular type of wood to build a cupboard, in web design, in open heart surgery, in flying an aircraft, in solving a plumbing problem in a heritage listed house - is the essence of a vocation.

Much finer discriminations can be made in relation to performance in vocational areas than in so-called academic or general subjects.

But much of it is one-to-one. The on-the-job verification and assessment of performance in surgery and in flying, for example, is incredibly labour intensive and draws upon very fine discriminations of performance against a multitude of indicators.

I don't believe that vocational education is interested only in mastery or competence. Go to the UK or Worldskills competitions and you will be quickly disabused. For four days, each competitor - many of them from our colleges and only 17 years of age - has an individual assessor watching and assigning value to every move.

Every qualified motor vehicle repairer is competent, there are some you would never return to.

Performance in vocational education can not only be graded, but in my view graded to a much greater and more perfect level of discrimination than, say, A level history. It is just so costly and labour-intensive that it is prohibitive.

So, in vocational education - which we need to grow rapidly in terms of participation and achievement over the next ten years to give us a stable platform on which to build our diplomas - we have immense work to do.

We talk of the involvement of industry in both designing the content of subjects and in the assessment process. 'Fitness-for-purpose' in vocational education flows off the tongue.

But it means more than doing away with written examinations. It means replacing them with something far more robust.

How are we are going to do it?

That takes me to the fourth point, which is e-assessment.

In the area of vocational education, City and Guilds have done some immensely important work that breaks new ground in fit-for-purpose assessment in vocational education.

The assessment of vocational attainment depends in large part on the quality of the verified product. City and Guilds are developing a global e-assessment process. The example I have seen is weaving, where - for example - the woven product of a student in Brazil might be shown on line to a C&G assessor in Singapore, who can ask for the knot four in on the nineteenth row down to be shown on screen, and then for the camera to show the reverse side. Just incredible.

E-assessment must lie at the heart of future developments in both vocational and general education, not least because it offers opportunities to lighten the assessment load.

The innovative work the QCA is undertaking on on-screen, on-demand testing in key and basic skills and in ICT testing at key stage 3 is therefore very important.

We have been asked by the Secretary of State to facilitate and communicate a plan for e-assessment over the next 2-5 years, including greater use of computer-delivered assessments in some of the most popular GCSE qualifications.

We will shortly invite a range of assessment providers and key stakeholders to a launch event on 20 April. We are particularly interested in giving this a European dimension, in involving cutting-edge providers such as City and Guilds, and including our fellow regulators, most notably in Northern Ireland, in which great advances are being made.

The fifth matter which I need to touch upon is recording and grading achievement.

The interim report states that '...there has been a clear demand for grading of individual components within the diploma.' It goes on to say that the case for grading the overall diploma is finely balanced.

The number of university applicants with three or more A grades is increasing. In 2002, the number of students achieving three or more A grades was around 20,000, representing about 2.5 per cent of the age cohort.

The problem for the university courses where selection is an issue, might be solved by extending the grade scale to differentiate more finely amongst the highest attaining candidates.

However, I am wary about short term changes to the system that require the development and understanding of new standards. The risks are great.

It would, in my view, be more appropriate to make better use of the evidence already available from the current system, until alternative approaches have been developed and tested.

There is presently much additional information that could be made available to those universities that need it.

For example, the results for each unit could be made available routinely to universities, in addition the overall qualification grade.

Far smaller proportions of candidates achieve a grade A in each unit than in the overall qualification. In OCR English Literature in 2002, for example, 26.5 per cent of candidates were awarded A grades. Only 6.2 per cent obtained the equivalent of a grade A in each of the six units.

In addition to information about unit grades, it would be possible to make information about UMS marks available.

In other words, the problem of differentiating amongst those candidates obtaining grade A can be solved in the short and medium-term through making better use of the information already available.

We want to sort out how and when this can be done with the awarding bodies and UCAS. I think it is a quick win: it might well be possible to make this information available almost immediately.

And of course, the interim report is right on the proposals to develop transcripts of achievement. Work on this should start now, building as the report suggests on initiatives such as the Progress File.

My sixth and final point is about participation and retention.

We currently have a quarter of our 16-19 year olds not in education and training. It is important that we increase that proportion steadily over the next decade, in preparation for the new diploma qualification.

Achieving this is not simply a matter of the availability of curriculum and qualifications.

Equally important are the parallel arrangements that must be made to facilitate the take-up: industry-referenced vocational curriculum; teachers with industry experience; teaching in industrial standard facilities; flexibility to move between school, college and workplace; and the involvement of business and industry in workplace provision.

The recently published OFSTED report on 'Pathways to Parity', which looks at vocational education for 14-19 olds in Denmark, the Netherlands, and New South Wales demonstrates the importance of such arrangements.

As Mike Tomlinson has said, 'retaining the status quo is not an option'. Neither is piecemeal or poorly managed development over the next ten years.

Piecemeal reform is one of the reasons for the large number of qualifications within the national qualifications framework. There are, for example, at present within the NQF:

• 350 vocationally-related qualifications at level 2;
• 386 general qualifications at level 1;
• 508 occupational qualifications at level 3;
• 59 entry level qualifications.

The interim report and the vocational qualifications element of the Skills Strategy provide a unique opportunity to build a more coherent, simpler qualifications system that will meet the needs of learners, employers and higher education more successfully.

Careful redesign and rationalisation of these qualifications incrementally but with determination over the next ten years, with the needs of adults and 14-19 year olds firmly in mind, will enable genuine lifelong learning pathways from 14-90 to be established.

And that takes us into the territory of the reform of vocational qualifications, which is important territory but not the ground I have been asked to cover today.

So, what is the timeline? What are the milestones for the next ten years?

That is essentially a matter for the owner and the architect, but as the builder I can advise them on what is possible and what is not.

Provided the awarding bodies, the sub-contractors, can deliver, the number of assessment units in A level could be reduced from September 2006, although 2007 would be a safer target.

Designing coursework requirements and tasks that span more than one subject can be taken forward within our GCSE pilot programme. New cross-subject arrangements could be in place, in a limited number of subjects, from September 2006.

On new forms of assessment within qualifications:

• work on increasing the use of teacher-led assessment within qualifications is already underway in some of our GCSE pilots, and will continue;
• development of computer-delivered assessment is already underway in key and basic skills testing and in ICT tests at key stage 3, and could be extended to a number of GCSE subjects from September 2006.

On providing better differentiation amongst the highest attaining advanced level students:

• the provision of more information to higher education about unit grades and marks should be possible for the summer 2005 examinations series, provided the UCAS and awarding body data systems can support it;
• extending grade scales and integrating AEA-type questions into A level is complex, and our advice to the architect would be that this is not an easy quick win.

On integrating 'key skills' into English, mathematics and ICT qualifications, work could start immediately in relation to new GCSEs in English, mathematics and ICT, with new qualifications available for teaching from September 2006 - clearly the recommendations from Adrian Smith's review of mathematics provision will influence developments in this area.

Finally, moving towards a unit-based credit-rated set of 14-19 qualifications, which I believe to be pivotal to the success of the proposed reforms:

• pilots of new unit-based, credit-rated general and vocational qualifications in pioneer subjects and sectors could start in 2005;
• the first awards of such qualifications could take place in 2007;
• the principles could be extended to all vocational sectors and GCSE subjects by 2010 and to all advanced level qualifications by 2012.

That would be consistent with the parallel development of the new National Qualifications Framework, as part of the Skills Strategy.

Thank you.

Ken Boston