Education CommitteeWritten evidence submitted by The Royal Statistical Society

Introduction

1. The Royal Statistical Society (RSS) is a professional body and learned society devoted to the interests of statistics and statisticians.

2. Within its broader remit of promoting the public understanding of statistics and the competent use and interpretation of statistics, the Society works to ensure that the discipline of statistics and the use of data are competently taught, learned and assessed at school, further education and higher education levels. The Society also promotes statistics and the appropriate use of data as a core competency in the workplace and through lifelong learning.

3. The Society believes that, for modern life, it is essential that all learners should be acquainted with the basic principles of probability, risk and problem-solving using statistics. We take the view that statistical skills are essential tools across the curriculum and for the sciences (including the social sciences), and they are a vital life skill.

4. The Society itself has over 60 years of experience in curriculum development, setting and awarding internationally recognised statistics qualifications at a range of levels (the Ordinary Certificate which is approximately between GSCE and A level; the Higher Certificate at approximately the level of the first two years of UK university statistics and the Graduate Diploma which is a full degree-equivalent in statistics).

5. Although statistics pervades many school subjects, the Society notes that statistics is largely taught, learned and assessed within the mathematics curriculum. Hence many of the points below refer to the mathematics curriculum and its development.

6. The Society has considered its responses to the questions in the consultation invitation after seeking views of members who are educators and education research specialists, including those who have experience of the assessment process at both GCSE and A level.

7. The Society is unanimous in its view that curriculum development, delivered by qualified subject specialists is crucial and that no future awarding body framework or specification structure should put curriculum development at risk.

1. What are the arguments in favour of and against having a range of awarding bodies for academic and applied qualifications (including A’levels, GCSEs. Diplomas, BTECs and OCR Nationals), and the merits of alternative arrangements, such as having one national body or examination board franchised to offer qualifications in particular subjects or fields?

8. Noting that the “best and fairest” educational outcomes for young people are the target, we believe the best way to help to achieve these outcomes is through a system which strongly incentivises high-quality curriculum development. Curriculum research and improvement are vital to help achieve this. Overall, the matter of how this is being delivered is not something the Society takes a strong view on; indeed we believe this key outcome could be delivered through different types of systems.

9. The Society is an awarding body and uses its own members and others brought in because of their experience and expertise to develop and assess its qualifications. This is an example of subject specialists successfully contributing to the development of a curriculum within a single awarding body.

10. Other countries successfully develop innovative curricula in a context with single awarding body. A good example in the area of statistics is New Zealand which has developed one of the most promising mathematics and statistics curricula in the world.

11. Clearly, if a single awarding body should be established there would need to be an understanding that its remit includes curriculum research and development by subject specialists. In addition, to ensure there is room for input from key experts in the different subject areas such a system would need to have due checks and balances in place to avoid government-driven control of the curriculum. The Scottish model is one that may be worth pursuing, and it would be essential that a single awarding body retains suitable independence from both Government and commercial interests.

12. A situation with competing awarding bodies, such as the current system in England, has created an environment which stimulates the development of innovative and forward thinking approaches. For example, the range of awarding bodies currently in place means they are prepared to establish project syllabuses (such as Nuffield Physics and SMP and MEI Mathematics) and to develop curricula in new and different ways (such as the pilot schemes developed in mathematics recently). Such work has delivered a healthy national suite of qualifications, in a system with multiple awarding bodies, through competition-driven innovation in curriculum research and development.

13. If the recommendations of the recent Mathematics Taskforce report1 are implemented, a considerable amount of curriculum development work will be needed in the coming years. The Society concurs with the Taskforce’s view that this should be led from within the mathematics community rather than by a government agency.

2. How to ensure accuracy in setting papers, marking scripts, and awarding grades?

14. It is the Society’s view that current problems with setting and marking examinations and fixing thresholds will not be solved by reducing the number of awarding bodies. Other solutions need to be sought. An extensive discussion of these difficult issues can be found in a recent report from QCA.2

15. Current problems with accuracy in setting papers may have come about partly as a result of the wish of the awarding bodies to streamline their operations and so improve efficiency (ie not educationally driven); this has resulted in some cases in dispensing with what has subsequently turned out to be essential proof reading checks. In our view current mechanisms to ensure the accuracy of papers are not as effective as they could be.

16. Marking of examination papers in mathematics is moving towards on-line marking. The Society recommends a review of this process as there is little evidence currently that this development has improved accuracy (or validity) in marking scripts.

17. Currently all GCSE Mathematics specifications are required to conform to the pattern of two tiers only. Additionally, there has to be an overlap of C grade in the two tiers so that all candidates are being assessed in a tier where a C grade is possible. This has three negative effects:

(i)there are a significant number (well over a quarter) of candidates for whom a C grade is likely to be always beyond reach. Forcing such candidates to sit a paper in which some if not most of the paper is inaccessible cannot be a positive experience;

(ii)there are, likewise, a significant number of candidates who will obtain A or A* grades (about one sixth). For them being made to sit papers in which a proportion of the work is possibly trivial is also not helpful to the process of differentiating ability; and

(iii)the result of having to offer five grades at each tier means that a potential C grade candidate cannot access a large proportion of the papers. For example, currently, at the higher tier (with grades A*, A, B, C and D), 50% of the marks have to be available to grade C candidates, but this causes a reduction in the amount of differentiation for the most able within a tier. This can only be resolved by allowing Awarding Bodies to develop specifications which are not constrained by the need for all subjects to conform to the same rules.

18. There are other issues which the Society would like to draw to the attention of the consultation committee, loosely connected to the question. Mathematics, and within this, statistics, forms part of the National Curriculum and is taken by all pupils up to age 16. The unique nature of the subject needs to be recognised and understood.

19. A particular concern of the Society is the rule that all units can have only one mode of assessment. This prevents work in which students collect and analyse their own data from forming part of the assessment of a statistics unit; in practice this means that such work, which lies at the heart of statistics, is rarely done. It must be made clear to curriculum developers and to regulators that not all subjects can be processed through the same filter and/or be subject to all the same rules of examination and assessment. OFQUAL (and its predecessor bodies) has closed down valuable qualifications because they do not fit the regulatory rules of the day. OFQUAL should not be seeking to process subjects through a “one size fits all” filter when it comes to accrediting qualifications and ensuring that they represent reliable and consistent levels of attainment. This approach by OFQUAL has already:

Discounted new and relevant qualifications such as the planned GCSE Use of Mathematics (aimed at post-16 students) which OFQUAL did not approve as a GCSE in 2010 as it did not match statutory regulations.

stopped a proven-to-be-successful model of assessing GCSE mathematics via a combination of examination and coursework as part of learning. Similarly, OFQUAL has indicated that at the next revision of AS and A2 Mathematics specifications any unit (or module) may have only one mode of assessment; thus a Mathematics/Statistics unit may be 100% examination or 100% coursework (or controlled assessment) but it may not contain a mixture. Similarly, successful units containing significant amounts of numerical analysis that have used computers are to be stopped. It is a concern that there is now almost no coursework option offered in any GCSE, AS or A2 unit in Statistics. Neither is there the possibility of using technology and the rich software it offers in assessment regimes for mathematics or statistics. This means that young people are already going on to further or higher education or work with no experience of, for example, handling real data in context, solving problems which are relevant to themselves or to employers.

3. Thoughts about the commercial activities of awarding bodies, including examination fees and textbooks and their impact on schools and pupils?

20. In our view the commercial impact of Awarding Bodies should be examined. In the interest of developing high-quality textbooks some of the current practices, such as using examination board branding or examiner’s names on textbooks should be reviewed. It is clear this practice gives these textbooks a commercial advantage over other textbooks, which may be of an equal or better quality but do not have the branding power of an examination board behind them. It is clear there are many talented textbook authors who deliver high quality teaching materials and their innovative contributions should be encouraged in an open market.

November 2011