Lies, damned lies and statistics?

07 November 2013

Statistics can be your best friend, and your worst enemy. At present they are being used quite freely in the press in an attempt to influence the way people think about the relative value of studying different subjects at GCSE. It is suggested that some subjects have more value than others, that some are “hard” and some are “soft”.

A core element in the government’s case for the English Baccalaureate is the argument that certain subjects are more helpful than others for getting into university. This is based on the fact that the Russell Group of 24 leading universities publishes a list of “facilitating” subjects at A Level, which, they say: “opens doors to more degrees and more professions than others.” These are all EBacc subjects, but the Russell Group also says that while hundreds of courses require one of more of these facilitating subjects, many students will study a combination of these and “non-facilitating” subjects. Their publication, Informed Choices, lists the subjects that are required more often than others, and points out that there are some degrees that are open without any specific subject background: “it is not about “hard” or “soft” subjects, but the right ones.”

This leads us to ask what other statements and statistics are being used to try to justify changes to our education system, and whether they are being presented in their right context. Two examples of the use of statistics this month are:

  • Liz Truss, Under-Secretary of State for Education and Children: “Maths, for example, is the only school subject which has been proven to add to earnings, by up to 10% at A Level, when every other factor is taken into account.”
  • BBC News: “Some 82.4% of social science graduates were employed three years after graduating, compared with 79% of arts and humanities graduates and 78% of graduates with science degrees, the figures suggest.”

As with the idea of “facilitating subjects”, statistics like these are used to suggest which subjects children should be encouraged to study at school. We think these arguments are misleading, because they encourage us to look at education in a very narrow and instrumental way. The important point is that we should be thinking about more than percentages and specialisms. A discussion that places one subject above another benefits nobody. What children need is a combination of subjects that will equip them for the world, but which will also make the most of their interests and talents.

That means a curriculum that is both broad and balanced, that develops them creatively and intellectually, as well as giving them the necessary skills in numeracy and literacy. Reports from the Confederation of British Industry, the Young Foundation and the European Commission all suggest that without the understanding, skills and knowledge that come from engaging in many different disciplines, our children will fail in the coming world. Creative, innovative thinkers who have been encouraged to explore and implement their ideas, however, will succeed – and the arts are as important in nurturing creative thinking as science. The CEO of Google, Eric Schmidt has said’ “Over the past century, the UK has stopped nurturing its polymaths. You need to bring art and science back together.” The EBacc is helping to drive them apart.

So do the statistics quoted above stand up to scrutiny? At the CLA we love statistics, because it is essential to know the evidence base for making decisions. All the statistics the CLA collates and passes on to you are carefully tested and checked. If there is a statistic that makes an impact but that we think is misleading, we will not pass it on to you.

So we have stress-tested the two statements quoted earlier. The source for Liz Truss’s claim that pupils who study Maths A Level earn 10% more than peers is pretty obscure, but it seems to come from a 1999 piece of research from the University of Newcastle upon Tyne. Anna Vignoles of the Centre for Economic Performance and Peter Dolton of the University of Newcastle upon Tyne based the research on a 6,000 strong sample from the National Child Development Study (NCDS) and from a postal survey of graduates. They looked at the earnings at the age of 33 of the NCDS sample, while the postal survey looked at the earnings six years on of people who had graduated in 1980. It looks like a sound piece of research, but we would be interested to see it updated, as the sample is from people who took their Maths A Level in the 1970s, and the employment market is significantly different today.

The BBC piece reporting that Social Science Graduates had higher employment rates confirms a trend. Social studies graduates have a higher employment rate than Maths graduates three and half years after graduation – 81.8% compared to 75.4%. The employment figure for arts and design graduates, is also above Maths. But then when you look at the figures, you find that the low employment rate for Maths graduates is because 17% of them are going on for further study.

A better statistic is to look at is the variation in employment prospects according to what kind of degree has been taken. As the Office of National Statistics points out, having a degree means you are much less likely to be unemployed than if you had no qualifications: “by the age of 24, 13% of those with only GCSEs were unemployed, compared with 7% of those who had left education with A-Levels and just 5% of those with a degree.”

So any degree increases the chances of being employed – but what is interesting is that differences in employment chances between subjects are small. A report by the Higher Education Statistical Authority shows unemployment rates of graduates of graduates ranging for 1.2% for Medicine and Dentistry to 5.4% for Computer Science. Within this range other examples are:

  • Maths 2.3%
  • Science subject areas 2.9%
  • Social studies 3.0%
  • Creative arts and design 5.2%

(In other areas of statistics a difference of 3% in unemployment rates would not be regarded as statistically significant.)

So we can safely say that employment prospects and future earnings vary very slightly between which degree is taken. The important message is that we should want every child to get an education and qualifications, whatever subject they choose. A broad and balanced curriculum is essential to enable people to make the choices about what suits them best. The better their motivation, the better their results, the better their results, the better their life-chances.

Driving children down a narrow path, and cutting them off from the opportunities for creativity provided by the arts, is not going to provide society with the complete citizens that it needs. But it will create the wrong kind of statistics: for educational failure, unfulfilled talent, and cultural decline.