A measure of freedom

James Tooley explains how the return of IQ testing could liberate education from credentialism.

It would be disingenuous to say that I wrote an essay mentioning IQ without thinking that it would get some attention. But the strength of the reaction surprised me. My essay was not just about IQ: it also focused on the purposes of assessment. But using the controversial book The Bell Curve (Herrnstein and Murray, 1994) as a springboard into the discussion had the media champing at the bit in the sure expectation that the educational academic community would be angry and hostile. The media were right.

My argument, while accepting that formative assessment is an integral part of education, questioned the four main justifications given for summative assessments such as Standard Assessment Tasks, GCSEs, A levels and degrees. The justifications given may be there to shore up institutional power: but it is through deconstructing these justifications that the power structures can be undermined. One modern justification is in terms of the need for information required in the so-called market, allowing for better and worse schools to be identified. But this raises the vexed question of how we are to show the value added by schools; crucially this justification assumes that summative assessments do measure what is educationally valuable, which is precisely the issue at stake here. A second justification is that summative assessments motivate learning. But many young people are turned off education precisely by the threat of humiliating public examinations. The third justification would be that it is rather nice for individuals to receive a certificate to show they have understood something.

The final justification is that summative assessments provide information for employers. Now, a MORI survey has shown that in general employers are interested in only rather basic skills and knowledge that can be picked up in the first couple of years of schooling and are not the stuff of summative assessments in English literature and mathematics. Moreover, employers say they are more interested in the personal qualities of potential recruits than in their academic qualifications. But are these summative assessments useful ways of measuring these qualities? If someone has a GCSE grade C in mathematics, say, does this mean he or she is very lazy but bright, or of average intelligence and very diligent, or what? And why are we happy to see the educational process corrupted by the needs of employers in this way?

But employers have got to have some way of selecting among potential applicants. It is here that The Bell Curve comes into play. For what Herrnstein and Murray contend is that employers use summative assessments as a proxy for IQ, and, moreover, that they are right in doing so, because IQ is the single best predictor of job performance and productivity. It is better than educational qualifications, or targeted tests of particular job skills, or interview performance, or any other single measure. If Herrnstein and Murray are right, then this paves the way for the liberation of the educational process from the needs of employers. It allows for a disaggregation of all the information muddled together in educational qualifications. If it is intelligence we are interested in then we do not need to sort students with GCSEs, A levels, and so on: IQ tests are far more efficient. And as IQ stabilises, on average, at ten years of age, childhood scores are as good predictors as those obtained after years of schooling. If it is punctuality and diligence, then letters of recommendation are more useful indicators than educational qualifications. And for employment requiring particular specialist knowledge or skills, the intellectual demands of these jobs can be genuinely separated from the phoney and wasteful demands of credentialism.

While my musings on assessment seemed harmless enough, educationists did not support this further twist in my argument. One cannot it seems mention IQ in the academic educational establishment without encountering hostility; to mention IQ as potentially liberatory leads to derision. Why? There are three sets of objections.

The first questions whether IQ tests measure anything useful and whether they are culturally, ethnically or social class biased. IQ tests measure IQ, clearly. But psychometricians have developed them precisely to isolate intelligence understood as cognitive ability. Many educationists now question whether there even is such a thing, preferring fashionable Howard Gardner's "multiple intelligences". But the psychometrician's usage has at least a distinguished vintage, with the Greeks recognising cognitive ability as one of two (Aristotle) or three (Plato) faculties of mind. It is no good arguing, too, that IQ tests are just "paper and pencil" tests, hence useless in capturing what is important about the human condition - many high points of our culture, music, literature and science are "paper and pencil", too; we are "paper and pencil" beings. As for bias, the evidence in The Bell Curve deserves careful examination. The IQ difference between blacks and whites, the authors say, is greater on test items that appear culturally neutral than on those that appear culturally loaded. Moreover, although black IQ scores get higher with higher socioeconomic status, the difference between blacks and whites of the same higher socioeconomic status does not shrink. Also, is it not odd that IQ tests are never accused of being sexist? Men and women have nearly identical mean IQs. But if IQ tests are deliberately constructed so as to put certain groups down, why were women not put down too? Perhaps there is something more objective lurking in these tests that cannot so easily be manipulated to satisfy the whims of the white males who invented them.

Second, it is objected that research around IQ uses at best sloppy, and at worst fraudulent, data. Some sloppy statistics are likely in any scientific pursuit; it is the accusation of fraud that is far more serious. A straw poll recently revealed that most of my academic educational colleagues believe Sir Cyril Burt was exposed as a fraud, particularly in inventing MZ twins to support his claims about IQ and heredity, and that this, above all, brings the whole psychology of individual differences into disrepute. This is very disturbing, and reveals the extent to which the discussion around IQ is closed and ideological. For while Hearnshaw's biography of Burt attempted to substantiate the claims of fraud, Joynson's The Burt Affair, and Fletcher's Science, Ideology and the Media, disputed the claims and raised some anomalies in the accusations themselves. Hearnshaw, for example, was accused of selectively misquoting Burt to back up the case of fraud. Yet the anti-IQ lobby remains oblivious to the debate, preferring to believe the discipline is in disrepute because of the misdeeds of one man.

Finally, there is the fear of the totalitarian implications of IQ testing. Using IQ tests, governments can sort children by ability, labelling them for life. Crucially this fear depends on governments being allowed such power over the educational process. But this seems to reveal a classic case of the neutrality of the technology itself. IQ testing historically was prompted by left-leaning social reformers as a way of overcoming rigid class structures in British society. Burt himself objected to the rigidity of the 11-plus exam on the grounds that it was only on average that children's IQ stabilised at ten, not that any given child had a stable IQ at that age. My own ideas point to the way IQ tests could liberate education from the stultifying sorting needs of employers. It is not IQ tests that should be feared, but the power of government to decide educational opportunities.

Educationists have a knee-jerk reaction to any mention of IQ, and it does not seem to matter that their objections can at least be challenged. But IQ cannot simply be avoided because it does not chime with what we as educationists want to hear. I seek to bring IQ back on to the educational agenda. By pointing to a potentially liberating aspect of IQ testing, the liberation of education from credentialism, I hope to reopen the debate from a novel perspective.

James Tooley is a research fellow, Centre for Social Ethics and Policy, University of Manchester, and will direct the new Education and Training Unit at the Institute of Economic Affairs.