Pygmalion is a play about the power of expectation. George Bernard Shaw’s linguistics professor Henry Higgins transforms flower-seller Eliza Doolittle’s prospects and status simply through the power of high expectation. Eliza works hard for sure but seems always to know she will succeed.

In 1968, another professor, Harvard academic Robert Rosenthal, teamed up with junior school head Leonore Jacobson to explore the extent they could develop “Pygmalion in the classroom”. They gave an intelligence test to a group of young children and then told a randomly selected 20 per cent their test results indicated unusually high potential and that they would “bloom” during the year. Remarkably, they did. Other studies since have confirmed the importance of high expectations in raising performance and also the self-fulfilling nature of low expectations.

This year’s A-level and GCSE chaos may unwittingly provide us with an opportunity to test “Pygmalion in the exam room”. We know that the grade inflation resulting from the decision not to moderate down teacher assessment has left about 1 in 4 students with an A level grade A that they would not have achieved had the exams gone ahead. About 1 in 10 GCSE maths and English students also have a good pass in those subjects that they wouldn’t have achieved had they sat the actual exam. Of course, we do not know who is the one in four, or one in ten, but that matters little in an experiment given the large population of students.

Tes FE podcast: The exams fiasco, with UCU’s Jo Grady

Opinion: Be warned, Mr Williamson - students won’t forget this

Background: A-level unconditional offers raising uni drop-out rates

Grade inflation

In the days following the government U-turn, there were some dire predictions that this grade inflation would come out in the wash. Many would struggle if they progressed to A levels or to high-tariff universities on the strength of their good fortune. But is that likely?

Surely the logic of Pygmalion will mean these students make progress in the way we would expect of people with those grades? It would be worth us constructing now a way of checking the progress and achievement of this cohort over the next five years to see if that is the outcome. It might be particularly interesting to track those students whose grades were moderated sharply upwards by the algorithm as these have benefited significantly.

The conjuring trick uncovered by Rosenthal and Jacobson does have to be believable. Most A-level students initially moderated downwards (nearly 40 per cent) were affronted, suggesting they believe their assessed grades were accurate. Credibility meant only 20 per cent of children in the original experiment were told their results were exceptional. So even if grade-inflated students do as well as their accurately-graded peers we could not simply give great grades to all in future. But maybe we could consider a range of Pygmalion-in-the-college experiments. 

Colleges might prove fertile ground for such work. Every year, I meet thousands of new students who lack confidence and self-belief. When I show these new students examples of young people who have won world skills competitions or international arts prizes and point out they were no different to them when they started with us, they begin to sit up straighter. What if we told a random 20 per cent that their maths diagnostic test indicated they have unusually good understanding and should improve significantly when they resit the exam, or tell another random 20 per cent their early work indicates they will achieve mastery of their skill and command high pay on completion?

Of course, the student still has to put in the hard yards. This is about motivation and belief, not a magical substitute for effort. Perhaps most importantly, the teachers also have to be convinced that the end result is real and not part of the trick. Eliza reserves her most damning criticism for her professor: “The difference between a lady and a flower girl is not how she behaves, but how she’s treated. I shall always be a flower girl to Professor Higgins, because he always treats me as a flower girl, and always will.” Thankfully, teacher assessment showed that our staff generally thought more highly of our students than might an examiner, suggesting we are content to play catalyst and believe the end result.

Unlike Higgins, college staff already have a great track record when it comes to treating students well.  They often see the future, well-adjusted, highly skilled individual where a school saw the same person as a disruptive trouble maker, but I’m not sure we’ve tried using random praise as a means to higher performance. What have we got to lose?

Ian Pryce is principal and CEO of the Bedford College Group