England’s performance in the Programme for International Student Assessment (Pisa) league tables could be boosted if pupils were encouraged to study certain GCSE subjects, new research has suggested.

A study by Cambridge Assessment found that “the popularity of subjects…could have some influence on England’s performance in Pisa”.

Cambridge Assessment looked at the correlation between grades in individual GCSE subjects with Pisa ability estimates.

The analysis found that some subjects had a surprisingly weak correlation with Pisa results. “Our correlation analysis shows that performance on the Pisa reading test is at least as closely aligned with achievement in GCSE science as it is with GCSE English,” the researchers note.

However, certain GCSE subjects seemed to be associated with a higher Pisa score. “The strongest of such relationships was found with whether students studied three separate science GCSEs or whether they studied the combined science GCSEs (core and additional science),” the study says.

“Before accounting for any pupil characteristics, the mean Pisa score of those studying separate sciences was around 70 Pisa points ahead of those studying combined science. After accounting for differences in prior attainment at key stage 2, the difference was around 35 points.”

The study found that even after accounting for differences in achievement at GCSE, “those studying separate sciences were around 20 Pisa points [in reading, maths and science] ahead of those studying combined science” - equivalent to about six months’ progress.

While it is possible that this might be explained by “unmeasured aptitudes” of pupils choosing to study separate sciences, the researchers note that if the differences are interpreted “in a purely causal manner they would suggest that, without even needing to improve GCSE results, if the two-thirds of pupils studying combined science moved to studying separate sciences, the performance of the UK in Pisa could rise to being close to that of Korea”.

The researchers conclude that “an individual’s performance in Pisa is likely to depend upon their choices with regard to what they study”.

“As such, the popularity of subjects at a national level could have some influence on England’s performance in Pisa.”

They say that these findings “warn against uncritical use of the Pisa results without careful consideration of exactly what is being measured”.