SPI does provide valuable comparative data but there will be varying factors that will have an influence on the SPI data depending on the membership of class/groups in each school i.e. in Science, if a school doesn't offer Separate Science, their whole cohort may well feature in Combined Science. There will more likely than not be students with higher prior attainment competing for the Combined Science grades than there would be if there was a Separate Science option too. Equally, in a non-selective state school, the students are competing for the same grades as those in a selective school. Prior attainment will likely subdue the range of grades towards the top end achieved due to the way grades are assigned in the summer which will in turn subdue the amount of progress or value added that will be represented in this measure. Similar issue in MFL (and other subjects that not all students take) is possible - a school with a group of 20 linguists may be more likely to secure a higher SPI/progress compared to a school which enters its whole cohort - the range of prior attainment will be much broader and the motivation may well be different.