A cross-country analysis of the results of the OECD’s 2009 Programme for International Student Assessment (PISA) shows that school systems which group students by ability tend to have lower overall average results and larger achievement gaps between rich and poor than those that provide for all students regardless on their socio-economic backgrounds.
School systems which have selective schools, stream students by ability, make students repeat grades and transfer weak or disruptive students tend to perform below average and have more socially unequal results. In contrast, school systems that perform above-average and have below-average socio-economic inequalities tend to be those in which socio-economically advantaged and disadvantaged students attend the same schools.
The 13 OECD countries whose school systems have low levels of student selection and grouping had an average student performance of 505 score points compared to the average for all OECD countries of 403 score points. None of the school systems participating in PISA which had above-average performance and below-average socio-economic inequalities had high levels of student differentiation between schools.
On average across OECD countries, 2.6% of the variation in student performance is solely related to schools’ policies on selecting and grouping students and 9.6% of the variation is jointly related to both schools’ policies on selecting and grouping students and on socio-economic and demographic background.
The variation in student performance among schools in Australia explained by student selection and grouping is well below the OECD average. The variation in student performance solely related to schools’ policies on selecting and grouping students is negligible and 3.3% of the variation is jointly related to both schools’ policies on selecting and grouping students and on socio-economic and demographic background. Selection and grouping of students explains a negligible proportion of the variation in student performance in other countries such as Finland, Iceland and Norway.
Grade repetition
In countries, and in schools within countries, where more students repeat grades, overall results tend to be worse and socio-economic differences in performance tend to be wider, suggesting that people from lower socio-economic groups are more likely to be negatively affected by grade repetition.
This finding is consistent with the literature on grade repetition which shows that it is students from socio-economically disadvantaged backgrounds who appear to be hurt most by grade repetition.
Although the objective of grade repetition is for these students to have more opportunities to learn, underperforming students do not seem to benefit from repeating a grade. One hypothesis to explain this negative relationship is that having the option to have low-performing students repeat a grade places fewer demands on teachers and schools to help struggling and disadvantaged students improve their performance.
Streaming students
The OECD report also shows that school systems which stream students by ability into different education programmes have larger achievement gaps but do not achieve higher levels of average performance than systems that stream students later in their school careers. The younger the age at which streaming first occurs, the greater the achievement gaps in student performance by age 15.
This finding is consistent with prior research showing that inequality is greater in more differentiated school systems on balance. Early selection into different institutional tracks is associated with larger socio-economic inequalities in learning opportunities without being associated with better overall performance.
Student transfers
In school systems where it is more common to transfer weak or disruptive students out of a school, performance and equity both tend to be lower as well. The more frequently schools transfer students to another school because of students’ low academic achievement, behavioural problems or special learning needs the lower the school system’s performance in PISA. The report shows that over one-third of the variation in student performance across countries can be explained by the rate at which schools transfer students.
One hypothesis is that school transfers may hurt student achievement because transferring out of a school is usually difficult for students. Another hypothesis is that in systems where transferring students is a common policy or practice, teachers and the school community have less of an incentive to commit themselves to helping lower-achieving students improve.
School systems that transfer students to other schools more frequently also tend to show a stronger relationship between students’ socio-economic background and performance. This suggests that transferring students tends to be associated with socio-economic segregation in school systems, where students from socio-economically advantaged backgrounds end up in higher-performing schools while students from disadvantaged backgrounds end up in lower-performing schools.
Within schools, however, transferring students leads to more homogeneous student populations and the effect of students’ socio-economic background on performance is mitigated. Not surprisingly, therefore, in these types of systems, PISA shows levels of socio-economic inequity between schools to be relatively larger than levels of inequity within schools.
The report also found that individual schools which make more use of transfers also perform worse in some countries.
Selective schools
Highly selective schools which use students’ academic records or recommendations from feeder schools to decide who will be admitted tend to perform better than non-selective schools in many countries. However, the prevalence of selectivity in the education system does not relate to the system’s overall performance level. Instead, education systems that contain a large proportion of selective schools tend to have greater variation in performance between schools.
Trevor Cobbold
PISA 2009 Results: What Makes a School Successful? Resources, Policies and Practices, OECD.