England’s score in Year 5 maths has stopped improving for the first time, a major international comparison shows, but the science score has increased significantly.

Year 5 pupils scored 552 in maths in the Trends in International Mathematics and Science Study 2023 (Timss). This was a fall of four points from 2019, and the first time the Year 5 maths score has not increased between cycles.

However, the authors of the international report said the decrease was not statistically significant.

Year 9 students’ maths scores increased from 515 in 2019 to 525 in 2023, though this increase was also not described as being significant.

A Department for Education report responding to the latest Timms findings says that the lack of any significant falls in England’s scores this year reflects “schools’ strong commitment to recovery” after the Covid-19 pandemic.

England did see significant increases in science scores for both Year 5 and Year 9 pupils compared with 2019. Year 5 scores increased from 537 to 556 and Year 9 scores from 517 to 531.

This follows 2019‘s Year 9 science score, which had been a significant drop compared with previous years. The 2023 score published for England today remains below scores achieved prior to 2019.

The scores mean England ranks ninth in primary maths for Timss 2023, sixth in secondary maths and fifth in both Year 5 and Year 9 science. England rose up the rankings from 2019 in the latter three, but several countries that had been ahead did not participate in 2023.

England remains behind several East Asian participants in maths - including Singapore, Chinese Taipei, Korea, Hong Kong and Japan. England is ahead of Hong Kong in secondary science, and Japan and Hong Kong in primary science.

• More on Timss 2023: How England compares with the rest of the world on attendance
• Pisa 2022: UK’s Pisa scores fall in maths, science and reading
• Explainer: What is Timss and why do we take part?

Run by the International Association for the Evaluation of Educational Achievement, Timss measures knowledge of understanding of curriculum content. This sets it apart from the Programme for International Student Assessment, which measures applying education to real-life problems.

The two cohorts measured in this Timss cycle were both severely disrupted by the pandemic and by school closures. In summer 2020, Year 5 pupils were in Year 2 and Year 9 students were in Year 5.

The DfE’s national report into England’s scores says that there has been an increased range of performance seen across both maths and science, driven by the highest-performing pupils increasing their scores.

The report authors said more work must be done to establish the groups of pupils who have seen less resilience in their learning during the pandemic disruption.

The report adds that stagnation in the lower end of performance for secondary pupils is “likely to be disproportionately borne by disadvantaged pupils”.

Strengths and weaknesses

Timss also measures specific domains in maths and science. Year 5 pupils were weakest in the measurement and geometry domain for maths, which was a strength this year for the majority of high-performing countries.

The DfE’s report says that England’s Year 5 pupils were strongest in data. Year 9 students were stronger in “data and probability and in number” and were weaker in algebra and geometry.

Both Year 5 and Year 9 pupils scored higher than their average maths scores in a domain that measures “knowing” of maths concepts, and Year 9 also scored higher in applying those concepts. Year 9 students scored significantly lower than average in maths reasoning.

Making recommendations on policy, the authors of England’s national report suggested continued attention should be given to weak performance in geometry, measurement and algebra. The authors also said developing cognitive skills would support both maths and science development.

Authors said the maths mastery approach promoted since 2014 has been accompanied by increases in Timss maths performance, but increases in performance over time may also be driven by teacher professional development and investment.

Attendance compares well with other nations

The study also measured socioeconomic status from headteacher reports and absence levels.

Socioeconomic data was not available for all pupils, but score gaps between the more affluent pupils and more disadvantaged pupils generally saw small decreases from 2019.

England has seen an attendance crisis since the pandemic, however, the Timss data shows that it actually compares well with many other nations.

Timss 2023 saw a slightly lower percentage of Year 9 students eligible for SEN support or free school meals in the sample than in the past six years.

The report states this could lead to an over-estimation in performance, but added that it was likely to be balanced out by private schools being under-represented in the Year 9 sample.

Timss 2023 involved 131 primary schools in England and 136 secondaries with 4,091 Year 5 pupils and 4,239 Year 9 students. The tests were taken between March and June 2023.

A total of 66 jurisdictions took part.

Scotland announced it would leave Timss in 2010 but said last year it would rejoin the study. The next cycle of Timss will take place in 2027.

School standards minister Catherine McKinnell said tax-free incentives to attract teachers to subjects like maths, physics, chemistry and computing will help drive up standards in these subjects.

For the latest education news and analysis delivered every weekday morning, sign up for the Tes Daily newsletter