PISA 2025: Science achievement and experiences of 15-year-olds Publications
Publication Details
This report covers the science achievement and learning experiences of 15-year-old students in New Zealand English-medium schools in 2025. Each cycle of PISA focuses on 1 learning area and in PISA 2025 the focus was on science, providing in-depth achievement and contextual data.
Author(s): Rachel Borthwick, Ministry of Education
Date Published: September 2026
Key findings
Achievement
New Zealand students continue to demonstrate strong performance in science over time, even as the OECD average declines. New Zealand is one of the higher-performing countries in science with a mean score of 509, well above the OECD average of 482. Only 7 systems had higher science achievement: B-S-J-Z (China), Singapore, Macao (China), Chinese Taipei, Japan, Estonia, and Korea. (B-S-J-Z (China) refers to the Chinese provinces of Beijing, Shanghai, Jiangsu, and Zhejiang.)
New Zealand’s science score was similar to that of the United Kingdom (511), Canada (510), Australia (509), Finland (504), and the United States (502).
New Zealand continues to have inequities in achievement for students in different socioeconomic groups and in different ethnic groupings. We also have a large score gap between the highest and lowest achievers (284). Girls and boys, however, have similar achievement in science over time.
Attitudes and beliefs
New Zealand students were more likely than the OECD average to report being curious and enjoying science. However, they reported lower perseverance across a range of areas than their peers across the OECD and were less likely to express a belief they could adapt to new or challenging situations.
They were more likely to disagree that intelligence was fixed and also more likely to report seeking help with their learning.
Between 2022 and 2025, New Zealand students’ reported belief in the value of school increased slightly across the 3 measures. In 2022, 72% of students disagreed that school was a waste of time; in 2025, 77% of students disagreed with this statement.
Resources for learning
For digital resources, New Zealand principals were more likely than their international counterparts to report having policies in place about the use of digital devices in schools, though socioeconomically disadvantaged schools had an overall lower level of preparedness for digital learning than advantaged schools.
Although New Zealand was above the OECD average for the availability of most digital resources, there was an increase between 2022 and 2025 in the proportion of students in New Zealand schools where the principal reported that a shortage or inadequacy of digital resources hindered the school’s capacity to provide instruction.
New Zealand students averaged 4.4 hours of class time in science per week. New Zealand students had more time on digital devices for learning than the OECD average. Where students reported more leisure time on devices at school, their achievement was lower. New Zealand also had one of the higher rates – 38% – of reports that students get distracted when using digital resources in science lessons.
New Zealand students reported using Artificial Intelligence (AI) for learning at a similar rate to the OECD average, although more New Zealand students (42%) had never or almost never used AI to draft text for their assignments, compared to 32% across the OECD. Students who had never used AI for learning had significantly higher average science scores than those who had used it at all.
Learning environment
A large majority of New Zealand students reported feeling safe while at school: over 90% report feeling safe in their classrooms and on their way to and from school, and 80% agreed when asked if they felt safe at school overall. However, New Zealand students were more likely to report classroom disruption compared to the OECD average, and were among the least likely to never or almost never experience bullying (39%, about the same as Germany, Estonia, and Denmark). About 15% of New Zealand students reported experiencing racism at their school at least once a month.
Compared to 2022, there was an overall increase in bullying exposure in 2025 (as measured by an index combining students’ responses to multiple items), but rates were lower than in 2018.
New Zealand students were less likely to report missing school or class in 2025 than in 2022, but were more likely to skip an entire school day than the OECD average. Students who had missed a class or school day in the fortnight before PISA had lower science scores than those who had not.
New Zealand students tended to have a lower sense of belonging than their peers across the OECD.
Compared to the OECD average, New Zealand students reported higher levels of teacher support in science classes. This had, however, decreased since 2015.
Environmental science
Achievement in environmental science reflected overall science performance: only 7 countries scored significantly higher than New Zealand.
New Zealand students were slightly more likely than the OECD average to believe in their capacity to generate environmental change. They also consistently agreed with statements about collective efficacy at a higher rate than their international peers. They were, however, less likely to take part in any practical environmental activities.
A higher proportion of New Zealand students than the OECD average attended schools where their principal reported opportunities to participate in environmental activities, though students in socioeconomically disadvantaged schools were less likely to have these opportunities.
New Zealand students had somewhat higher awareness of some environmental issues than the OECD average and reported at slightly higher rates that they could explain the following issues well: increasing greenhouse gases, the consequences of clearing forests, and the extinction of plants and animals.
Response rates and bias
New Zealand was 1 of 5 countries asked to complete a non-response bias analysis due to low participation. The analysis for New Zealand found that the participating schools represented the population well across different categories, but that the student sample was slightly biased toward higher achievers. The bias is similar to 2022, and similar, though likely smaller, bias may affect past results. OECD analysis has shown that this occurs broadly: as student participation drops, there is a bias toward higher achievers taking part across the countries and economies in PISA.
Navigation
Contact us
PISA
If you have any questions about PISA (Programme for International Student Assessment) please email the EMA Team Mailbox