Skip to content

Presentation

The comments included in termly school reports enable teachers to provide personalised feedback on each pupil’s progress. But are they always a strict evaluation of pupils’ work and behaviour? A statistical analysis of school reports issued to more than 600,000 sixth-form students in the science stream between 2013 and 2017 reveals, on the contrary, differences in the comments depending on the pupil’s gender.

These differences stem primarily from the vocabulary used by teachers to describe the behaviour of girls and boys. However, in science subjects – and particularly in those leading to degree programmes where men are over-represented – these differences can also be explained by the use of distinct terms to describe the skills of girls and boys at the same academic level. On the one hand, the gendered vocabulary associated with behaviour reflects assessments relating to the perceived immaturity of boys and aims to encourage girls for their efforts and diligence. On the other hand, the gendered vocabulary associated with skills reflects assessments of girls’ shortcomings on the one hand, and boys’ aptitudes on the other.

The influence of teachers who make more extensive use of gendered vocabulary tends to be limited in the short term, both in terms of performance in the baccalaureate and pupils’ post-baccalaureate educational choices.

Key Results

  • An analysis of more than 600,000 school reports for Year 13 science students (2013–2017) shows that, at a comparable academic level, teachers do not use the same vocabulary to describe the work of girls and boys.
  • The use of gendered vocabulary is more pronounced in science subjects leading to degree courses where men are over-represented (mathematics, physics and chemistry), and less pronounced in humanities subjects (philosophy, modern languages) or in science disciplines with a predominantly female student body (life and earth sciences).
  • In mathematics, for example, the vocabulary used in school assessments can predict the gender of 6.5 out of 10 pupils on average, whereas gender-neutral vocabulary would predict the gender of 5 out of 10 pupils (random prediction).
  • In this subject, three-quarters of gendered terms relate to behaviour: girls are praised for their effort and diligence, whilst boys’ immaturity is highlighted. The remaining quarter describes skills: at the same academic level, teachers place greater emphasis on girls’ shortcomings and boys’ abilities, in line with gender stereotypes in science.
  • Conversely, in subjects with a predominantly female student body, terms relating to skills are used without distinction between girls and boys.
  • Teachers who use more gendered vocabulary slightly improve performance in mathematics in the baccalaureate (a gain of 0.8 ranks out of 100 for girls, and 0.4 ranks out of 100 for boys) but do not influence trajectories in higher education.

Method and Data

The analyses presented in this study are based on several sources provided by the Directorate for Évaluation, Forward Planning and Performance at the Ministry of National Education (MEN-DEPP) and by the Sub-Directorate for Information Systems and Studies of the Ministry of Higher Education and Research (MESR-SIES). These include: 1) anonymised data extracts from the ‘Admission Post-Bac’ platform (2013 to 2017 sessions), which contain all school reports for final-year secondary school pupils and their higher education course choices; 2) the OCEAN-DNB database (results of the national brevet examination, 2010 to 2022 sessions) and the OCEAN-BAC database (baccalaureate results, 2013 to 2022 sessions); 3) the SYSCA and SISE databases (2013 to 2023), which enable the tracking of pupils’ educational pathways through upper secondary school and into higher education.

Partners

Agence nationale de la recherche (ANR)Chaire PEMS

The authors would like to thank the French National Research Agency (ANR), through project ANR-17-CE28-0001-01, and the Women and Science Chair at PSL-Paris Dauphine University for their financial support.

This paper has also benefited from the support of the Chair in Educational Policy and Social Mobility. Established in 2021 as part of a partnership between the Ardian Foundation, the Directorate for Evaluation, Foresight and Performance of the Ministry of National Education (MEN-DEPP) and PSE-Paris School of Economics, this chair aims to promote high-level research and the dissemination of knowledge on educational policies and social mobility.

Last modified: July 21, 2026