Are girls always good for boys? Short and long term effects of school peers’ gender
Introduction
Social scientists have long pointed out that peers’ influence is a key determinant of student achievement. Recent empirical studies consistently find significant causal effects of school peers’ ability on student own achievement,1 and there is some evidence that these effects vary considerably by student gender (Landaud, Ly, & Maurin (2020)). Nevertheless, until recent years, the effects of peers’ gender has only received little attention by economic researchers. Student gender is a crucial determinant of social interactions at school, especially among 14–15 years old students.2 These interactions may not only affect students’ behaviour and performance at school but they may also shape students’ preferences in a persistent manner. From a theoretical point of view, the effects of being exposed to a higher proportion of female peers at school are unclear. On the one hand, a higher proportion of girls could benefit to all students through a positive effect on the quality of the learning environment, as girls tend to be more conscientious and disciplined than boys on average.3 On the other hand, a higher proportion of opposite-gender peers may provide greater distraction (Coleman (1961); Hill (2015)), reduce cooperation and positive role-model effects among students (Lu & Anderson (2014); Mouganie & Wang (2020)) or lead teachers to change their behaviour and practices, resulting in negative effects on student learning.
This paper builds on administrative data to investigate the effects of the proportion of girls among school peers in middle school on student achievement and long-term educational choices in France. To address causality issues, I take advantage of two key features of the French educational system, namely compulsory schooling up to age 16 and the existence of a catchment area system implying very little school choice in the public education system. Due to this institutional setting, demographic shocks in the gender ratio of cohorts living in a school catchment area generate idiosyncratic variation in the gender ratio of school cohorts. The empirical strategy implemented in this paper therefore exploits variation in the proportion of girls which occur across cohorts of the same middle school.
I first show that within school cross cohort variation in the proportion of girls is unrelated to other changes in peers’ socio-economic and educational backgrounds, as measured by the proportion of students who have been held back a grade. I further show that variation in the proportion of girls is unrelated to changes in class size or teacher quality, as measured by teacher experience or tenure. By contrast, I provide clear evidence that the proportion of girls among school peers not only affects student contemporaneous achievement and school behaviour, but also affect their subsequent educational attainment. Generally speaking, being surrounded by more female peers benefits to female students while it has detrimental effects on male students, both in the short and the long run. A 20 pp increase in the proportion of girls in the classroom in 9th grade (i.e, 5 students) increases girls’ average performance at the end-of-middle school national exam by 2% of a SD, and decreases boys’ one by 1.6% of a SD. It also increases girls’ probability to attend high school (+1.2%) and to graduate from a high school academic track (+3%) (especially from the scientific track (+3.8%)), and decreases their probability to attend a vocational school (-2.5%) and their dropout rate (-4%). By contrast, it decreases boys’ probability to attend high school (-2.2%) and to graduate from a high school academic track (-3.9%), and increases their probability to attend a vocational school (+3.6%).
I then explore some potential mechanisms that may explain such short and long term effects. Building on a behavioural grade school principal give to each 9th grade student to measure civic behaviours and positive school attitudes, I first provide suggestive evidence that a higher proportion of female peers improves girls’ behaviour. I then exploit the availability of both blind and non blind test scores in my dataset to build a measure of teacher biased behaviour in favor of girls, and I show that this measure is strongly related to the proportion of girls in the classroom. This finding suggests that teachers may tend to adapt their practices and behaviours based on the gender composition of the class, resulting in an indirect negative effect of opposite-gender peers.
This paper contributes to the growing literature on the causal effects of school peers’ gender on student achievement and educational choices. Generally speaking, this literature focuses on short-term outcomes and tends to find that a higher proportion of girls benefits to all students. In particular, Hoxby (2000), Whitmore (2005), Lavy & Schlosser (2011), Hu (2015) and Gong, Lu, & Song (2019) find that a higher proportion of girls among school peers increases both girls’ and boys’ contemporaneous test scores and provide suggestive evidence that these effects operate through an improved learning environment. Related work further show that the proportion of girls in middle school increases both girls’ and boys’ probability to chose STEM over language subjects in high school (Schøne, Simson, & Strøm (2019)) and reduces girls’ and boys’ dropout rates (Anil, Guner, Delibasi, & Uysal (2016)).4 One notable exception is Black, Devereux, & Salvanes (2013), who exploit Norwegian registry data and find that the proportion of girls in middle school positively affects girls’ employment rate and wages several years later but has slightly negative effects on boys’ educational attainment.5 Eventually, a related literature investigates the effects of school peers’ gender in the specific context of single-sex education. Generally speaking, this literature finds positive effects of attending a single-sex schools for both boys and girls (Jackson, 2012, Jackson, 2021; Park, Behrman, Choi, 2013, Park, Behrman, Choi, 2018; Dustmann, Ku, & Kwak (2018)) but the evidence is more mixed as regards the effects of being assigned to single-sex classes in mixed-gender schools (Strain (2013); Lee, Turner, Woo, & Kim (2014); Eisenkopf, Hessami, Fischbacher, & Ursprung (2015); Booth, Cardona-Sosa, & Nolen (2018)).
This paper adds to the literature by providing clear evidence on both short and long term effects of school cohort gender composition on student achievement and educational attainment, in the broad context of mixed-gender education and on the basis of administrative data. These results are suggestive of persistent effects of school peers’ gender on students’ skills and preferences. This is key to understand the long-term consequences of gender imbalances at school, especially in the context of considerable gender imbalances across fields of study observed in developed countries (OECD (2017)). In addition, this paper is one of the first to provide clear evidence of negative effects of female peers on boys’ achievement and educational attainment in a context of mixed-gender education, with the notable exceptions of Black et al. (2013) and Hill (2015). Finally, it sheds new light on both direct and indirect mechanisms through which school peers’ gender can affect students’ academic performance and trajectories, namely the effects of peers’ gender on student behaviour and school attitudes and the adjustment of teaching practices to the gender composition of the classroom.
The remainder of the paper is organized as follows. Section 1 describes the French institutional context. Section 2 presents the datasets exploited in this paper and the outcomes under consideration. Section 3 presents the empirical strategy. Section 4 provides evidence on the validity of the identifying assumption and presents the estimations of school peers’ gender effects on student outcomes. Section 5 discusses potential mechanisms driving these effects. The final section concludes with a discussion of the implications of the main results.
Section snippets
School system, examinations and track choices
Secondary education in France consists of four years of lower secondary education in middle school (from 6th to 9th grade), and three years of upper secondary education in high school (from 10th to 12th grade). Schooling is compulsory up to age 16 in France. This paper focuses on the population of 9th grade students (14–15 years old) enrolled in public middle schools. An important feature of French middle schools is that students stay in the same class with the same teachers throughout the
Datasets and sample restrictions
This paper exploits a comprehensive panel dataset of students enrolled in secondary education in France over the 2007–2012 period.10 This dataset provides detailed information on student enrollment status every year, basic student demographics (age, gender, nationality and financial aid status) as well as the unique identifier of the class in which the student is enrolled and student performance at national examinations at
Empirical strategy
The identification of peer effects raises several identification challenges that have been well documented in the literature. As outlined by Manski in a seminal paper (Manski (1993)), behaviours and outcomes of peer group members may be similar for several reasons. This may be due to (1) the effect of peers behaviour on individual behaviour (“endogenous effects”), (2) the effect of peers characteristics on individuals behaviour (“exogeneous effects”), or (3) endogenous peer group selection and
Evidence on the validity of the identification assumption
The key identifying assumption to interpret the results as causal is that within school cross cohorts variation in the gender ratio is not correlated with changes in peers’ and teachers’ quality. To provide evidence on the validity of this assumption, I thus implement balancing tests on student and teacher characteristics, using the main specification. The results of these tests are displayed in Table 1. Consistently with the identification assumption, the proportion of female students in the
Student and teacher behavioural responses
Previous studies who find positive effects of female school peers on both girls’ and boys’ learning suggest that such an effect could be due to an improved quality of the learning environment (Hoxby (2000); Lavy & Schlosser (2011); Hu (2015)). In line with this idea, there is ample evidence in the literature that girls exhibit lower levels of disruption and tend to be more conscientious than boys on average (Jacob (2002); Duckworth & Seligman (2006); Bertrand & Pan (2013); Cornwell et al. (2013)
Conclusion
This paper shows that school cohort gender composition in middle school have persistent effects on student achievement and human capital acquisition. Being surrounded by more female peers has a positive influence on girls’ performance at the end-of-middle school national examination, whereas it has a detrimental effect on boys’ performance. Furthermore, it decreases girls’ dropout rate after middle school and increases their probability to graduate from high school, especially from the
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (71)
- et al.
Do single-sex classes affect academic achievement? an experiment in a coeducational university
Journal of public economics
(2018) - et al.
Why are single-sex schools successful?
Labour economics
(2018) - et al.
Academic performance and single-sex schooling: evidence from a natural experiment in switzerland
Journal of Economic Behavior & Organization
(2015) - et al.
Gender gaps in primary school achievement: A decomposition into endowments and returns to iq and non-cognitive factors
Economics of education review
(2014) The positive influence of female college students on their male peers
Labour economics
(2017)Do girl peers improve your academic performance?
Economics letters
(2015)Single-sex schools, student achievement, and course selection: Evidence from rule-based student assignments in trinidad and tobago
Journal of public economics
(2012)Where the boys aren’t: Non-cognitive skills, returns to school and the gender gap in higher education
Economics of education review
(2002)Do gender stereotypes reduce girls’ or boys’ human capital outcomes? evidence from a natural experiment
Journal of public economics
(2008)- et al.
On the origins of gender gaps in human capital: short-and long-term consequences of teachers’ biases
Journal of public economics
(2018)
Causal effects of single-sex schools on college entrance exams and college attendance: Random assignment in seoul high schools
Demography
L’Impact de la taille des classes sur la réussite scolaire dans les écoles, collèges et lycées français: estimations à partir du panel primaire 1997 et du panel secondaire 1995
Girls, girls, girls: Gender composition and female school choice
Economics of education review
Peer gender and educational choices
Empirical economics
Boys lag behind: How teachers gender biases affect student achievement
Economics of education review
Resource and peer impacts on girls’ academic achievement: Evidence from a randomized experiment
American Economic Review
The effects of high school peers gender on college major, college performance and income
The Economic Journal
Does Classroom Gender Composition Affect School Dropout?
Technical Report
From proof of concept to scalable policies: challenges and solutions, with an application
Journal of Economic Perspectives
Does gender matter?
Nature
The trouble with boys: Social influences and the gender gap in disruptive behavior
American Economic Journal: Applied Economics
The long-term impacts of low-achieving childhood peers: Evidence from project star
Journal of the European Economic Association
Under pressure? the effect of peers on outcomes of young adults
Journal of labor economics
Reclaiming school mathematics: The girls fight back
Gender and education
Gender differences in competition: The role of single-sex education
Journal of Economic Behavior and Organisation
From Teacher Quality to Teaching Quality: Instructional Productivity and Teaching Practices in the US
Technical Report
Does evaluating teachers make a difference?
Technical Report
Gender, competitiveness, and career choices
The quarterly journal of economics
The adolescent society: The social life of the teenager and its impact on education
Noncognitive skills and the gender disparities in test scores and teacher assessments: Evidence from primary school
Journal of Human Resources
The effects of person versus performance praise on childrens motivation: Gender and age as moderating factors
Educational psychology
Self-discipline gives girls the edge: Gender in self-discipline, grades, and achievement test scores.
Journal of educational psychology
Peer effects, teacher incentives, and the impact of tracking: Evidence from a randomized evaluation in kenya
American Economic Review
Rapport dévaluation de lassouplissement de la carte scolaire
Technical Report
Gender and gender role development in adolescence
Handbook of adolescent psychology
Cited by (9)
Gender peer effects in high schools: Evidence from India
2024, Journal of Economic Behavior and OrganizationLearning effects of conditional cash transfers: The role of class size and composition
2024, Journal of Development EconomicsPeers’ race in adolescence and voting behavior
2023, Economics of Education ReviewOpposite-gender friendships and learning performance of students: Evidence from China
2023, International Journal of Educational DevelopmentBoys left behind: The effects of summer camp and follow-up strategies on academic, personal, and social competencies
2023, Economics of Education ReviewHealth benefits of having more female classmates: Quasi-experimental evidence from China
2022, Economics of Education ReviewCitation Excerpt :Second, the rich information in the CEPS data enables us to address contextual confounding by controlling for a large set of covariates at the student, parent, household, and class levels, as well as a full set of school fixed effects in the estimation. There has been substantial evidence across cultures that adolescents benefit from having more female classmates in terms of academic performance (Gong et al., 2021; Hill, 2015; Hoxby, 2000; Hu, 2015; Lu and Anderson, 2015; Briole, 2021), cognitive skills (Black et al., 2013; Lavy and Schlosser, 2011), non-cognitive skills (Gong et al., 2021), and mental health (Guo et al., 2021). However, to the best of our knowledge, little is known about whether having more female classmates would benefit adolescents’ physical health.
I would like to thank Luc Arrondel, Manon Garrouste, Marc Gurgand, Elise Huillery, Fanny Landaud, Eric Maurin, Sandra McNally, Ariell Reshef and seminar participants at the EALE 2018 conference (Lyon), the AFSE 2018 annual meeting (Paris) and the Paris School of Economics seminars for their valuable comments. I also thank the French Ministry of Education for providing me with access to the data exploited in this paper. Declarations of interest: none.