Are girls always good for boys? Short and long term effects of school peers’ gender

https://doi.org/10.1016/j.econedurev.2021.102150Get rights and content

Highlights

  • This paper investigate the effect of school peers gender in 9th grade.

  • It exploits within school cross cohort variations in French middle schools (2008–2011).

  • Female peers positively affect female students and negatively affect male students.

  • The impact is observed on short term performance and long-term educational attainment.

  • Evidence suggests that student and teacher behaviour may drive this effect.

Abstract

This paper exploits idiosyncratic variations in school cohorts’ gender composition to investigate the short and long-term effects of school peers’ gender. Using French administrative data over the 2008–2012 period, it shows that the proportion of female peers’ in middle school not only affects students’ contemporaneous performance but also influences their subsequent educational attainment. More specifically, a larger share of girls among school peers increases girls’ test scores, reduces their dropout rates and increases their probability to graduate from high school several years later, especially in the scientific track. By contrast, it increases boys’ probability to attend a vocational school and decreases their high school graduation rate. I find suggestive evidence that these effects partially operate through a negative effect of opposite-gender peers on students’ school behaviour and through an adjustment of teacher behaviour based on the gender composition of the classroom.

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)

  • H. Park et al.

    Causal effects of single-sex schools on college entrance exams and college attendance: Random assignment in seoul high schools

    Demography

    (2013)
  • T. Piketty et al.

    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

    (2006)
  • N. Schneeweis et al.

    Girls, girls, girls: Gender composition and female school choice

    Economics of education review

    (2012)
  • P. Schøne et al.

    Peer gender and educational choices

    Empirical economics

    (2019)
  • C. Terrier

    Boys lag behind: How teachers gender biases affect student achievement

    Economics of education review

    (2020)
  • D. Whitmore

    Resource and peer impacts on girls’ academic achievement: Evidence from a randomized experiment

    American Economic Review

    (2005)
  • M. Anelli et al.

    The effects of high school peers gender on college major, college performance and income

    The Economic Journal

    (2017)
  • B. Anil et al.

    Does Classroom Gender Composition Affect School Dropout?

    Technical Report

    (2016)
  • A. Banerjee et al.

    From proof of concept to scalable policies: challenges and solutions, with an application

    Journal of Economic Perspectives

    (2017)
  • B.A. Barres

    Does gender matter?

    Nature

    (2006)
  • M. Bertrand et al.

    The trouble with boys: Social influences and the gender gap in disruptive behavior

    American Economic Journal: Applied Economics

    (2013)
  • J. Bietenbeck

    The long-term impacts of low-achieving childhood peers: Evidence from project star

    Journal of the European Economic Association

    (2020)
  • S.E. Black et al.

    Under pressure? the effect of peers on outcomes of young adults

    Journal of labor economics

    (2013)
  • J. Boaler

    Reclaiming school mathematics: The girls fight back

    Gender and education

    (1997)
  • A.L. Booth et al.

    Gender differences in competition: The role of single-sex education

    Journal of Economic Behavior and Organisation

    (2009)
  • S. Briole

    From Teacher Quality to Teaching Quality: Instructional Productivity and Teaching Practices in the US

    Technical Report

    (2019)
  • S. Briole et al.

    Does evaluating teachers make a difference?

    Technical Report

    (2019)
  • T. Buser et al.

    Gender, competitiveness, and career choices

    The quarterly journal of economics

    (2014)
  • J.S. Coleman

    The adolescent society: The social life of the teenager and its impact on education

    (1961)
  • C. Cornwell et al.

    Noncognitive skills and the gender disparities in test scores and teacher assessments: Evidence from primary school

    Journal of Human Resources

    (2013)
  • H. Corpus et al.

    The effects of person versus performance praise on childrens motivation: Gender and age as moderating factors

    Educational psychology

    (2007)
  • A.L. Duckworth et al.

    Self-discipline gives girls the edge: Gender in self-discipline, grades, and achievement test scores.

    Journal of educational psychology

    (2006)
  • E. Duflo et al.

    Peer effects, teacher incentives, and the impact of tracking: Evidence from a randomized evaluation in kenya

    American Economic Review

    (2011)
  • G. Fack et al.

    Rapport dévaluation de lassouplissement de la carte scolaire

    Technical Report

    (2012)
  • N.L. Galambos

    Gender and gender role development in adolescence

    Handbook of adolescent psychology

    (2004)
  • Cited by (9)

    • Gender peer effects in high schools: Evidence from India

      2024, Journal of Economic Behavior and Organization
    • Peers’ race in adolescence and voting behavior

      2023, Economics of Education Review
    • Health benefits of having more female classmates: Quasi-experimental evidence from China

      2022, Economics of Education Review
      Citation 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.

    View all citing articles on Scopus

    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.

    View full text