The school schedule is the main influence on the organization of time in the lives of children and young adults. It determines when students enter and leave school, when they eat, when they are indoor or outdoor, and in doing so also influences the timing and duration of sleep, when they socialize, and when they play sports, do homework, and have free time. All of those activities have an influence in the circadian health or circadian harmony. In a recent work, Madrid (2022) makes a very complete review of the state of the art in relation to circadian health, which is only achieved when there is harmony between four clocks: the internal, the environmental, the metabolic and the social clock. In this paper we propose an analysis to understand the effects of the school schedule in academic performance.

Spain is one of the few European countries where a double-shift school system is implemented in public schools, either following the compact schedule, morning shift only CS], where students attend all sessions before lunch; or in extended hours, morning and afternoon shift [ES], where student attend sessions in both morning and afternoon hours. The typical compact schedule is 9am to 2pm for school (ages 6 to 12) and 8:30am to 2:30pm for high school (ages 12 to 16). The usual extended hours are from 9 a.m. to 12:00 p.m. and then from 3 p.m. to 5 p.m. for school and from 9 a.m. to 1 p.m. and then from 3 p.m. to 5 p.m. for high school (Egido, 2011; Galván, 2013; Gabaldón-Estevan, 2021).

Data connecting school hours to student performance in Spain are scarce. An exception is the study by Escribano and Díaz-Morales (2014) on Spanish 12 to 16 years old students found a similar attention pattern for compact schedule (increased throughout the school day till midday) for both chronotypes; they also found that boys tended to reach on average and at different times of-day (in the middle and at the end of the school day) higher attention than girls, but no comparison between compact and split schedule was made in this study. Other small-scale longitudinal data do not seem to yield statistically significant results for or against a given type of schedule (Fernández Enguita, 2002; Hóspido, et al. 2019).

The objective of this article is to analyse the effect of school schedule on academic performance, placing special emphasis on the differences between boys and girls. To do this, we use data from the 2019 Diagnostic Tests of the Valencian Community. The results show that, although on average the compact schedule do not affect performance, they do accentuate the gender gap between boys and girls.

The data for our empirical analysis come from a standardized exam carried out each year to third grade students in the Valencian Region in Spain. The test is called a “Diagnostic Test.” This exam is mandatory for all primary publicly funded schools). The exam has no academic consequences for students; its objective is to provide objective information, to educational authorities. We analyse the scores for this exam for the 2018/2019 academic cohorts. The exam is very rich and consists of several parts. For mathematics, a single test is taken, but for Spanish and Valencian and the foreign languages, , four tests are taken: oral expression, written expression, oral comprehension and written comprehension. LOOK. Our measures of student performance are mathematical competence and the average of the four variables for Spanish, Valencian and foreign language, . Additionally, each student and his/her family fill out an extensive questionnaire. The questionnaire asks students to answer questions about themselves and the environment in which they live. The responses to the questionnaire offer rich information about the students' individual characteristics and their family background and living environment. Children's performance is a combination of individual and family characteristics, as well as educational resources. We have two school variables: whether the course is taught in Spanish or in Valencian, and our predictor of interest, whether the school of the student follows a extended schedule (ES) with two classroom sessions, one in the morning and one in the afternoon. or a compact schedule in a single session (CS). We use an Ordinary Least Squares (OLS) estimation model. This procedure presents biases since the choice of school in Spain is not random, as we explain below. The estimated models would be the following: S_i=α+β〖CS〗_i+γ〖LP〗_i+u_(i ) (1) S_i=α+β〖CS〗_i+δ〖CS〗_i×〖female〗_i+ρX_i+γ〖LP〗_i+u_i (2) where Si is the academic result of interest, CSi is a binary variable that is equal to 1 if the individual attends a class with CS and equal to 0 if it is otherwise. LPi is a dummy variable that takes the value 1 if the teaching is taught in Spanish and zero otherwise. The variable femalei is a dummy variable that takes the value 1 if the gender is female and the value zero if the gender is male, Xi is a set of observable characteristics of the students and their families and ui is a random component. Our variable of interest is the effect of CS on Si, for boys and girls separately, which is captured in model (2) by coefficients β and δ.

With respect to the model (1) we can say that there is a negative effect of CS on academic results on average, being only significant for foreign language and Valencian. It should also be noted that teaching classes in Spanish has a negative effect on all subjects, except for Spanish language. The estimates that we have obtained with the first model show the effect for all students. However, this could mask the results. There is literature that evinces that there are potential heterogeneous effects depending on gender, see for example Dahmaun (2017) and Hek et al (2018). The estimates of model (2) show the following. In mathematics we obtain a significant and positive effect for girls. Calculating the marginal effect, this is 0.307 for girls, who do worse in mathematics. For boys we get a negative effect. For Spanish, although they have the expected sign, we do not obtain any significant effect. On the other hand, the marginal effect is highly significant for men and with a negative sign in foreign language. Finally, there is a negative marginal effect for both (-0.545 for men and -0.251 for women) for Valencian Language

Alon, S. and D. Gelbgiser (2011) “The female advantage in college academic achievements and horizontal sex segregation” Social Science Research 40 , 107-119. Brindusa, A. Cabrales, A. ana J. M. Carro (2016), “Evaluating a bilingual education program in Spain: The impact beyond foreign language learning. Economic Inquiry 54(2), 1202-1223. Brindusa, A. and A. Cabrales (2010), “Los determinantes del éxito en la educación primaria en España” Fedea Working Paper. Buchmann, C. DiPrete T.A and A. McDaniel (2008), “Gender Inequalities in Education” Annual Review of Sociology, 34(3), 319-37. Carrell, S.E., Maghakian, T. and West J. E. (2011), “The causal effect of school start of school start time on the academic achievement of adolescents”. American Economic Journal: Economic Policy, 3, 62, 81. Crespo, L. Fernandez, M. Hospido, L. and J. Montalban. (2020) “¿Qué sabemos sobre el efecto del tipo de jornada escolar en el rendimiento académico?”. Dahmann, S.C. (2017), ”How does education improve cognitive skills? Instructional time versus timing of instruction” Labour Economics 47, 35-47. Edwards, F. (2012), “Early to rise? The effect of daily start times on academic performance” Economics of Education Review, 31, 970-983. Ewert, S. (2012), “Fewer Diplomas for Men: The Influence of College Experiences on the Gender Gap in College Graduation” The Journal of Higher Education, 83,6. Pope, N. G (2016), “How the time of day affects productivity: Evidence from school schedules”. The Review of Economics and Statistics 98(1), 1-11. Raza, U. and U. Hazir (2019), “Boys versus girls’ educational performance; Empirical evidences from global north and global south”. African Educational Research Journal, 7(4), 163-167.