The effects of month of birth on student achievement and academic performance

Form for thesis proposal

Research question

To what extent does the month of birth affect students’ results and academic performance?

Motivation

With the general purpose of increasing performance and maximizing outcomes, previous research has been conducted on whether the month of birth affects students’ academic success and achievements. It appears that students born at the end of the academic year tend to have lower results than those born at the beginning of the academic year (Crawford, Dearden, & Greaves, 2013). The month of birth is a main factor causing most of the differences in educational attainment between students (Russell & Startup, 1986). This is however a relevant issue which should concern the policy makers because there are both short- and long-term effects on the students’ level of confidence, behavior, and skills, as a result of these detrimental effects. Moreover, there could be potential repercussions in terms of performance during adulthood (Crawford et al., 2013). Nevertheless, with an increased degree of flexibility within schools, parents would no longer have to attempt to improve their children’s academic achievements through costly activities, such as private schooling, tutoring services, or outside test preparation (Robertson, 2011). Literature review

The available scientific research agrees that these differences in performance are mainly caused by the school rules regarding the entry age for students. Moreover, there have generally been reached similar conclusions concerning the short-term outcome and slightly different results relating to the long-term effect. Crawford et al. (2013) found that there is a relatively large gap between students born in summer and students born in autumn. This gap slightly decreases as students approach the end of the compulsory schooling, but it is still statistically significant. However, there is reduced evidence as to whether these differences persist during adulthood. In accordance with these conclusions, another research conducted by Russell and Startup (1986) highlights that performance at school is a function of age and that age is a key driver in the differences of students’ results. Aliprantis (2014) encourages investment in early childhood education and argues that such educational investments provide a higher return if made at a younger age rather than later in the life cycle. Within their research, Bell and Daniels (1990) analyze whether students born during summer have a disadvantage when compared to students born in autumn and they reach similar conclusions to those stated above. However, a particular feature is that the outcome of their research suggests that the performance gaps persist also later in the life cycle. Robertson (2011) finds that over time the performance of the youngest and oldest students tends to be equal. However, the research cannot explain what factor or mechanism determines this later relevant increase in academic achievement of the youngest students.

Hypothesis

Within this paper, I will analyze whether the month of birth indeed affects students’ academic performance and I expect that it has a decreasing effect beginning on the students born in September, whom I expect to have the highest degree of academic success, and ending on those born in August, whom I await to prove to be disadvantaged and thus have lower results. Differently stated, I expect that students born during autumn will perform

better than those born in winter, who will have higher results than those born in spring, who in turn will succeed more than the remainder, the students born during summer. 𝐻": 𝛽&'()'*+', = 𝛽./)0+', = ⋯ = 𝛽2345 = 𝛽63738) 𝐻₉: 𝛽_&'()'*+',> 𝛽_./)0+', > ⋯ > 𝛽₂₃₄₅ > 𝛽₆₃₇₃₈₎ 𝛽;0<)= is a regression coefficient that measures the relationship between month of birth and the result obtained by each student. This, together with the research method used will be extensively discussed in the following section. Research method In order to support the empirical work, I will perform an OLS multiple regression and attempt to estimate the effect of month of birth on students’ academic results. I intend to analyze to what extent do these results persist in the later schooling period, and therefore I will investigate the month of birth effects on academic performance of college students. The data used is obtained from Dataverse Harvard and it comprises students’ month of birth, gender, race, class size, and results in fields such as mathematics, reading, word study skills, and listening. I start by considering a relatively general model, wherein the results of each student at a particular time are a function of age (month of birth), gender, and race 𝑅? = 𝑓(𝑤?, 𝑥? , 𝑦? , 𝑧?)

where 𝑅? is student 𝑖’s result, 𝑤? is student 𝑖’s month of birth, 𝑥? is student 𝑖’s gender, thus 𝑥_? ∈ {0; 1}, 𝑦? is student 𝑖’s race, white, black, Asian, Hispanic, native American, or other, and 𝑧_? is the class size to which student 𝑖 belongs. Following this relationship of dependency, I propose the following model, of the form of a regression equation, to be considered further in this paper: 𝑅?,) = 𝛼?,) + 𝛽;0<)=Q,R ∗ 𝑤? + 𝛽T'<U',Q,R ∗ 𝑥?+ 𝛽VW/'Q,R ∗ 𝑦? + 𝛽X4W88&?Y'Q,R∗ 𝑧?+ 𝜀?,)

where 𝑅?,) is student 𝑖’s result at time t, 𝑤? is student 𝑖’s month of birth, 𝑥? is student 𝑖’s gender, 𝑦? is student 𝑖’s race, 𝑧? is student 𝑖’s class size measured in number of students, 𝛼?,), 𝛽_;0<)=Q,R, 𝛽T'<U',Q,R, 𝛽VW/'Q,R, and 𝛽X4W88&?Y'Q,R are regression coefficients which encompass the structural relationships with the results of each student, and 𝜀?,) is the error term. Additionally, I assume that the error term 𝜀?,) is normally distributed with a mean 0 and a constant variance 𝜎\_{, which is the residual sample variance.}

By performing these regressions, I expect to find the highest 𝛽;0<)=Q,R for September, which should thereafter gradually decrease in the following months, reaching the lowest value for month August. Thus, I could indeed conclude that the students born in summertime have an academic and educational attainment disadvantage than those born during autumn, and that the month of birth has a concerning degree of influence on the academic performance.

References Aliprantis, D. (2014). When Should Our Children Start School? Journal of Human Capital, 8(4), 481–536. http://doi.org/10.1086/679109 Bell, J. F., & Daniels, S. (1990). Are Summer-Born Children Disadvantaged ? The Birthdate Effect in Education. Oxford Review of Education, 16(1), 67–80. Crawford, C., Dearden, L., & Greaves, E. (2013). When you are born matters : evidence for England. London: The Institute for Fiscal Studies. Robertson, E. (2011). The effects of quarter of birth on academic outcomes at the elementary school level. Economics of Education Review, 30(2), 300–311. http://doi.org/10.1016/j.econedurev.2010.10.005 Russell, R. J. H., & Startup, M. J. (1986). Month of birth and academic achievement. Personality and Individual Differences, 7(6), 839–846. http://doi.org/10.1016/0191-8869(86)90082-6