The Relationship Between School Starting Age and Academic Achievement in Later Schooling 

Abstract

School starting age varies across the world between four and seven years. A debate regarding the optimal starting age exists within the literature. Researchers measure school starting age as either state-permitted (absolute) school entry age, or relative age within a grade cohort. Research demonstrates that younger absolute school entry age does not predict better academic outcomes. Research into relative entry age most commonly finds that older students typically have better academic outcomes, and these persist into high school, and influence entrance into post-secondary institutions. Further research suggests that school-readiness should be measured individually, due to the fact that developmental stage, and chronological age, do not correlate precisely in every candidate for school entry. Various scholars have theorized that younger school entry may help disadvantaged students who start school cognitively behind their more privileged peers; however, there is no significant evidence to support this. Ongoing policy dialogue is necessary to help bring countries and states into alignment with current research findings and develop flexible school-readiness entry procedures.

The age that children start school varies around the world. The wide range of views on the subject is reflected in the different national school entry ages, varying from four to seven years old (Group, 2019; OECD, 2018). Within the literature, two areas of investigation emerge, based on measurements of starting age. School entry age (SAE), (sometimes termed absolute school entry age), refers to the age determined by a country or state at which children may start school. This is usually operationalized as an age cutoff, where children must reach an age by a specified date to enter an academic school year. For example, in Queensland (Australia), a child must be six years and six months old by December 31 of the proposed enrollment year for entry into grade one (Queensland, 2019). Relative school starting age reflects resultant age difference (in months) within a cohort of students from a single academic year. Research in this area uses naturally occurring differences created by either country/state school entry age or relative age differences within cohorts to group participants. A minority of researchers propose that a younger start to formal education may be beneficial, especially for students from disadvantaged backgrounds (Sharp, 2002; Suziedelyte & Zhu, 2015). However, a large body of literature exists suggesting older school starting age allows time for further maturation through which children typically become more “school ready” and able to make better use of educational resources than younger peers (Edwards, Taylor, & Fiorini, 2011). This essay argues that younger school entry age does not typically predict better academic achievements in later schooling. Furthermore, it argues that older relative age, within a starting year cohort, usually positively impacts later academic outcomes. However, this essay argues that substantial developmental differences exist in this period; therefore, optimal school staring age may vary individually.

Reading and writing (literacy) are considered fundamentals of academic ability gained through education, and literacy is considered a powerful determinant for quality of life across a range of domains, including employment success, and social well-being (Christie & Misson, 2012; Darcovich et al., 1997). Suggate (2009) investigated the relationship between school entry age (SAE) and later reading ability. This researcher utilized the variations in SAE between countries and examined the findings of the 2006 Programme for International Student Assessment (PISA) to measure reading competence as a function of SAE (PISA, 2007). PISA is an international comparative study commissioned by the Organization for Economic Co-operation and Development (OECD) to test 15-year-old students in reading, science, and mathematics (Turner & Adams, 2007). Suggate (2009) controlled for social and economic factors using an index (SEFI) consistent with previous research by Elley (1992). Suggate (2009) included 400,000 participants from 56 countries, with SAEs at every age from four, through to seven years, and tested participants on reading skills. Suggate (2009) concluded that SAE did not significantly predict reading skills at age 15. A criticism of this study may be that a single index of social and economic factors may not accurately capture all the socio-economic (SES) variables connected to literacy. However, the consistency of the findings suggests that this was not the case. This research provides extensive international evidence in favor of the argument that younger SAE does not predict literacy ability in later schooling. Those who would criticize Suggate’s (2009) SEFI, may appeal to prior research describing successful preschool literacy intervention as evidence for the merit of younger formal schooling (Bus & van Ijzendoorn, 1999). However, multi-country research of absolute SAE that concludes with the recommendation of lowering SAE is scarce; therefore, relative age research is offered to illustrate an opposing position.

The potential positive effects of younger exposure to formal education on literacy were examined in a cohort of children from low-income families in Virginia (USA) by Huang and Invernizzi (2012). These researchers utilized the difference in age created by the academic year entry cutoff date, of up to 10 months. They conducted testing at the beginning of preschool that demonstrated the literacy of the oldest in the class was initially more advanced than the youngest and followed participants through to the end of grade two, testing them at the end of each academic year. Their results showed that although the oldest in the class remained ahead of the youngest at the end of grade two, the gap had substantially narrowed. The authors interpreted this narrowing as evidence for the younger student’s capacity to learn more than older students over the same duration (Huang & Invernizzi, 2012). However, as this study only followed students for three years, predictions of performance at older ages are speculative. Additionally, the gap between older and younger students remained significant. This research does not jeopardize the principal argument that a younger school entry does not typically predict later improved outcomes, clearly supported by Suggate (2009) in aspects of literacy.

The effect of a relative age difference at school commencement on later academic achievement was further studied by Ponzo and Scoppa (2014). They considered domains beyond literacy, including math and science, which are potent predictors of adult socio-economic success, according to Ritchie and Bates (2013). Ponzo and Scoppa (2014) studied 3581 grade four students in the 2006 Progress in International Reading Literacy Study (PIRLS), 4470 grade four students in the 2007 Trends in International Mathematics and Science Study (TIMSS), and TIMSS results for 4400 grade eight students. Results for 30,780 grade ten students in the 2009 PISA were also analyzed. (Ponzo & Scoppa, 2014). These researchers concluded that on average, younger children within their cohort scored lower at each age period. They also investigated the effect of relative age on high school choice. In Italy, students choose between three final school options: Lyceum is the highest academic option, followed by Technical school, then Vocational school (Ponzo & Scoppa, 2014). The researcher’s results demonstrated a significantly higher probability of choosing the Lyceum when the students were older in the cohort, and conversely, more chance of the vocational pathway when younger. This study supports the thesis that starting younger at school does not deliver short, or long term, educational benefits. Furthermore, findings by Ponzo and Scoppa (2014) would suggest that starting younger may be a hindrance to academic success, whereas being older in a cohort may increase later academic achievement, which are the first and second points made in the opening argument in this paper.

The theory that disadvantage may endure as a reason to recommend an earlier school start age remained pertinent when considering outcomes beyond literacy. Suziedelyte and Zhu (2015) investigated whether this may help close the gap between lower SES children, who are not exposed to as much learning in their home life as higher SES children, for whom opportunities to learn at home are plentiful. They used data from the Longitudinal Survey of Australian Children (LSAC) and exploited the consistent school start dates from several Australian states as a way to divide 5-year-old children who were, or were not, able to start school in a given year, due to the legal age cutoff. They measured children at age 6-7 on cognitive and non-cognitive tests, at a point where early starters had completed two years of schooling and late starters, only one. Suziedelyte and Zhu (2015) analyzed several disadvantage models and argued that when disadvantage was measured as single motherhood status and low maternal education, earlier school starting age helped close the cognitive gap between advantaged and disadvantaged children. However, this effect did not reach statistical significance. Like Huang and Invernizzi (2012), Suziedelyte and Zhu (2015) raised a valid question regarding whether younger school commencement may address disadvantage but ultimately failed to show that it does. While it may be true that disadvantage and cognitive disparities coalesce and impact education; and as argued by OECD (2017) egalitarian societies should redress this, the findings by Suziedelyte and Zhu (2015) do not provide sufficient evidence to advocate a younger school starting age, nor disprove the primary assertions of this essay.

An Australian study by Martin (2009) is of interest because it contains findings that conflict with the larger body of research concerning relative age and academic outcomes. Martin (2009) investigated 3,684 high school students, and the impact of natural birthdate variation, inferred delayed primary school start, and grade repetition at a younger age. This researchers’ findings were that being age-appropriate for a grade leads to the best outcomes across all domains, while older students fared worst, and the youngest, best. Martin (2009) found that that the effect was strongest at the extremes, where markedly older-for-grade students showed little or no advantage (Martin, 2009). The conclusion that younger-for-cohort students fare better than older contrasts with a large body of worldwide research finding otherwise, and Martin’s (2009) unusual results must be viewed in the light of this. Additionally, Martin (2009) states that the most significant conclusion to be drawn from his research is that an analysis of the effects of grade repetition shows a series of negative outcomes associated with it. That finding does not intersect with any arguments defended in this paper.

Robust findings that starkly contrast with Martin (2009) are found in research conducted by Bedard and Dhuey (2006) in a multi-country study showing long term advantage in various academic outcomes as a function of older relative-age-within-cohort created by the standard practice of cutoff dates. These researchers examined the effect of relative age on TIMMS results, in participants from ten OECD countries at grade four and eight. They extended this analysis into older educational outcomes by examining the university pretest activity in Canadian and American high school students, and college enrolments in the USA. In further support of a central argument in this essay, these researchers found that throughout the OECD countries, at the fourth grade “the oldest students scored 4–12 percentiles higher than the youngest students” and at grade eight, “2–9 percentiles higher” (Bedard & Dhuey, 2006, p. 1468). This was strengthened by the additional findings from Canada and the USA, where older students were more likely to engage in pre-university academic programs while in high school, and more likely to enter a top-tier tertiary institution in the USA. These researchers conclude their paper with encouragement to policymakers to address the clear academic outcome gap between older and younger students, as younger students often come from lower socio-economic circumstances, where delayed school start is not financially viable, and thus suffer a double disadvantage (Bedard & Dhuey, 2006).

Although the study by Bedard and Dhuey (2006) is superior in length and breadth to the study by Martin (2009), the disagreement between the two studies is valuable, in that it asks the reader to question how this difference may occur. Of the many possible explanations, one potential answer is worthy of highlighting. Developmental age and chronological age are not necessarily the same concept, nor does child development always progress in a simple linear manner (Steiner, 2001; Suggate, 2009). For example, Boereboom and Tymms (2018) tested children in New Zealand at school starting age and found substantial differences in cognitive skills, e.g. reading, mathematics, and vocabulary. These skills not only varied at school commencement but persisted a year later, and also varied in extent and trajectory between boys and girls, suggesting that individual “school readiness” varies dramatically, and potentially should be measured carefully prior to discussions about age (Boereboom & Tymms, 2018).

In conclusion, the literature presented in this essay supports the arguments proposed at the outset. Younger school entry age does not typically predict higher academic achievements in later schooling. Older relative age, within a starting year cohort, usually positively impacts later academic outcomes. Substantial developmental differences exist in this period; therefore, optimal school staring age may vary individually. This essay has examined research into school entry age across numerous countries and confirmed that younger SAE is not a reliable predictor of academic achievement, specifically in domains of literacy. Research into relative school starting age was also investigated, and this established that younger school starting age was not associated with increased academic achievement. On the contrary, this research found that older students in their cohort gained and retained advantage across multiple academic domains, in high school and entry into tertiary institutions. Finally, this essay highlighted that the assumption of a linear correlation between chronological age and developmental stage might be problematic, and the assessment of individual school readiness may be beneficial. Further research into school starting age is necessary due to the policy inconsistencies noted across the world, and the educational outcome disparities created age cutoff policies. More widespread flexible entry protocols may be desirable based on individual developmental differences and school-readiness.