Illustrated by Evelyn Hsy
“I cannot tell you how many parents have come up to me over the years and said, ‘Because I read your book Outliers, I held my kid back from starting school, and it was the best decision I ever made.’”
— Malcolm Gladwell in his podcast Revisionist History
Many parents believe that holding their children back a year will benefit their academic performance. However, our survey of National Merit Semifinalists—students who score in about the top 1% on the PSAT/NMSQT—in our school district suggests that holding your kid back will not in fact help them become semifinalists, as the semifinalists were on average neither old nor young for their grade.
While some parents obtain exemptions for their children and some students subsequently skip or repeat grades, there’s a cutoff for how young students can be when they start kindergarten. In California, between 2005 and 2010, this cutoff was five years old on or before December 2nd. The Relative Age Effect (RAE) theorizes that younger children often perform worse than older children in their grade because they are less developmentally mature, particularly in lower grades when a one-year gap between them and their peers is more noticeable.
Among SUHSD National Merit Semifinalists there is no significant RAE. Admittedly, our sample size of 41 students who started school between 2005 and 2010 is small, and there are many confounding variables.
The graph to the right shows the number of semifinalists born in each relative month. Those between 0 and 12 are a normal age for their grade. Those with negative numbers are born after the typical cutoff for their grade, making them relatively younger, while those with numbers above 12 are born before the typical age cutoff for the grade above, making them relatively older. The average and median relative months among the sampled semifinalists are 6.44 and 7, respectively. These fall squarely within the normal ages for their grade, and thus do not indicate an RAE.
“The Relative Age Effects in Educational Development: A Systematic Review,” a review of studies that was published in the International Journal of Environmental Research and Public Health, found that “Relatively younger children in [the] same class groups: (a) obtain significantly lower mean scores in cognitive and motor tests, (b) have a higher repetition rate, and (c) have a less[er] capacity of socialization.”
The review concluded, “In the academic domain, there are significant differences in favor of relatively older schoolchildren. These differences are measurable in cognitive, motor, and socio-emotional performance. Although the RAE diminish at older ages, their long-term (adulthood) impact may be important in terms of further education, income and family savings capacity.”
Dr. Alar Urruticoechea, one of the review’s authors, said that there are studies “that go beyond the educational stage and report a relationship between salary, family life, suicide, etc. and relative age.” Even so, one of the studies in the review did not find any long-term effects.
What Can Reduce the RAE
The review also found that the RAE had a particularly large effect on boys, students in lower socio-economic levels, immigrants, and those with less family support. However, Dr. Urruticoechea noted, “There are variables that could be protective and make these [relative age] effects less important in the development of children, such as family support, teacher training, the adaptation of the center to the needs of the child and not the child having to adapt to the center, etc.”
Our SUHSD semifinalist data does not necessarily mean that the RAE is nonexistent. Because some M-A students come from wealthier neighborhoods like Menlo Park and Atherton, students’ socio-economic status may contribute to the absence of an RAE. In addition, since we looked only at those who scored highest on the PSAT, our data does not capture whether and to what extent the RAE affects those who score lower. Furthermore, essentially all students taking the NMSQT are at least 15, and the RAE tends to decrease as students get older. Some studies included in the review found an RAE in academics that disappeared by age 15.
That being said, other studies looking at similar metrics have found an impact from the RAE on older students. Dr. Urruticoechea said, “A study, not yet published, that we conducted shows how there are statistically significant differences in favor of relatively older children in the PISA [Program for International Student Assessment, an international exam taken by 15-year-olds] 2018 tests for Spain.”
What to Do
Dr. Urruticoechea said that people frequently overlook the RAE and “look for other answers to [students’] difficulties, such as ADHD or other pathologies that can be confused on many occasions with the brain’s own immaturity.”
That said, even if the RAE is correctly identified, reacting to it is not necessarily beneficial. Gladwell noted, “Parents holding their kids back doesn’t solve the problem, it just creates a relative age effect arms race.” If parents of all the younger kids hold their kids back, then those who used to be the older kids will then be the ones negatively affected by the RAE. Then parents of those kids might hold their kids back, disadvantaging the others, and so on.
Both the review and Gladwell offered suggestions of how systems can combat the RAE.
Gladwell said, “In Australia, they’ve invented something called Maturity-Based Corrective Adjustment Procedures—Mat-CAPs, as it’s known—for provisional use in the sport of swimming.” He suggested adapting Mat-CAPs for use in education by running test scores through “an age correction algorithm.”
Dr. Urruticoechea highlighted a greater variety of options. He said, “There are authors who propose to make groups and passage between groups more flexible depending on the relative age and abilities of each child, to evaluate each child exactly when they have a certain relative age, to delay the enrollment of these children, [and] to generate support plans, among others.”
Dr. Urruticoechea added, “I really believe that we should consider the RAE as a variable when researching education and see it for what it is: one more factor among those that can help or not in the teaching/learning process and make decisions to help education be more inclusive and equitable for all people.”
There is no clear consensus about the extent to which the RAE exists. It would seem from the findings of experts that the RAE exists even when controlling for some other variables. However, there are a variety of other factors that are likely far more important in determining a student’s academic success, including financial and familial support. Our spread of National Merit Semifinalists reflects the RAE’s insignificant influence on academic performance for at least this subset of SUHSD students.