Does a trait that happens to fit social evaluative criteria become an initial advantage? Evidence from studies of the fit between traits and evaluations, the conversion of evaluations, and the relative age effect

Assessment table and what the literature supports

The claim examined in this note is as follows. “An initial advantage tends to arise when an individual’s traits happen to fit society’s evaluative criteria and are recognized by others. For example, consider a child who happens to be good at logical thinking and a child who happens to be good at swimming. In contemporary society, logical thinking has more established evaluative criteria and is easier to evaluate, so the former child is more likely to gain an advantage.” We divided this claim into four parts and assessed each part.

The literature supports the following five statements.

1. The same trait is recognized differently depending on the culture and period, the evaluator, and the yardstick. Many studies consistently support this statement.
2. A process also exists in which early recognition leads to later advantages through the allocation of opportunities. However, classroom expectancy effects are small and often dissipate over time.
3. The statement that “logical thinking is easier to evaluate” is not supported at the childhood stage. Even in swimming, which is measured by times, early selection picks up the chance factor of birth month, and the bias becomes stronger as the selection level rises.
4. The evidence is more consistent with the view that the difference depends not on measurability but on how broad the routes are through which an evaluation is converted into grades, credentials, occupations, and wages. However, no study directly compares logical thinking with swimming ability, and this view is an inference from indirect evidence.
5. For the claim as a whole (a chance fit produces early evaluations, which become the starting point of cumulative advantage), each component has prior research, but in this survey we found no source that states the components as one chain (provisional).

Returning to the logical-thinking child and the swimming child

A child who happens to be good at swimming is, if anything, identified earlier during childhood. Competition times make the ranking clear, and the child enters a selected squad and gains good coaching and strong competitors. At this stage, it would be fair to say that the swimming child is the one who is “more easily recognized.”

The difference can widen after recognition. Evaluations of logical thinking are fixed in the form of educational credentials through school grades and entrance examinations, and they are converted into occupations and wages. Evaluations of swimming remain as times and competitive records, but the routes through which they are converted into credentials or wages are narrow. However, the routes are narrow, not absent. Some studies report that the experience of school sports, through participation, is associated with later wages and college attendance (Persico et al. 2004; Stevenson 2010).

Therefore, the conclusion of the example (the logical-thinking child is more likely to gain an advantage) may be supportable if the reason is replaced. The reason is not that “the evaluative criteria are established and the trait is easy to evaluate” but that “the evaluation is converted to a greater degree into opportunities in other domains.” This replaced reason is consistent with indirect evidence, but no study has tested it directly.

What we examined and how

• Scope: definitions of giftedness and social values (implicit theories of giftedness), the fit between temperament and the environment (goodness of fit), cultural and historical differences in the evaluation of the same trait, what evaluators reward (teachers’ grades, sports scouting), institutional change and heritability, the labor market price of skills, the relative age effect (sports and school), sports experience and later wages, the conversion of capital (Bourdieu), and accumulation from early evaluations (expectancy effects, selection, funding thresholds).
• Collection: we gathered about 70 sources from academic databases (Crossref, OpenAlex, Europe PMC) and searches, and retained 36 after screening. For sources read in earlier notes (Do Those Who Have Receive More? Mechanisms, Causal Identification, and Limits of Cumulative Advantage and the Matthew Effect in the Academic Literature, Is the First Advantage Luck? Option Luck and Brute Luck, Ex Ante and Ex Post, and the Evidence on Initial Advantages and Fairness Judgments in the Academic Literature, Is the Ability to Make an Effort Also a Talent? The Heritability of Grit, Self-Control, and Motivation to Learn, and How Far These Traits Can Be Changed), we reused the full-text copies.
• Excluded: sources for which neither the full text nor the abstract was available, namely Chen, Rubin, and Sun (1992, shyness in China and Canada), Caspi, Elder, and Bem (1988, the life course of shy children), DiMaggio (1982), Rosen (1981, the copy was an image PDF from which no text could be extracted), and Heath et al. (1985). Books (Gardner on multiple intelligences, Thomas and Chess, a chapter by Csikszentmihalyi, Bourdieu and Passeron).
• Verification: for 24 of the 36 sources, we obtained the full text (published version, author version, working paper version, or the summary page of a research report). We checked the figures and claims against the relevant passages of these texts. For the sources read in working paper versions (Cornwell et al., Branigan et al., Deming, Autor et al., Persico et al., Case and Paxson, Stevenson, Papageorge et al., Black et al.), the version is stated with the figures. The remaining 12 sources were used within the range of their abstracts and are attributed to their authors with phrases such as “according to the abstract.”

P1: Do trait differences include a part the individual did not choose?

Here, “by chance” is limited to the sense of “not chosen by the individual.” In this sense, the supporting evidence is clear. Polderman et al. (2015) integrated 50 years of twin studies and estimated the heritability of human traits as a whole at 49%. Because people cannot choose their genotypes, at least part of the difference in traits is not chosen by the individual.

Birth month is also a typical difference that the individual did not choose. When children are placed in the same age group, a child born just after the cutoff and a child born just before it differ in age by almost one year, and they also differ in physical and psychological maturity (Musch and Grondin 2001). This difference affects early evaluations both in sports selection (Musch and Grondin 2001) and in teachers’ judgments at school (Campbell 2014).

However, heritability is not “the share determined by chance.” Genetic differences are inherited from parents and are also linked to the family environment (Is the Ability to Make an Effort Also a Talent? The Heritability of Grit, Self-Control, and Motivation to Learn, and How Far These Traits Can Be Changed). “Not chosen by the individual” and “randomly assigned” are different statements, and what P1 can establish is the former. The distinction between types of luck is discussed in Is the First Advantage Luck? Option Luck and Brute Luck, Ex Ante and Ex Post, and the Evidence on Initial Advantages and Fairness Judgments in the Academic Literature.

P2: Is the same trait recognized differently depending on whether it fits the evaluative criteria?

Social values are part of the definition of giftedness

Whether a person is regarded as gifted depends in part on whether society regards the domain as valuable. According to the abstract of Sternberg and Zhang (1995), the two authors proposed five conditions for people to call someone “gifted,” each of which is necessary and which are jointly sufficient. The summary of Sternberg’s (1995) research report lists the five conditions as excellence, rarity, productivity, demonstrability, and value, and states that society regarded people as gifted to the degree that they met these five conditions. The same report also states that teachers rated students whose thinking styles matched their own more highly, that students overestimated the degree to which they matched their teachers, and that students who were taught and assessed in ways that matched their ability patterns performed better. The “value” here refers to almost the same thing as the “evaluative criteria” in the claim.

Evaluations change with whether temperament fits the demands of the environment

Goodness of fit is the idea that a child’s adjustment depends not on the child’s temperament itself but on whether the temperament fits the demands of the environment¹. According to the abstract of Lerner, Lerner, and Zabski (1985), the authors predicted, in a sample of 194 fourth-grade children, that children whose temperament matched or exceeded teachers’ demands or expectations would function better than those whose temperament fell short. They obtained partial support for two of the three temperament attributes for which a direction was predicted. Teachers rated the children’s academic ability and adjustment, and objective measures of academic achievement were also administered. Because the support is only partial, this study shows no more than that fit can affect evaluation.

The same shyness receives different evaluations across periods and countries

Shyness in China is an example in which the evaluation of the same trait reversed with social change. According to the abstract of Chen, Cen, Li, and He (2005), among elementary school children (mean age 10 years) in three cohorts in 1990, 1998, and 2002, shyness was associated with social status and academic achievement in 1990; in 1998, the associations weakened or disappeared; and in 2002, shyness was associated with peer rejection, school problems, and depression. Sociability and cooperation were associated with peer acceptance in every year, and aggression was associated with difficulties in every year, so what changed over time was the evaluation of shyness. The study compares the three years as different stages of social change in China.

Differences also exist across countries. According to the abstract of Liu et al. (2015), among children in grades 4 to 8 in China and Canada in the 2010s (787 and 1,033 children), the relation between shyness and adjustment did not differ between the two countries, but unsociability, a preference for being alone, was more strongly associated with adjustment difficulties in China. The authors explained this difference by the contrast between group-oriented and individual-oriented values. The two studies differ in the years examined and in the type of comparison; when they are placed side by side, however, the trait whose evaluation differed between China and Canada in the 2010s was not shyness, whose evaluation changed in China between 1990 and 2002, but unsociability.

What do evaluators reward?

Teachers’ grades do not reward test scores alone. Cornwell, Mustard, and Van Parys (2013) showed, using a national survey of elementary school children in the United States (ECLS-K), that the grades given by teachers did not match test scores and that the discrepancy favored girls in every subject (abstract of the working paper version). Boys whose test scores were equal to those of girls were also graded less favorably, but the gap almost disappeared when noncognitive skills such as approaches to learning were controlled. A one-standard-deviation higher rating in approaches to learning was associated with reading grades that were 0.19 to 0.26 standard deviations higher (same source, full text). This means that, even with the same academic ability, recognition changes depending on whether another trait included in the teacher’s yardstick fits.

For evaluators in sports, an experiment has shown that the way the yardstick is presented changes evaluations. Mann and van Ginneken (2017) showed 25 scouts from elite soccer clubs a youth match and asked them to rank the players’ potential. In the group given no age information and in the group given dates of birth, a bias toward rating relatively older players within the age group more highly remained; in the group told that the shirt numbers were ordered by relative age, the bias disappeared. Because the players were the same and only the presentation of the yardstick differed, this is the experiment closest to P2.

Christensen (2009), based on interviews with 8 coaches of national youth soccer teams, showed that coaches identify talent not through measurement but through a practical sense of recognizing familiar patterns of movement, and that they prefer players who are likely to learn, practice, and develop; the study concluded that talent is socially constructed within top-level soccer.

Whether abilities appear in grades depends on what is valued

The values of the school as a whole also change which abilities are recognized. According to the abstract of Coleman (1960), the study examined, in 10 schools with different values, the hypothesis that in adolescent groups where academic achievement is valued, students with high grades include many who actually have high intelligence, whereas this is not the case in groups where academic work is not valued. The study then presented evidence of a strong effect of interscholastic athletics as a source of the schools’ values. The way the question is framed (whether the same intelligence is recognized in the form of grades depends on what the surrounding group values) corresponds directly to P2. However, the abstract does not report the result of testing the hypothesis, so we do not count this study as evidence supporting P2.

Institutional change also changes which differences appear in outcomes. Rimfeld et al. (2018) reported, for 12,500 people in Estonia, that the proportion of variance in educational attainment and occupational status explained by DNA differences was about twice as large after independence as during the Soviet era (abstract of the author-accepted version). For example, the variance explained by the polygenic score for educational attainment rose from 2.1% in the Soviet era to 6.1% after independence (in the analysis with a 10-year cutoff). Branigan, McCallum, and Freese (2013), in a meta-analysis of 15 samples, concluded that the heritability of educational attainment differs by country, sex, and birth cohort and varies with the environment. Even the same genetic differences appear in outcomes to different degrees depending on the social sorting mechanisms.

Bourdieu (1986) described this relation for family culture. He stated that what is acquired early in the family becomes either a positive value (an advantage of being ahead) or a negative value (lost time), depending on how far it is from the demands of the school market. In a footnote, he also wrote that dispositions given a negative value in the educational market can receive a very high value in other markets.

Assessment of P2

The direction of the relation is supported by the same conclusion from separate lines of evidence: cultural differences, historical differences, evaluators’ yardsticks, and institutional differences. However, most of the evidence is correlational or comes from comparisons across contexts, and studies that place children with the same size of difference under different evaluative criteria are limited to Mann and van Ginneken (2017) and a few others. P2 is supported, but the size of the effect remains undetermined.

P3: Is logical thinking easier to evaluate than swimming ability?

P3 can be read in two ways. One is the measurability reading, in which “logical thinking is easier to measure”; the other is the conversion reading, in which “evaluations of logical thinking are more easily converted into opportunities in other domains.” Here, conversion refers to an evaluation received in one setting becoming an opportunity in another setting, such as grades, entrance examinations, educational paths, occupations, and wages. The evidence points in opposite directions for these two readings.

The measurability reading is contradicted

In swimming, rankings are determined by times, and swimming can be said to be easier to measure than logical thinking. Nevertheless, in swimming, early evaluation and selection pick up birth-month differences. According to the abstract of Cobley et al. (2019), among 2,141 male 100 m freestyle swimmers aged 13 to 16 who competed in state and national competitions, relative age effects were observed in raw times at ages 13 and 14 and became stronger as the selection level rose. At age 15, the odds ratio between the oldest and youngest quartiles within an age group was 2.28 for the top 50% and 6.02 for the top 10%. When times corrected for age-related development were used, this bias almost disappeared. Times are precise, but they measure not only talent but also differences in maturity.

The relative age effect in swimming is not an exception. In the systematic review by Lorenzo-Calvo et al. (2021), a relative age effect was observed in 58.65% of the swimmers analyzed (121 samples, 24,906 swimmers); it was stronger in younger categories and among males, weakened from around ages 17 to 18, and disappeared in elite competition. Across sports, according to the abstract of Cobley et al. (2009), a meta-analysis of 253 samples from 14 sports in 16 countries found that relative age effects were consistently present but small in effect size, and largest among males aged 15 to 18, at the regional and national representative levels, and in highly popular sports.

These results indicate that, at the childhood stage, the swimming child is also evaluated early and strongly. The statement in P3 that “logical thinking has more established evaluative criteria and is easier to evaluate” is inconsistent with this evidence as long as it is read as a statement about measurability.

Given the strong effect of interscholastic athletics that Coleman (1960) identified as a source of school values (abstract), there may be situations in which children who are good at sports are more readily recognized on the axis of peer evaluation; however, this abstract does not permit a conclusion specific to swimming.

The conversion reading has indirect support

On the other hand, some evidence indicates that the routes through which evaluations are converted into opportunities in other settings are broader for cognitive skills close to logical thinking. Bourdieu (1986) wrote that educational credentials confer institutional recognition on cultural capital, make holders comparable and exchangeable, and set the conversion rate between cultural capital and economic capital. He also stated that, because the value of credentials also depends on their scarcity, the conversion rate itself can change.

Cognitive skills have a price in the labor market. Hanushek et al. (2015), using the adult skills survey (PIAAC) in 23 countries, showed that a one-standard-deviation increase in numeracy skills is associated with hourly wages that are 17.8% higher on average. That price differs greatly across countries: it was below 15% in 8 countries, including the Nordic countries, and about 28% in the United States. Returns to skills were lower in countries with higher union density, stricter employment protection, and larger public sectors. Even for the same skill, the price into which it is converted is determined by social institutions.

The price also changes over time. Autor, Levy, and Murnane (2003) showed that computerization reduced routine cognitive and manual tasks and increased nonroutine problem-solving and complex communication tasks. Deming (2017) showed that, from 1980 to 2012, jobs requiring high levels of social interaction grew by about 12 percentage points as a share of the U.S. labor force, whereas jobs that are math-intensive but less social shrank by 3.3 percentage points. Deming also showed that the labor market return to social skills was much greater in the 2000s than in the 1980s and 1990s (working paper version). The fact that “logical thinking is valued” in contemporary society is therefore not a fixed fact but the price of this period.

Studies of whether physical characteristics are converted into wages show the conversion route more directly. Case and Paxson (2008) showed, using four data sets from the United States and the United Kingdom, that the higher wages of taller people are explained by cognitive scores in childhood, and that taller adults select into occupations that require more cognitive skills and fewer physical skills (abstract of the working paper version). The physical advantage was converted into wages not directly but through cognitive skills.

Sports experience also has conversion routes

This does not mean that sports experience is never converted into later opportunities. Persico, Postlewaite, and Silverman (2004) showed, for white men in the United States, that height at age 16, rather than adult height, predicts wages, and that about half of this difference is explained by differences in participation in high school athletics and clubs (working paper version). After controlling for family background and other factors, participation in high school athletics was associated with adult wages that were 11.7% higher (same source). The authors cautioned that, because participation is the individual’s choice, it cannot be concluded that making participation mandatory would raise wages. Stevenson (2010), using the rapid expansion of girls’ high school sports under Title IX in the United States, estimated that a 10-percentage-point increase in girls’ sports participation in a state raised girls’ college attendance by 1 percentage point and their labor force participation by 1 to 2 percentage points (abstract of the working paper version). In contrast, according to the abstract of Barron, Ewing, and Waddell (2000), evidence that athletic participation directly affects wages and educational attainment was only partial, and much of the effect reflected individual differences in ability and in preferences for leisure.

What is converted in these studies is not swimming times themselves but participation in school sports. This can be read as indicating that sports evaluations are difficult to convert into other settings as competitive results, but that routes for conversion exist as the experience of participation.

A theory in which rewards shift toward what can be measured

A theory in which measurability affects rewards does exist for another setting. Holmström and Milgrom (1991) used as an example the argument that basing teachers’ pay on students’ test scores shifts effort toward teaching measurable basic skills at the expense of developing curiosity and creative thinking. Using this example, they showed that jobs should be designed so that tasks that are easier to measure carry stronger incentives. However, this is a theory of how employers reward workers’ effort, and it does not address how children’s traits are recognized. It can be used only indirectly for the assessment of P3.

Assessment of P3

The measurability reading is contradicted at the childhood stage. The conversion reading is consistent with evidence that cognitive skills are converted into credentials and wages and with evidence that physical advantages are converted into wages through cognition, but we found no study that compares logical thinking with swimming ability in the same study. Because sports experience also has conversion routes, it cannot be said that “evaluations of swimming are not converted into other opportunities.”

P4: Does early recognition accumulate?

The process by which early evaluations produce later advantages has itself been confirmed in experiments and quasi-experiments (see Do Those Who Have Receive More? Mechanisms, Causal Identification, and Limits of Cumulative Advantage and the Matthew Effect in the Academic Literature for details). Merton (1968) named the Matthew effect: for the same contribution, large recognition accrues to scientists who are already well known, whereas recognition is less readily given to scientists who are not yet known. van de Rijt et al. (2014) showed that, for projects that received a randomly assigned donation, subsequent donations increased from an average of 1.11 in the control group to 2.49. Bol et al. (2018) showed that, even with nearly identical review scores, early-career researchers just above the funding threshold obtained more than twice as much research funding over the following 8 years as those just below it.

Some studies identify the route from the evaluation of traits to accumulation. Musch and Grondin (2001) summarized that relatively older players within an age group are more likely to be selected for selected squads, that once selected they gain good coaching and strong competitors, and that competing at a higher level itself increases motivation. According to the abstract of Hancock, Adler, and Côté (2013), the authors proposed a theoretical model that explains relative age effects through the Matthew effect operating through parents, the Pygmalion effect operating through coaches (expectations elicit performance), and the Galatea effect operating through the athletes themselves (expectations for oneself elicit performance). At school, Campbell (2014) showed, among 5,481 seven-year-old children in the United Kingdom, that in-class ability grouping strengthens the tendency of teachers to judge relatively older children born early in the school year (autumn) as more able. Papageorge, Gershenson, and Kang (2020), using the difference between the expectations that two teachers held for the same student, estimated the elasticity of the probability of completing a four-year college degree with respect to teacher expectations at about 0.12 (abstract of the working paper version). The abstract of Caspi, Elder, and Bem (1987) states that childhood behavioral styles continued into adult occupational and family life through the accumulation of their own consequences (cumulative continuity) and through continuing to evoke the same reactions from others (interactional continuity).

On the other hand, some evidence shows that early recognition does not accumulate. Jussim and Harber (2005), reviewing 35 years of research, concluded that self-fulfilling prophecies occur in classrooms, but that their effects are typically small, do not accumulate greatly across evaluators or over time, and are more likely to dissipate than to accumulate. Black, Devereux, and Salvanes (2011), using data on the entire population of Norway, reported that the short-term positive effect of starting school younger on earnings largely disappeared by age 30. However, differences within school do not disappear quickly. According to the abstract of Bedard and Dhuey (2006), in OECD countries the youngest children in a grade scored 4 to 12 percentiles lower in fourth grade and 2 to 9 percentiles lower in eighth grade than the oldest children. The relative age effect in swimming also weakened from around ages 17 to 18 and disappeared at the elite level (Lorenzo-Calvo et al. 2021). The abstract of DiPrete and Eirich (2006) defines cumulative advantage as a general mechanism in which an advantageous relative position becomes a resource that produces further relative gains, and states that the term has come to have several meanings in the sociological literature. That the process exists and that the difference widens over time are separate claims that must be tested separately (Do Those Who Have Receive More? Mechanisms, Causal Identification, and Limits of Cumulative Advantage and the Matthew Effect in the Academic Literature).

P4 is partly supported. The process exists, but whether it continues to widen the difference depends on how much selection allocates opportunities, how much an evaluation is converted into other settings, and whether there are later opportunities to catch up.

Provisional assessment of novelty

We made this assessment after completing the collection and verification. Because confirming novelty is a task for the paper-writing process, the assessment here is provisional.

The parts that already exist in prior research are as follows.

• The process by which evaluations produce initial advantages and the process by which these advantages accumulate: the Matthew effect of Merton (1968) and the review by DiPrete and Eirich (2006).
• Attributes that the individual did not choose becoming, through evaluators, the starting point of accumulation: the model of relative age effects by Hancock et al. (2013) (the attribute is birth month) and Musch and Grondin (2001).
• Social values entering into whether a person is regarded as gifted: Sternberg and Zhang (1995) and Sternberg (1995).
• The fit between temperament and environmental demands affecting evaluation: Lerner et al. (1985).
• The value of what is acquired in the family being determined by its distance from the demands of the school market and becoming either an advantage of being ahead or a loss, and educational credentials setting the conversion rate between cultural and economic capital: Bourdieu (1986).

The parts we did not find in this survey are as follows.

• The chained claim that, for an individual’s chance traits rather than social class, whether the domain of the trait fits society’s evaluative criteria divides early evaluations, which become the starting point of cumulative advantage. Bourdieu (1986) wrote about the culture of family class, and Hancock et al. (2013) wrote about birth month within the same domain; neither compares across domains.
• The alternative account that explains differences in advantage between domains not by measurability but by the degree to which evaluations are converted into opportunities in other settings. The idea of conversion is found in Bourdieu (1986), but we did not find it in the form of a proposition contrasted with measurability.

The search covered title and abstract searches in OpenAlex (0 results for the combination of “cumulative advantage” and “socially valued,” and 62 results checked for the combination of “cumulative advantage” and giftedness), Crossref, Europe PMC, and general searches. For Japanese-language sources (Hiroshi Takeuchi’s discussion of the social construction of ability, Yuki Honda, Takehiko Kariya, and Takahiro Nakamura), we were unable to check the full texts. The possibility of having overlooked relevant work remains. Within this range, we found no prior research for the chained form of the claim or for the conversion account, and we therefore provisionally judged that the claim may be novel.

In a later audit in the paper-writing process (2026-11), this provisional assessment was overturned: the claim was judged not novel because it can be stated with Bourdieu’s (1986) conversion rate, Bol et al. (2018), and Merton (1968).

Gaps (unresolved issues)

• We found no study that compares logical thinking and athletic ability in the same group of children, following them from early evaluations through later opportunities to adult outcomes.
• Within the range of this survey, there was no study of whether sports evaluations, as competitive results (rather than participation), are converted into later credentials or wages.
• We did not collect Japanese-language empirical studies showing what academic and athletic evaluations are each converted into under the institutions of Japanese schools and school club activities.
• We also found no study that compares, across domains, indicators of the degree to which evaluations are converted (for example, how well a given evaluation predicts educational paths or wages).
• Changes over time in the wage premium for noncognitive traits (administrative data from Sweden and Finland) and the finding that teachers’ effects on grades and behavior correlate only weakly with their effects on test scores are discussed in Where Has Research on Non-Cognitive Skills Moved? Trends in Naming, Wage Returns, Interventions, and Measurement (2015–2026).

Unverified items

• Twelve sources were used within the range of their abstracts because their full texts could not be obtained. These sources are Sternberg and Zhang 1995; Lerner et al. 1985; Chen et al. 2005; Liu et al. 2015; Coleman 1960; Cobley et al. 2009; Cobley et al. 2019; Bedard and Dhuey 2006; Barron et al. 2000; Hancock et al. 2013; Caspi et al. 1987; and DiPrete and Eirich 2006. In the text, all of them are attributed to their authors with phrases such as “according to the abstract.”
• Sternberg (1995) was read on the summary page of the research report, and the full text of the report (PDF) could not be obtained.

References

All URLs were accessed on 2026-10-09.

Trait differences and evaluation institutions

• Polderman, T. J. C., Benyamin, B., de Leeuw, C. A., Sullivan, P. F., van Bochoven, A., Visscher, P. M., & Posthuma, D. (2015). Meta-analysis of the heritability of human traits based on fifty years of twin studies. Nature Genetics, 47(7), 702–709. https://doi.org/10.1038/ng.3285
• Rimfeld, K., Krapohl, E., Trzaskowski, M., Coleman, J. R. I., Selzam, S., Dale, P. S., Esko, T., Metspalu, A., & Plomin, R. (2018). Genetic influence on social outcomes during and after the Soviet era in Estonia. Nature Human Behaviour, 2(4), 269–275. https://doi.org/10.1038/s41562-018-0332-5
• Branigan, A. R., McCallum, K. J., & Freese, J. (2013). Variation in the heritability of educational attainment: An international meta-analysis. Social Forces, 92(1), 109–140. https://doi.org/10.1093/sf/sot076

Definitions of giftedness, fit, and evaluators

• Sternberg, R. J., & Zhang, L.-F. (1995). What do we mean by giftedness? A pentagonal implicit theory. Gifted Child Quarterly, 39(2), 88–94. https://doi.org/10.1177/001698629503900205
• Sternberg, R. J. (1995). A triarchic approach to giftedness (Research Monograph 95126). The National Research Center on the Gifted and Talented, University of Connecticut. https://nrcgt.uconn.edu/research-based_resources/sternber/
• Lerner, J. V., Lerner, R. M., & Zabski, S. (1985). Temperament and elementary school children’s actual and rated academic performance: A test of a ‘goodness-of-fit’ model. Journal of Child Psychology and Psychiatry, 26(1), 125–136. https://doi.org/10.1111/j.1469-7610.1985.tb01633.x
• Chen, X., Cen, G., Li, D., & He, Y. (2005). Social functioning and adjustment in Chinese children: The imprint of historical time. Child Development, 76(1), 182–195. https://doi.org/10.1111/j.1467-8624.2005.00838.x
• Liu, J., Chen, X., Coplan, R. J., Ding, X., Zarbatany, L., & Ellis, W. (2015). Shyness and unsociability and their relations with adjustment in Chinese and Canadian children. Journal of Cross-Cultural Psychology, 46(3), 371–386. https://doi.org/10.1177/0022022114567537
• Cornwell, C., Mustard, D. B., & Van Parys, J. (2013). Noncognitive skills and the gender disparities in test scores and teacher assessments: Evidence from primary school. Journal of Human Resources, 48(1), 236–264. https://doi.org/10.3368/jhr.48.1.236
• Campbell, T. (2014). Stratified at seven: In-class ability grouping and the relative age effect. British Educational Research Journal, 40(5), 749–771. https://doi.org/10.1002/berj.3127
• Mann, D. L., & van Ginneken, P. J. (2017). Age-ordered shirt numbering reduces the selection bias associated with the relative age effect. Journal of Sports Sciences, 35(8), 784–790. https://doi.org/10.1080/02640414.2016.1189588
• Christensen, M. K. (2009). “An eye for talent”: Talent identification and the “practical sense” of top-level soccer coaches. Sociology of Sport Journal, 26(3), 365–382. https://doi.org/10.1123/ssj.26.3.365
• Coleman, J. S. (1960). The adolescent subculture and academic achievement. American Journal of Sociology, 65(4), 337–347. https://doi.org/10.1086/222722
• Bourdieu, P. (1986). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241–258). Greenwood. https://www.marxists.org/reference/subject/philosophy/works/fr/bourdieu-forms-capital.htm

Relative age effects and selection in sports

• Musch, J., & Grondin, S. (2001). Unequal competition as an impediment to personal development: A review of the relative age effect in sport. Developmental Review, 21(2), 147–167. https://doi.org/10.1006/drev.2000.0516
• Cobley, S., Baker, J., Wattie, N., & McKenna, J. (2009). Annual age-grouping and athlete development: A meta-analytical review of relative age effects in sport. Sports Medicine, 39(3), 235–256. https://doi.org/10.2165/00007256-200939030-00005
• Cobley, S., Abbott, S., Eisenhuth, J., Salter, J., McGregor, D., & Romann, M. (2019). Removing relative age effects from youth swimming: The development and testing of corrective adjustment procedures. Journal of Science and Medicine in Sport, 22(6), 735–740. https://doi.org/10.1016/j.jsams.2018.12.013
• Lorenzo-Calvo, J., de la Rubia, A., Mon-López, D., Hontoria-Galán, M., Marquina, M., & Veiga, S. (2021). Prevalence and impact of the relative age effect on competition performance in swimming: A systematic review. International Journal of Environmental Research and Public Health, 18(20), 10561. https://doi.org/10.3390/ijerph182010561
• Hancock, D. J., Adler, A. L., & Côté, J. (2013). A proposed theoretical model to explain relative age effects in sport. European Journal of Sport Science, 13(6), 630–637. https://doi.org/10.1080/17461391.2013.775352
• Bedard, K., & Dhuey, E. (2006). The persistence of early childhood maturity: International evidence of long-run age effects. Quarterly Journal of Economics, 121(4), 1437–1472. https://doi.org/10.1162/qjec.121.4.1437
• Black, S. E., Devereux, P. J., & Salvanes, K. G. (2011). Too young to leave the nest? The effects of school starting age. Review of Economics and Statistics, 93(2), 455–467. https://doi.org/10.1162/rest_a_00081

The price of skills and the conversion of evaluations

• Hanushek, E. A., Schwerdt, G., Wiederhold, S., & Woessmann, L. (2015). Returns to skills around the world: Evidence from PIAAC. European Economic Review, 73, 103–130. https://doi.org/10.1016/j.euroecorev.2014.10.006
• Autor, D. H., Levy, F., & Murnane, R. J. (2003). The skill content of recent technological change: An empirical exploration. Quarterly Journal of Economics, 118(4), 1279–1333. https://doi.org/10.1162/003355303322552801
• Deming, D. J. (2017). The growing importance of social skills in the labor market. Quarterly Journal of Economics, 132(4), 1593–1640. https://doi.org/10.1093/qje/qjx022
• Holmström, B., & Milgrom, P. (1991). Multitask principal-agent analyses: Incentive contracts, asset ownership, and job design. Journal of Law, Economics, & Organization, 7(special issue), 24–52. https://doi.org/10.1093/jleo/7.special_issue.24
• Case, A., & Paxson, C. (2008). Stature and status: Height, ability, and labor market outcomes. Journal of Political Economy, 116(3), 499–532. https://doi.org/10.1086/589524
• Persico, N., Postlewaite, A., & Silverman, D. (2004). The effect of adolescent experience on labor market outcomes: The case of height. Journal of Political Economy, 112(5), 1019–1053. https://doi.org/10.1086/422566
• Stevenson, B. (2010). Beyond the classroom: Using Title IX to measure the return to high school sports. Review of Economics and Statistics, 92(2), 284–301. https://doi.org/10.1162/rest.2010.11623
• Barron, J. M., Ewing, B. T., & Waddell, G. R. (2000). The effects of high school athletic participation on education and labor market outcomes. Review of Economics and Statistics, 82(3), 409–421. https://doi.org/10.1162/003465300558902

Accumulation from early evaluations

• Merton, R. K. (1968). The Matthew effect in science. Science, 159(3810), 56–63. https://doi.org/10.1126/science.159.3810.56
• DiPrete, T. A., & Eirich, G. M. (2006). Cumulative advantage as a mechanism for inequality: A review of theoretical and empirical developments. Annual Review of Sociology, 32, 271–297. https://doi.org/10.1146/annurev.soc.32.061604.123127
• van de Rijt, A., Kang, S. M., Restivo, M., & Patil, A. (2014). Field experiments of success-breeds-success dynamics. Proceedings of the National Academy of Sciences, 111(19), 6934–6939. https://doi.org/10.1073/pnas.1316836111
• Bol, T., de Vaan, M., & van de Rijt, A. (2018). The Matthew effect in science funding. Proceedings of the National Academy of Sciences, 115(19), 4887–4890. https://doi.org/10.1073/pnas.1719557115
• Jussim, L., & Harber, K. D. (2005). Teacher expectations and self-fulfilling prophecies: Knowns and unknowns, resolved and unresolved controversies. Personality and Social Psychology Review, 9(2), 131–155. https://doi.org/10.1207/s15327957pspr0902_3
• Papageorge, N. W., Gershenson, S., & Kang, K. M. (2020). Teacher expectations matter. Review of Economics and Statistics, 102(2), 234–251. https://doi.org/10.1162/rest_a_00838
• Caspi, A., Elder, G. H., Jr., & Bem, D. J. (1987). Moving against the world: Life-course patterns of explosive children. Developmental Psychology, 23(2), 308–313. https://doi.org/10.1037/0012-1649.23.2.308

Footnotes

1. This idea derives from the clinical research of Thomas and Chess. Because the original sources are books, we used only the empirical papers here. ↩