Scand J Med Sci Sports 2010: 20: 509–515 & 2009 John Wiley & Sons A/S

doi: 10.1111/j.1600-0838.2009.00979.x

Relative age effect in female sport: a diachronic examination

of soccer players
N. Delorme1, J. Boiché2, M. Raspaud1
1
Laboratoire Sport et Environnement Social (SENS), Universite´ J. Fourier – Grenoble I, UFRAPS, Grenoble, France, 2Centre de
Recherche en Activite´s Physiques et Sportives (CURAPS), Universite´ de La Re´union, Le Tampon, France
Corresponding author: N. Delorme, Laboratoire Sport et Environnement Social (SENS), Universite´ J. Fourier – Grenoble I,
UFRAPS, BP 53, 38041 Grenoble cedex 9, France. Tel: 133 4 76 63 50 94, E-mail: [email protected]
Accepted for publication 30 April 2009

The relative age effect (RAE) is a biased distribution of elite whether relative age accounts for dropout from the activity.
athletes’ birthdates, with an overrepresentation of those The analyses revealed significant differences between the
born at the beginning of the competitive year and an under- expected and the observed distributions for all age cate-
representation of those born at the end. Despite an abundant gories. Furthermore, a significantly biased distribution of
literature, the impact of sex on this phenomenon remains dropout players’ birthdates was found for the o10, o14,
neglected by most researchers. This study investigated the and o17 categories. On the whole, dropout players are
whole sample of female soccer players affiliated to the underrepresented in Q1 and Q2, but overrepresented in Q3
French Soccer Federation for the 2006–2007 season and Q4. This study completes the literature on RAE among
(n 5 57 892). It first aimed at testing the presence of RAE females, but the inconsistency of the results calls for more
depending on age. Next, we looked at the birthdates of research on this population.
dropout players during the next season (n 5 15 285), to test

The relative age effect (RAE) is a phenomenon are more often selected to be a part of elite groups or
initially observed by Grondin et al. (1984), and to play within national youth teams. They benefit
consists in a biased distribution of elite players’ from this early exposition, which may bring them
birthdates. What is generally noticed is an over- determinant technical and strategic skills (Williams,
representation of athletes born at the beginning of 2000; Ward & Williams, 2003).
the competitive year but an underrepresentation of This differentiated access to high-level training and
people born by the end of it. It is considered to be competition is considered as a kind of discrimination,
derived from the age categories distinguished by because it disadvantages players born late after the
sport instances to organize and try to balance youth cutoff date, reducing drastically their chances to
competition. Most of the time, children and adoles- access to the elite (Musch & Hay, 1999; Simmons
cents born during two consecutive years are gathered & Paull, 2001; Edgar & O’Donoghue, 2005). Even if
in a common category. Thus, within a system using such discrimination seems involuntary, it needs to be
the first of January as a cutoff date, a child born in examined cautiously, given the profits that can be
January of the first year of a category will show a achieved through the practice of certain sports.
discrepancy of up to 23 months with a child born in
December of the second year. This difference in
relative age is likely to induce important variations RAE and dropout
in terms of physical (Tanner & Whitehouse, 1976)
and cognitive development (Bisanz et al., 1995), the In addition to partly conditioning the access to high-
discrepancies being particularly marked during pub- level practice, the differences of experience induced
erty (Delorme & Raspaud, 2009). The advantages by relative age are likely to be a factor of the dropout
regarding physical characteristics (i.e., in height, for young participants born late in the competitive
weight, or strength) can undoubtedly impact those year. Firstly, because of a less advanced state of
young players’ perceived potential in the activity physical development, they might experience a re-
(Helsen et al., 2005). Those born at the beginning duced time of play during official games, as suggested
of the competitive year are more easily identified as by the study of Vaeyens et al. (2005). Additionally,
‘‘talented’’ or ‘‘promising’’, and consequently, they relative age may be accompanied by divergences in

509
 16000838, 2010, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1600-0838.2009.00979.x by Universidade De La Coruña, Wiley Online Library on [18/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Delorme et al.
terms of self-perceptions. Indeed, players born at the was systematically observed for ice hockey (e.g.,
end of the competitive year may encounter more Barnsley et al., 1985; Barnsley & Thompson, 1988;
frequent situations of failure or inferiority, and such Grondin & Trudeau, 1991), which represents the
negative situations are likely to undermine these most popular activity in this country, whereas no
players’ feelings of competence and self-worth. effect was observed for volleyball (Grondin et al.,
Past literature in sport psychology reveals that 1984), a low-developed sport. Some authors also
reduced opportunities to play (Guillet et al., 2002), underlined the link between the strength of the
as well as low perceptions of sport competence RAE and the level of practice studied in a given
significantly contribute to the cessation of the prac- sport; the more the level increases, the stronger the
tice of team sports (Ommundsen & Vaglum, RAE. This correlation has particularly been shown
1991a, b; Guillet et al., 2006). Thus, the RAE tradi- in ice hockey and soccer (Musch & Grondin, 2001).
tionally observed among elite samples could also be
the result of more important rates of dropout among
players born late in the competitive year. In this vein, Factors of success in the activity and RAE
[Delorme & Raspaud, 2008] show that among male
The second main factor underlined by Musch and
basketball players aged between 9 and 15 years, and
Grondin (2001) is the prevalence of physical attri-
female players aged from 8 to 15 years, those born at
butes for performance in the activity. Most of the
the end of the year are overrepresented among
studies in which a RAE was observed concerned
dropout players, whereas those born early after the
activities in which height, weight, and strength re-
cutoff date are underrepresented. Such a biased
present key factors of success, like team sports
distribution in the birthdates of dropout players
involving physical contact such as ice hockey (e.g.,
also emerged among male soccer players aged from
Grondin & Trudeau, 1991), football (e.g., Daniel &
8 to 17 years [Delorme & Raspaud, 2008].
Janssen, 1987), soccer (e.g., Helsen et al., 2005), or
other individual activities, like tennis or swimming
(Baxter-Jones, 1995). Conversely, in other activities,
Competition and RAE the physical attributes cited can be seen as being
relatively independent from success, because other
Even if the role of the cutoff date chosen to create age
parameters, such as motor skills, are more prevalent,
categories among young participants is deemed cen-
or can even be seen as counterproductive, because
tral in the RAE phenomenon, it is not considered as
being short and thin is preferable for performance.
the only necessary condition for its presence. Instead,
This is the case, for example, for dance or gymnas-
in their review of the literature, Musch and Grondin
tics, which should therefore, not be prone to a RAE.
(2001) present it as the result of ‘‘a mixture of
Accordingly, Van Rossum (2006) did not find RAEs
physical, cognitive, emotional, and motivational
among Dutch dancers. This hypothesis was also
causes work together to produce the effect’’ (p. 159).
sustained by the work of Baxter-Jones and Helms
The first element they underline is competition. In-
(1996) and of Malina et al. (2004), who showed that a
deed, in the case of a low competition, any player can
late puberty was an advantage for young dancers and
belong to the team and may benefit from numerous
gymnasts. Besides, a positive correlation was found
opportunities to play, because there is a place for
between late puberty and elite gymnastics practice
everyone. As the authors advance, ‘‘the larger the
(Malina, 1994; Baxter-Jones et al., 1995).
pool of potential players, for a given sport in a given
category, the strongest the resulting RAE should be’’
(p. 154). In this vein, if the number of potential players Sex and RAE
is superior to the actual structures (i.e., number of
teams and number of players in a team) and infra- Musch and Grondin (2001) stated that stronger
structures (i.e., equipment available and time slots), RAEs should be observed in male youth sports,
there will be an accrued competition among them to compared with female samples. According to Bax-
belong to a team. Under these conditions, priority will ter-Jones (1995), such differences would result from a
be given to the more performing players, potentially complex interaction between the biological differ-
those born early in the competitive year, who benefit ences due to puberty and sociological factors. In
from a better perception from coaches regarding their fact, while considering the two factors presented
potential in the activity, compared with players born above, as sport represents a masculine area, competi-
later in the year (Helsen et al., 2005). tion is expected to be higher among boys than girls,
This said, it is logical to expect a RAE to appear in leading to a greater RAE (Musch & Grondin, 2001).
popular activities, which is subject to cultural dispa- Indeed, if an early physical development gives an
rities (i.e., the popularity of one sport varies from one important advantage to young sportsmen in many
country to another). For instance, in Canada, a RAE activities, it appears to be a disadvantage for young

510
 16000838, 2010, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1600-0838.2009.00979.x by Universidade De La Coruña, Wiley Online Library on [18/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
RAE in French female soccer
sportswomen, given the existing social pressures In this regard, the purpose of the present study was
around their body (Vincent & Glamser, 2006). This twofold. Investigating the female soccer players af-
stereotyped representation of the female body may filiated to the French Soccer Federation (FSF) for
lead early-developed girls to drop out from sports the 2006–2007 season, it first aimed at examining the
activities (Shakib, 2003), thus reducing the potential presence or absence of a RAE depending on age.
strength of the RAE. Finally, because of an earlier This was carried out by looking at every age category
puberty in girls and the larger variance in the devel- in a diachronic perspective. Next, we looked at the
opment of puberty in boys, at the moment of the birthdates of the players who had not reiterated their
selection process, there would be greater physical affiliation for the 2007–2008 season, in order to test
disparities due to relative age among boys, compared whether relative age accounts for female dropout
with girls (Baxter-Jones, 1995). from the activity.
However, in spite of an abundant literature on this According to the previous research as discussed
phenomenon in the sport setting, the impact of sex above, the two following hypotheses were formu-
remains neglected by most researchers (Musch & lated:
Grondin, 2001; Wattie et al., 2007). Studies concern-
(1) We expect a systematic RAE in all youth-age
ing female samples are not only scarce to date, but
categories but no effect in the adult category.
have also revealed inconsistent results. To our
(2) We expect an overrepresentation of late-born
knowledge, only five studies exist on the examination
children in dropouts and an underrepresentation
of RAE in female samples. Two of them concerned
of those who are born early in the competitive year
popular activities, i.e., where an important competi-
for each youth category. Concerning the adult
tion should enhance RAE, but where physical attri-
category, a uniform distribution of dropouts is
butes are considered as handicapping or at least not
expected.
advantageous, namely gymnastics and dance. Bax-
ter-Jones (1995), studying elite British gymnasts,
observed no RAE either in males or in females.
Van Rossum (2006) reached the same conclusions Materials and methods
with pre-professional Dutch dancers. Data collection
Conversely, the three other studies concerned team For the purpose of the present study, the birthdates of all
sports where a RAE was traditionally observed French female players affiliated to the FSF (n 5 57 892) during
among male athletes. First, Delorme et al. (2009), the 2006–2007 season were collected from the federation
investigating the French elite championships in team database. Among this sample the FSF also communicated
the birthdates of the players who did not reiterate their
sports, report no RAE for female soccer, basketball, affiliation for the 2007–2008 season (n 5 15 285). This infor-
and handball players, whereas several significant mation was communicated after the 2007–2008 season was
effects emerged for male players in ice hockey and over so as to avoid considering dropout players who had taken
trends were also observed for male rugby union and their licenses late during the season. Among female players,
handball. In the same vein, Vincent and Glamser the FSF distinguished six age categories: ‘‘o8 years,’’ ‘‘o10
years,’’ ‘‘o12 years,’’ ‘‘o14 years,’’ ‘‘o17 years,’’ and
(2006), analyzing a sample of 17-year-old American ‘‘adults.’’
soccer players, report a more systematic and stronger
RAE for males, compared with females. Finally,
Delorme and Raspaud (2009) observed a significant Data analysis
RAE in all youth categories of French basketball In the past literature, the presence of a RAE was determined
players, aged from 7 to 17 years. The effects were by showing a significant difference between the expected
even more pronounced for girls compared with boys. number of players born by month or quarter (i.e., three
These results suggest that among female adult, elite consecutive months period) and the observed distribution.
For each of the six categories presented, the players’ birthdates
samples, there is actually no or weak RAEs. How- were classified into four quarters. Since the cutoff date used to
ever, the only study conducted on regular young form age categories has been modified by the FSF (Jullien et
sport participants reveals a significant effect for both al., 2008), the players born before 1982 were classified from Q1
girls and boys. (August–October) to Q4 (May–July) and the players born in
1982 and after were classified from Q1 (January–March) to Q4
(October–December).
A chi-square goodness-of-fit test was then performed,
Purpose of the study and hypotheses comparing the expected and the observed distribution of
birthdates. In line with previous research, for each age
Given these contradictory results and the reduced category, the expected distribution was calculated based on
number of studies having investigated female athletes the national birth statistics by month and year for females,
using weighted mean scores. These data were obtained
in the past, it seems crucial to look at original through the National Institute of Economical Statistics and
samples of this population in order to better under- Studies. Regarding dropout players, the expected distribution
stand the impact of sex on the occurrence of RAE. considered was based on the birthdates observed for the whole

511
 16000838, 2010, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1600-0838.2009.00979.x by Universidade De La Coruña, Wiley Online Library on [18/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Delorme et al.
Table 1. Distribution of French female soccer players (2006–2007)

Category Q1 Q2 Q3 Q4 Total w2 P

Adults 5425 5905 5748 5686 22 764 62.14 o0.0001

(D) ( 168) (1434) (191) ( 357)
Under 17 2154 2338 2208 2084 8784 9.54 o0.05
(D) (126) (1106) ( 54) ( 78)
Under 14 1450 1478 1429 1259 5616 25.50 o0.0001
(D) (1114) (145) ( 18) ( 141)
Under 12 1860 2058 1968 1888 7774 8.72 o0.05
(D) (18) (1102) ( 38) ( 72)
Under 10 1938 2023 1896 1663 7520 41.35 o0.0001
(D) (1133) (1118) ( 42) ( 209)
Under 8 1495 1446 1331 1162 5434 64.58 o0.0001
(D) (1190) (182) ( 78) ( 195)
Total U8–U17 8897 9343 8832 8056 35 128 103.92 o0.0001
(D) (1472) (1415) ( 202) ( 685)

D is the difference between the observed distribution and the theoretical expected distribution.

Table 2. Dropout in French female soccer (2006–2007) Table 3. Distribution of dropout in French female soccer (2006–2007)

Category Total Dropout % Category Q1 Q2 Q3 Q4 Total w2 P

Adults 22 764 6743 29.62 Adults 1547 1747 1770 1679 6743 4.92 o0.18
Under 17 8784 2404 27.37 (D) ( 60) ( 2) (167) ( 5)
Under 14 5616 1689 30.07 Under 17 568 612 679 545 2404 12.33 o0.01
Under 12 7774 2159 27.77 (D) ( 21) ( 28) (174) ( 25)
Under 10 7520 1686 22.42 Under 14 412 414 441 422 1689 8.72 o0.05
Under 8 5434 604 11.12 (D) ( 24) ( 31) (111) (144)
Total 57 892 15 285 26.40 Under 12 536 579 557 487 2159 3.66 o0.30
(D) (119) (18) (110) ( 38)
Under 10 402 458 395 431 1686 13.71 o0.01
(D) ( 33) (14) ( 30) (158)
population of female licensed soccer players by month and Under 8 176 168 146 114 604 2.70 o0.44
year, using weighted mean scores. For this type of analysis, (D) (110) (17) ( 2) ( 15)
this procedure is deemed more adequate than the use of Total 2094 2231 2218 1999 8542 3.05 o0.38
national standards for comparison [Delorme & Raspaud, U8-U17
2008]. (D) ( 4) ( 14) (164) ( 46)

D is the difference between the observed distribution and the theoretical

expected distribution.

Results
Table 3 presents the birthdate distribution by
Table 1 presents the birthdates’ distribution by quarter of dropout players for the 2006–2007 season
quarter for each age category identified for female for each age category.
players by the FSF, during the 2006–2007 season. A significantly biased distribution was found for
The analyses reveal statistical differences between the o10, o14, and o17 categories. As a whole,
expected and observed distributions for all age cate- the dropout players were underrepresented in Q1
gories. For all youth categories, the results reflect a and Q2, but more often overrepresented in Q3 and
classical RAE with an overrepresentation of players Q4. Conversely, the birthdate distribution was dis-
born in Q1 and Q2, and an underrepresentation of tributed as expected for the o8, o12, and adult
players born in Q3 and Q4. For the ‘‘adult’’ category, categories.
however, the asymmetry observed was derived from
this classical pattern; players born in Q2 and Q3 were Discussion
overrepresented, but those born in Q1 and Q4 were
underrepresented. A RAE has traditionally been observed among male
Table 2 presents the rates of dropout for each age athletes of the elite level in sports where physical
category for the 2006–2007 season. attributes such as weight, height, and strength repre-
The dropout rate increased in the first age cate- sent key factors for success. This phenomenon can
gories and from the ‘‘o14 years’’ and it approxi- thus be seen as the consequence of a selection bias
mated 30%. The general turnover rate was equal favoring players born at the beginning of the com-
to 26.4%. petitive year and potentially higher rates of dropout

512
 16000838, 2010, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1600-0838.2009.00979.x by Universidade De La Coruña, Wiley Online Library on [18/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
RAE in French female soccer
among players born at the end. Very little is known among whom the elevation of the level of practice
about RAE among female athletes, and the handful was found to enhance the magnitude of the RAE
of studies that investigated this population report (Musch & Grondin, 2001), it appears that among
heterogeneous results. This research thus aimed at females, the higher the level, the lower the RAE is. A
exploring the French female soccer players for the social explanation seems necessary to justify this
2006–2007 season. We studied their birthdate dis- apparently paradoxical result.
tribution compared with the national standards. As suggested by Vincent and Glamser (2006),
Moreover, we distinguished a sample of dropout social pressures that encourage adolescents to con-
players in order to examine whether their quarter form to gender-based stereotypes could prevent fe-
of birth could account for this behavior. males from achieving excellence in competitive sport,
especially early maturing females. The physical char-
acteristics needed for athletic accomplishments are
A systematic RAE sometimes opposite to the representation of the ideal
Because a high competition among players and the female body, which should be thin and tiny (Choi,
prevalence of physical attributes for performance 2000). Shakib (2003) noted that the consequence of
have been pointed out as necessary conditions for a such an opposition may be for some girls to drop out
RAE to occur (Musch & Grondin, 2001), we ex- from sports at some point during puberty. Thus, such
pected a significant RAE to emerge among youth role conflicts could lead outstanding female players
categories, as was observed for French basketball to drop out from soccer or to renounce to an elite
players by Delorme and Raspaud (2009). Conversely, practice. Because of the physical differences due to
because a very weak effect appeared for 17-year-old relative age, the pressures, and the associated embar-
American soccer players (Vincent & Glamser, 2006) rassment would be particularly high for players born
and no effect at all was observed for French soccer at the beginning of the year, which may explain why
players in the national championship (Delorme et al., the birthdate distribution reveals no RAE among
2009), we posited that there should be no significant elite players. In brief, if an early physical develop-
effect in the adult category. ment acts as an important advantage for young
These hypotheses were partially supported because males in many sports, it also acts as a socially
a significant distortion emerged between the ob- constructed disadvantage for young females that
served distribution by quarter and the national dis- could facilitate their dropout from sports activities.
tribution for all age categories, including adults. The Nevertheless, this interpretation remains specula-
distortion observed among youth players respects the tive, and more research is needed to specifically
traditional RAE, with players born in Q1 and Q2 examine whether, among the female players involved
overrepresented and players born in Q3 and Q4 at a good level of practice, there are actually more
underrepresented. This result echoes the results ob- dropouts among those born at the beginning of the
tained in basketball by Delorme and Raspaud (2009) competitive year compared with those born in the
and reinforces the idea that such a phenomenon last quarter.
particularly arises in popular team sports involving
physical contact. On the other hand, the significant
effect obtained for adults contrasts with the past A dual mechanism resulting in RAE
literature in this activity with female samples (Vin- The second purpose of the study was to examine
cent & Glamser, 2006; Delorme et al., 2009). Even if whether the birthdate distribution of female players
the pattern is less clear than in the youth categories, who dropped out from soccer would significantly
the main irregularities observed go in the sense of a differ from the values observed in the entire sample
RAE, with a very important overrepresentation of of licensed players. Because of the physical inferiority
players born in Q2 and a very marked underrepre- occasioned by relative age, and the potentially reduced
sentation of players born in Q4. opportunities to play and lowered self-perceptions, we
Moreover, it should be outlined that in the past hypothesized the dropout players to be overrepre-
two studies conducted on female soccer, the partici- sented in the last two quarters of the competitive
pants were elite players involved in the national year, and those born during the first two quarters to
championship, whereas the sample of the current be underrepresented. Because this phenomenon may
study includes players of all levels. Given the size of cease with physical growth, we set this hypothesis
this sample, which includes all the French female only for youth categories, and expected no significant
soccer players, one can reasonably think that it effect among adults.
represents an ‘‘average’’ level. In this vein, the This hypothesis was partially supported, since in
systematic RAE observed in this study disappears three age categories (o10, 14, and 17 years), the
in the French female elite level where Delorme et al. players born in the last 6 months of the competitive
(2009) did not find any RAE. Contrary to males year were overrepresented, whereas those born in the

513
 16000838, 2010, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1600-0838.2009.00979.x by Universidade De La Coruña, Wiley Online Library on [18/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Delorme et al.
first 6 months were underrepresented in the dropout from it and have fewer chances to be selected and
sample. The reverse asymmetry observed in the reach the elite.
global population (in fact, the classical RAE distri-
bution) would thus partly be due to this unequal Perspectives
distribution of dropouts. A fruitful perspective of
research would be to verify whether the factors cited Based on the whole population of the FSF players
earlier (e.g., opportunities to play and self-percep- for the 2006–2007 season, this study investigated,
tions) can actually be considered as mechanisms from a diachronic perspective, the RAE among
through which more important dropout rates are females. A significant RAE was observed for all
obtained among players born in Q3 and Q4. categories, suggesting that this effect seems to de-
It should also be underlined that in spite of a crease with the level among females, contrary to what
homogeneous distribution of dropout players in the was observed among males. Besides, the analysis of
o8 and o12 sub-groups, these categories were the birthdate distribution suggests that this phenom-
subject to a significant RAE. In other words, girls enon results from a dual mechanism: a higher rate of
born in Q1 and Q2 are more likely to begin playing dropout among players born late in the competitive
soccer compared with their younger counterparts. year, and a larger proportion of players born early in
Those born in Q3 and Q4, potentially because of the competitive year among beginners. These results
their less advantageous physical attributes, show a demonstrate the importance for RAE research to
kind of ‘‘self-selection’’ process before even trying the conduct more studies on female athletes, who remain
activity. The RAE can thus be seen as the result of neglected to date. The contradictory results obtained
two simultaneous phenomena. On the one hand, in the rare studies involving female participants show
children born at the end of a competitive year would that the current knowledge on the mechanisms
be less likely to undertake the practice of an activity explaining the presence or absence of RAE still needs
in which weight, height, or strength could be seen as to be improved.
relevant for performance. On the other hand, those
who begin to practice are more likely to drop out Key words: relative age effect, female, soccer, dropout.

References
Barnsley RH, Thompson AH. Birthdate Daniel TE, Janssen CTL. More on Guillet E, Sarrazin P, Carpenter PJ,
and success in minor hockey: the key to the relative age effect. Can Ass Trouilloud D, Cury F. Predicting
the NHL. Can J Behav Sci 1988: 20: Health, Phys Educ Recreat 1987: 53: persistence or withdrawal in female
167–176. 21–24. handballers with social exchange
Barnsley RH, Thompson AH, Barnsley Delorme N, Boiché J, Raspaud M. The theory. Int J Psychol 2002: 37: 92–104.
PE. Hockey success and birthdate: the relative age effect in elite sport: the Guillet E, Sarrazin P, Fontayne P,
RAE. Can Ass Health, Phys Educ French case. Res Q Exerc Sport 2009: Brustad RJ. Understanding female
Recreat 1985: 51: 23–28. 80: 336–344. sport attrition in a stereotypical male
Baxter-Jones A. Growth and Delorme N, Raspaud M. The relative age sport within the framework of Eccles’
development of young athletes. effect in French youth basketball: a expectancy-value model. Psychol
Should competition levels be age study on the whole population. Scand J Woman Q 2006: 30: 358–268.
related? Sports Med 1995: 20: Med Sci Sports 2009: 19: 235–242. Helsen WF, Van Winckel J, Williams
59–64. Delorme N, Raspaud M. Influence of the MA. The relative age effect in youth
Baxter-Jones A, Helms P. Effects of relative age effect on the dropout of soccer across Europe. J Sports Sci
training at young age: a review of the sports activities. Proceedings of the 2005: 23: 629–636.
Training of Young Athletes (TOYA) 13th Annual Congress of the ECSS: Jullien H, Turpin B, Carling C. Influence
study. Pediatr Exerc Sci 1996: 8: Estoril: European College of Sport de la date de naissance sur la carrière
310–327. Science, 2008: 72–73. professionnelle des joueurs de football
Baxter-Jones A, Helms P, Maffull N, Edgar S, O’Donoghue P. Season of birth français. Sci Sports 2008: 23: 149–155.
Baines-Preece J, Preece M. Growth and distribution of elite tennis players. J Malina RM. Physical growth and
development of male gymnasts, Sports Sci 2005: 23: 1013–1020. biological maturation of young
swimmers, soccer and tennis players: a Grondin S, Deshaies P, Nault LP. athletes. Exerc Sport Sci Rev 1994: 22:
longitudinal study. Ann Hum Biol Trimestres de naissance et participation 389–434.
1995: 22: 381–394. au hockey et au volleyball. La Revue Malina RM, Bouchard C, Bar-Or O.
Bisanz J, Morrison F, Dunn M. Effects Québécoise de l’Activité Physique 1984: Growth, maturation, and physical
of age and schooling on the acquisi- 2: 97–103. activity. Champaign: Human Kinetics,
tion of elementary quantitative Grondin S, Trudeau F. Date de naissance 2004.
skills. Dev Psychol 1995: 31: et ligue nationale de hockey: analyses Musch J, Grondin S. Unequal
221–236. en fonction de différents paramètres. competition as an impediment to
Choi PYL. Femininity and the physically Revue des Sciences et Techniques des personal development: a review of the
active woman. London: Routledge, Activités Physiques et Sportives 1991: relative age effect in sport. Dev Rev
2000. 26: 37–45. 2001: 21: 147–167.

514
 16000838, 2010, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1600-0838.2009.00979.x by Universidade De La Coruña, Wiley Online Library on [18/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
RAE in French female soccer
Musch J, Hay R. The relative age effect in childhood throught puberty. Am Vincent J, Glamser FD. Gender
soccer: cross-cultural evidence for a Behav Sci 2003: 46: 1405–1422. differences in the relative age effect
systematic discrimination against Simmons C, Paull GC. Season-of-birth among US Olympic Development
children born late in the competition bias in association football. J Sports Sci Program youth soccer players. J Sports
year. Sociol Sport J 1999: 16: 54–64. 2001: 19: 677–686. Sci 2006: 24: 405–414.
Ommundsen Y, Vaglum P. The influence Tanner JM, Whitehouse RH. Clinical Ward P, Williams AM. Perceptual and
of attributional style on the soccer- longitudinal standards for cognitive skill development in soccer:
related self-esteem and persistence in height, weight, height velocity and the multidimensional nature of expert
soccer of young boys. Scand J Med Sci weight velocity and the stages of performance. J Sport Exerc Psychol
Sports 1991a: 1: 45–50. puberty. Arch Dis Child 1976: 51: 2003: 25: 93–111.
Ommundsen Y, Vaglum P. The influence 170–178. Wattie N, Baker J, Cobley S, Montelpare
of low perceived soccer and social Vaeyens R, Philippaerts RM, Malina WJ. A historical examination of
competence on later dropout from RM. The relative age effect in soccer: a relative age effects in Canadian hockey
soccer: a prospective study of young match-related perspective. J Sports Sci players. Int J Sport Psychol 2007: 38:
boys. Scand J Med Sci Sports 1991b: 2005: 23: 747–756. 178–186.
1: 180–188. Van Rossum JHA. Relative age effect Williams AM. Perceptual skill in soccer:
Shakib S. Female basketball revisited: findings from the dance implications for talent identification
participation. Negotiating the domain. Percept Mot Skills 2006: 102: and development. J Sports Sci 2000: 18:
conflation of peer status from 302–308. 737–740.

515