Content uploaded by Michael Romann
Author content
All content in this area was uploaded by Michael Romann on Jun 14, 2019
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=rjsp20
Journal of Sports Sciences
ISSN: 0264-0414 (Print) 1466-447X (Online) Journal homepage: http://www.tandfonline.com/loi/rjsp20
Relative age effects in Swiss talent development –
a nationwide analysis of all sports
Michael Romann, Roland Rössler, Marie Javet & Oliver Faude
To cite this article: Michael Romann, Roland Rössler, Marie Javet & Oliver Faude (2018): Relative
age effects in Swiss talent development – a nationwide analysis of all sports, Journal of Sports
Sciences, DOI: 10.1080/02640414.2018.1432964
To link to this article: https://doi.org/10.1080/02640414.2018.1432964
View supplementary material
Published online: 02 Feb 2018.
Submit your article to this journal
View related articles
View Crossmark data
Relative age effects in Swiss talent development –a nationwide analysis of all
sports
Michael Romann
a
, Roland Rössler
b,c
, Marie Javet
a
and Oliver Faude
c
a
Swiss Federal Institute of Sport Magglingen SFISM, Section for Elite Sport, Magglingen, Switzerland;
b
Amsterdam Collaboration on Health and
Safety in Sports, Department of Public and Occupational Health & Amsterdam Movement Sciences, VU University Medical Center, Amsterdam,
Netherlands;
c
Department of Sport, Exercise and Health, University of Basel, Basel, Switzerland
ABSTRACT
Relative age effects (RAE) generate consistent participation inequalities and selection biases in sports. The
study aimed to investigate RAE across all sports of the national Swiss talent development programme (STDP).
In this study, 18 859 youth athletes (female N = 5353; mean age: 14.8 ± 2.5 y and male N = 13 506; mean age:
14.4 ± 2.4 y) in 70 sports who participated in the 2014 competitive season were evaluated. The sample was
subdivided by sex and the national level selection (NLS, N = 2464). Odds ratios (ORs) of relative age quarters
(Q1-Q4) and 95% confidence intervals (CI) were calculated. In STDP, small RAE were evident for females (OR
1.35 (95%-CI 1.24, 1.47)) and males (OR 1.84 (95%-CI 1.74, 1.95)). RAE were similar in female NLS athletes (OR
1.30 (95%-CI 1.08, 1.57)) and larger in male NLS athletes (OR 2.40 (95%-CI 1.42, 1.97)) compared to athletes in
the lower selection level. In STDP, RAE are evident for both sexes in several sports with popular sports showing
higher RAE. RAE were larger in males than females. A higher selection level showed higher RAE only for males.
In Switzerland, talent identification and development should be considered as a long-term process.
ARTICLE HISTORY
Accepted 19 January 2018
KEYWORDS
Youth sports; talent
selection; athletic
development; maturation
Introduction
The phenomenon of relative age effects (RAE) is well-known in
youth sports. Children and adolescents are commonly pooled
into annual age groups to account for developmental differ-
ences and, thus, to allow for more equal inter-individual
opportunities for being successful in a particular sport. There
remains, however, a potential gap of up to 12 months in
chronological age between individuals. RAE is defined as the
overrepresentation of chronologically older participants within
one selection year relative to their chronologically younger
counterparts. This effect occurs during the early development
of youth athletes (Cobley, Baker, Wattie, & McKenna, 2009;
Musch & Grondin, 2001). RAE may lead to a biased view of
the potential of children in a particular sport as early-born
athletes may have advanced physical and cognitive abilities
compared to their late-born opponents and are, therefore,
more likely be identified as more talented (Cobley et al.,
2009; Delorme, Boiché, & Raspaud, 2010; Gil et al., 2014;
Hancock, Adler, & Côté, 2013; Wattie, Schorer, & Baker, 2015).
Consequently, these children may have a higher chance of
being selected for representative teams or talent centres and
may receive more comprehensive future training. It has been
shown in youth soccer and basketball players that those who
are born late in the selection year are more likely to drop out
of these sports than players who are born early in the
selection year (Delorme et al., 2010; Delorme, Chalabaev,
& Raspaud, 2011).
In a comprehensive meta-analytical review, Cobley et al. (2009)
reported a consistent risk for RAE, which is apparent across a
variety of different sports. These authors presented data on RAE
in 14 different sports (ice hockey, volleyball, basketball, American
football, Australian rules football, baseball, soccer, cricket, swim-
ming, tennis, gymnastics, netball, Rugby Union and golf) for male
and female athletes from 4 years of age to the senior professional
level. Most studies were conducted in soccer and ice hockey. The
largest RAE were found in basketball and soccer. RAE were present
in all age categories and increased with age until late adolescence.
In adult athletes, the RAE were lower, comparable to (pre)pubertal
children. Further, RAE were apparent in all levels of play. Whereas
the RAE were marginal at the lowest levels, they increased with
higher representative levels. However, on the elite level, the effect
size decreased to values of lower competitive levels. Meanwhile,
data on RAE in other sports, such as alpine skiing (Müller,
Hildebrandt, & Raschner, 2015), handball (Schorer, Cobley, Busch,
Brautigam, & Baker, 2009) and athletic sprinting (Romann
&Cobley,2015) are available, confirming the presence of RAE in
thesesportsaswell.
Data on RAE in the above-mentioned meta-analysis were,
however, not consistent between studies, and evidence for
some sports is based on only small samples. Particularly, for
female athletes, the number of available studies is limited and
CONTACT Michael Romann michael.romann@baspo.admin.ch Swiss Federal Institute of Sport Magglingen (SFISM), Alpenstrasse 18, Magglingen
CH-2532, Switzerland
Co-Authors: Roland Rössler, Department of Sport, Exercise and Health, University of Basel, CH-4052 Basel, Switzerland. roland.roessler@unibas.ch
Marie Javet, Swiss Federal Institute of Sport Magglingen (SFISM), CH-2532 Magglingen, Switzerland. marie.javet@baspo.admin.ch
Oliver Faude, Department of Sport, Exercise and Health, University of Basel, CH-4052 Basel, Switzerland. oliver.faude@unibas.ch
Supplemental data for this article can be accessed here
JOURNAL OF SPORTS SCIENCES, 2018
https://doi.org/10.1080/02640414.2018.1432964
© 2018 Informa UK Limited, trading as Taylor & Francis Group
data are inconsistent. For instance, Delorme et al. (2010)found
RAEinFrenchfemalesoccerplayers,particularlyintheyouthage
categories. Interestingly, there was also an inverse RAE in drop-
outs who were overrepresented in the second half of the selec-
tion year and underrepresented in the first half. In contrast,
Vincent and Glamser (2006) observed marginal RAE among
female youth soccer players, whereas in their male counterparts
the RAE was clearly larger. The authors discussed complex inter-
actions of biological and maturational differences with socializa-
tion influences as possible reasons for these sex differences.
Goldschmied (2011) reported no RAE in elite level female soccer,
basketball and handball players. This report, however, was not
published in a peer-reviewed scientific journal, is based on a
limited number of players and contains data from three different
countries and seasons. Overall, the inconsistent evidence on RAE
in females warrants further research in this population (Delorme
et al., 2010).
Methodological confounding through differences in the talent
identification and development systems between countries or
country-specific popularity of a particular sport may also affect
results. Cohort studies analysing large homogeneous samples of
youth athletes over a variety of different sports from a single
country are rare. In most studies with large nationwide samples,
only a limited number of different sports (in most instances a single
sport) were analysed (Delorme et al., 2010,2011;Romann&
Fuchslocher, 2013). It has been suggested that the popularity of
a particular sport, the number of active participants, the impor-
tance of physical development and the competitive level affect the
existence of a RAE (Musch & Grondin, 2001). There is evidence
supporting these hypotheses. For example, Delorme and Raspaud
(2009) showed that no RAE is present in shooting sports, i.e. a sport
in which physical capabilities areofminorrelevanceforsports
performance. Similarly, it has been shown that in sports with
weight categories, RAE are not present (Albuquerque et al., 2012;
Delorme, 2014). Weight categories may counterbalance matura-
tional and physical differences between young athletes within age
categories and, thus, may prevent the occurrence of RAE.
Altogether, important questions remain, for instance, whether
RAE are indeed consistent through all sports or whether there
are inverse RAE in some sports as late-born talented age-groupers
might tend to change to specific sports with fewer competitive
demands and less rivalry. The analysis of RAE within a national
talent development programme including all selected youth ath-
letes of all organised sports, may facilitate to answer such ques-
tions within a homogenous talent development context.
The aim of the study was to investigate the RAE in all
organized sports based on a countrywide Swiss database
where youth athletes (7 to 20 years of age), who were selected
into a nationwide talent programme, were registered. It was
hypothesized that overall RAE are apparent in sports with high
rates of participation and high selection pressure (e.g. soccer,
alpine skiing), whereas sports with less public attention and
fewer participants (e.g. fencing, curling) do not necessarily
show RAE or even inverse RAE. As Olympic sports are usually
more popular, include more participants and show a high
competitive level already in young athletes, RAE should be
more pronounced in Olympic as compared to Non-Olympic
sports. Furthermore, it was expected that there are more
pronounced RAE in male as compared to female youth
athletes and that RAE are less pronounced in sports with
weight categories.
Methods
Participants
The Swiss system of talent identification, selection and develop-
ment is based on three levels of performance (Figure 1)(Romann
& Fuchslocher, 2013). The first level is a nationwide extracurricu-
lar programme called “Jugend und Sport”(J + S), which is offered
to all young children and adolescents aged 5 to 20 years. The
second level is the national Swiss talent development pro-
gramme (STDP) within J + S in 70 different sports starting at
the age of 7 years. All Swiss youth athletes in the STDP (N = 18
859; female N = 5353; age: 14.8 ± 2.5 y and male N = 13 506; age:
14.4 ± 2.4 y) who participated in the 2014 competitive season
were included in the present analysis. Athletes in STDP are
selected by the national talent selection instrument (PISTE),
which includes six major assessment criteria (competition per-
formance, performance tests, performance development, psy-
chological factors, athlete’s biography and biological
development) and a number of sub-components (e.g. resilience,
anthropometry, achievement motivation) (Fuchslocher, Romann,
&Gulbin,2013). Licensed coaches perform the practices in this
programme, and the athletes are expected to train more than
400 hours per year. J + S and Swiss Olympic jointly established
this cut-off criterion. The third level is the subgroup of the
national level selection (NLS). In the PISTE process, athletes
with the potential to successfully perform in national and inter-
national competitions are selected in the NLS.
Procedures and data analysis
Anonymised information on participant’s age, sex, date of
birth and sport disciplines was retrieved from the database
of J + S (Swiss Federal Office of Sport, 2015). The study was
approved by the institutional ethical review board of the Swiss
Federal Institute of Sport.
In Switzerland, the cut-off date for all sports is January 1
st
.The
athletes were categorized into four relative age quarters (Q)
according to their birth month independently of birth year (i.e.,
Q1 = January to March; Q2 = April to June; Q3 = July to
September; and Q4 = October to December). The observed
birth-date distributions were calculated for every relative age
quarter. The expected birthdate distributions were obtained
from the actual corresponding distributions (1994–2009) as the
number of live births registered with the Swiss Federal Office of
Statistics. The relative age quarters of the Swiss population were
as follows: Q1 = 24.5%; Q2 = 25.2%; Q3 = 26.1%; and Q4 = 24.2%.
Sports were categorized into Olympic and Non-Olympic sports
and for clarity, into large (25 largest sports according to the
number of involved athletes for each sex separately; sports
with more than 43 selected athletes for females and more than
74 selected athletes for males on the STDP level) and small sports
(lower number of athletes in STDP). The reason for this categor-
isation is a higher participation, more funding and more scientific
support for Olympic sports (Swiss Olympic Association, 2017).
Data of the latter are presented in the supplementary files only.
2M. ROMANN ET AL.
Based on these data, odds ratios (OR) with 95% confidence
intervals (95% CI) were calculated between Q1/Q4 as com-
monly used in RAE studies (Cobley et al., 2009). A relevant
RAE was assumed if the confidence interval of the OR did not
include 1. The OR for the Q1 vs. Q4 comparison was inter-
preted as follows: OR < 1.22, 1.22 ≤OR < 1.86, 1.86 ≤OR <
3.00, and OR ≥3.00, indicating negligible, small, medium and
large effects, respectively (Olivier & Bell, 2013). If the OR was <
1 and the confidence interval did not include 1, this finding
was interpreted as an inverse RAE. As population-based data
were analysed, inferential statistics were not applied (Gibbs,
Shafer, & Dufur, 2012).
Results
In total, 18 859 youth athletes (5353, 28% girls) between 7 and
20 years of age were included in the STDP in the year 2014
(Figure 1). Thirteen percent (N = 2477) of these athletes (female,
N = 970; male, N = 1507) were included in the subgroup of NLS.
Tables 1 and 2show the relative age quarter distribution of
female and male youth athletes together with the odds ratios
for athletes born in the first quarter vs. last quarter of the
selection year for the 25 sports with the largest number of
participants in Switzerland. Detailed data for all sports
included in the STDP are shown in the appendices in the
online supplementary material (Tables SI and SII).
In female athletes in the STDP (Table 1), medium RAE in track
and field and synchronized swimming were observed. Small RAE
were present in tennis, volleyball, soccer and alpine skiing. No
relevant RAE were found in the NLS in female athletes.
In male athletes in the STDP (Table 2), track and field, soccer,
shooting, basketball, ice hockey, field hockey and volleyball
showed medium RAE. Small RAE were present in cross-country
skiing, alpine skiing, tennis, swimming, handball and floorball.
In athletes in the NLS, large RAE were found in tennis, rowing,
soccer, alpine skiing and ice hockey and medium effects in
basketball and handball.
When comparing small sports with the 25 biggest sports
for male youth athletes, on average medium RAE were found
in the 25 biggest sports for male athletes of the STDP and the
NLS, small RAE in small sports of the STDP and inverse RAE in
small sports in the NLS. The only single small sport with large
RAE was trampoline (OR = 9.92 (95%-CI 1.27, 77.50)). In female
youth athletes, no relevant RAE was present in small sports
(see online supplementary Tables SI and SII).
The ORs for selected athletes born in the first quarter vs. the
last quarter of the selection year are categorised for the STDP
and the NLS for Olympic and Non-Olympic sports separately in
Figure 2. Whereas there is medium RAE for Olympic sports in
male youth athletes with larger RAE in the NLS, female sports
and Non-Olympic sports merely showed small to negligible RAE.
As expected, no relevant RAE were found in sports with
weight categories (judo, karate, wrestling, boxing; female,
OR = 1.42 (95%-CI 0.87, 2.30); male, OR = 0.85 (95%-CI 0.64,
1.12)) in athletes in the STDP. In contrast, athletes in the NLS
showed medium RAE in these sports (females, OR = 4.46 (95%-
CI 0.96, 20.66); males, OR = 1.75 (95%-CI 0.84, 3.17)).
Discussion
In this study, data on RAE are presented in a population-based
approach comprised of all youth athletes who were registered
in the Swiss talent development programme (STDP) in the
year 2014. A large variability in RAE was found across different
types of sports and between girls and boys. The main findings
are (a) that in female athletes a small and in male athletes
medium overall RAE were present, (b) that in male athletes the
RAE were considerably larger in Olympic as compared to Non-
Olympic sports and (c) that in male athletes the RAE were
larger in athletes in the NLS as compared to the STDP. Small
sports showed negligible to small RAE. In sports with weight
categories, medium RAE were only present in the NLS athletes
of both sexes.
Key findings
To the best of our knowledge, this is the first study analysing a
complete sample of a nationwide talent development pro-
gramme including all organized sports. Medium RAE were
found over all 68 male sports in the STDP sample. The effect
was clearly larger in the higher selection levels (NLS). This
finding is supported by the literature, showing an increasing
risk for RAE with higher levels of competition (Cobley et al.,
2009). Several factors may increase the risk of RAE in a particular
sport. For instance, RAE might be affected by the sport’s popu-
larity, the amount/rate of participation, the level of competition,
higher selection pressure, early specialization and the expecta-
tions of coaches who are involved in the selection process
(Cobley et al., 2009; Hancock et al., 2013; Wattie et al., 2015).
Also, differences between late- and early-born athletes in cog-
nitive, social, physical and maturational development might
result in RAE (Musch & Grondin, 2001). For instance, Reed,
Figure 1. Overview of the different levels of selection in the Swiss organized
sport system (J + S), the Swiss talent development programme (STDP) and the
national level selection (NLS).
JOURNAL OF SPORTS SCIENCES 3
Parry, and Sandercock (2016) suggested that social agents may
contribute to RAE in English school sports. A recent study of
German youth national football teams found RAE, but there
were no relevant differences in anthropometric and perfor-
mance characteristics between players of different relative
age quarters (Skorski, Skorski, Faude, Hammes, & Meyer,
2016). Thus, physical or maturational differences might be of
minor relevance for RAE on the highest performance level in an
already highly selected population. An interesting finding is
that male athletes in some NLS in sports with low participation
rates showed inverse (not significant) RAE, i.e., a higher propor-
tion of athletes was born in the last quarter of the year as
compared to the first quarter of the year. It might be speculated
that talented athletes, who were born late in the year, move
from sports with high participation rates to those with lower
participation rates in order to have the possibility to compete
on a high performance level (Delorme, 2014).
In contrast to male sports, there were merely small RAE
over all 63 female sports and no relevant difference between
the STDP and the NLS. Previous data also revealed differences
in RAE between girls and boys. For instance, Cobley et al.
(2009) summarised the results of 38 studies published
between 1984 and 2007 and found a larger OR (Q1 vs. Q4)
of 1.65 (95%-CI 1.54, 1.77) in male athletes as compared to
female athletes (OR 1.21 (95%-CI 1.10, 1.33)). Similarly,
Raschner, Müller, and Hildebrandt (2012) observed a differ-
ence between males (OR 3.32) and females (OR 1.89) in more
than 1000 athletes participating in the Youth Olympic Games
2012. Recently, Reed et al. (2016) reported data from more
than 10 000 children participating at the London Youth Games
and also found lower RAE in many sports for girls as compared
to boys. Thus, these results are in line with the current
evidence. Females generally mature earlier than their male
peers and after peak height velocity differences in athletic
performance are reduced or disappear (Vincent & Glamser,
2006). A possible additional explanation might be that the
number of male athletes was 2.5 times the number of female
athletes in the STDP. Further, the proportion of NLS athletes
was about 11% boys and 18% girls. This reflects a larger pool
for selection and a higher selection pressure in male athletes
and may explain the difference in RAE between boys and girls.
A further explanation for the smaller RAE in female youth
athletes might be that changes in body shape, which are
associated with early maturation (e.g. greater body mass per
stature, shorter legs, wider hips, greater body fat) are disad-
vantageous for performance (Vincent & Glamser, 2006).
This study reveals that Non-Olympic sports showed a lower
risk for RAE than Olympic sports. It could be speculated that this
is due to the greater attractiveness of Olympic sports as a result
of their greater media presence and higher funding, whereas
Non-Olympic sports are less popular and, hence, attract fewer
young people (Fuchslocher et al., 2013; Swiss Olympic
Association, 2017). A greater attractiveness might lead to larger
pools of athletes in Olympic sports, which increases selection
pressure (Musch & Grondin, 2001). This view is strengthened by
the fact that only about 10% of the sample was involved in Non-
Olympic sports, and the remaining athletes performed in
Olympic sports. Interestingly, 12.4% of the Olympic sports ath-
letes were selected to the NLS, whereas 21.6% of Non-Olympic
athletes achieved the NLS, confirming the higher selection pres-
sure in Olympic sports. Further, a higher professionalism in
Olympic sports in general, as well as their talent selection instru-
ments might be another underlying reason for an increased risk
of RAE (Fuchslocher et al., 2013).
Table 1. Nationwide data of relative age effects in female youth athletes in the Swiss talent development programme (STDP) and in the national level selection
(NLS) in the year 2014 for the 25 largest sports.
STDP NLS
Sport N Q1 (%) Q2 (%) Q3 (%) Q4 (%)
OR Q1 vs. Q4
(95% CI) N Q1 (%) Q2 (%) Q3 (%) Q4 (%)
OR Q1 vs. Q4
(95% CI)
Soccer 1042 27.6 26.2 27.7 18.4 1.49 (1.23, 1.79) 121 20.7 29.8 36.4 13.2 1.55 (0.83, 2.90)
Volleyball 557 29.3 27.5 24.2 19.0 1.53 (1.19, 1.95) 45 33.3 35.6 15.6 15.6 2.13 (0.87, 5.21)
Alpine skiing 412 31.6 24.5 22.3 21.6 1.45 (1.10, 1.90) 61 37.7 23.0 19.7 19.7 1.90 (0.95, 3.82)
Swimming 375 25.3 25.6 26.1 22.9 1.10 (0.82, 1.47) 12 25.0 25.0 25.0 25.0 0.99 (0.20, 4.91)
Handball 277 25.6 33.9 20.6 19.9 1.28 (0.90, 1.82) 42 26.2 38.1 16.7 19.0 1.36 (0.55, 3.39)
Gymnastics artistic 240 28.3 23.3 27.5 20.8 1.35 (0.93, 1.95) 58 17.2 32.8 20.7 29.3 0.58 (0.27, 1.27)
Tennis 201 29.9 26.9 23.9 19.4 1.53 (1.02, 2.29) 34 29.4 17.6 26.5 26.5 1.10 (0.45, 2.71)
Athletics (track and field) 183 31.7 29.0 23.5 15.8 1.98 (1.27, 3.10) 29 27.6 37.9 17.2 17.2 1.59 (0.52, 4.85)
Basketball 174 25.3 26.4 26.4 21.8 1.15 (0.74, 1.78) 47 21.3 25.5 36.2 17.0 1.24 (0.49, 3.14)
Equestrian jumping 163 26.4 28.2 20.2 25.2 1.04 (0.68, 1.60) 38 28.9 21.1 21.1 28.9 0.99 (0.43, 2.29)
Figure skating 148 21.6 23.0 33.8 21.6 0.99 (0.61, 1.62) 21 33.3 14.3 38.1 14.3 2.31 (0.60, 8.95)
Cross-country skiing 116 26.7 31.0 19.0 23.3 1.14 (0.68, 1.91) 15 33.3 26.7 0.0 40.0 0.83 (0.25, 2.71)
Orienteering (running) 111 21.6 25.2 27.0 26.1 0.82 (0.48, 1.41) 11 18.2 27.3 36.4 18.2 0.99 (0.14, 7.04)
Synchronized schwimming 104 27.9 29.8 27.9 14.4 1.92 (1.03, 3.58) 42 26 28.6 33.3 11.9 2.18 (0.76, 6.28)
Judo 84 28.6 25.0 25.0 21.4 1.32 (0.72, 2.44) 13 23.1 46.2 30.8 0.0
Sports climbing 81 35.8 17.3 21.0 25.9 1.37 (0.78, 2.41) 13 23.1 15.4 30.8 30.8 0.74 (0.17, 3.32)
Rhythmic gymnastics 73 28.8 23.3 30.1 17.8 1.60 (0.80, 3.21) 33 33.3 30.3 21.2 15.2 2.18 (0.76, 6.28)
Shooting 71 36.6 23.9 18.3 21.1 1.72 (0.91, 3.25) 9 44.4 33.3 11.1 11.1 3.97 (0.44, 35.5)
Ice hockey 67 26.9 29.9 17.9 25.4 1.05 (0.54, 2.04) 26 26.9 23.1 23.1 26.9 0.99 (0.35, 2.83)
Badminton 65 24.6 40.0 21.5 13.8 1.76 (0.78, 3.99) 18 22.2 38.9 22.2 16.7 1.32 (0.30, 5.91)
Rowing 57 29.8 29.8 24.6 15.8 1.87 (0.83, 4.21) 13 7.7 30.8 38.5 23.1 0.33 (0.03, 3.18)
Karate 53 26.0 21.0 36.0 17.0 1.54 (0.67, 3.57) 17 29.4 0.0 58.8 11.8 2.48 (0.48, 12.8)
Synchronized skating 50 30.0 26.0 24.0 20.0 1.49 (0.67, 3.31)
Snowboarding 49 26.5 22.4 26.5 24.5 1.07 (0.49, 2.36) 17 29.4 11.8 17.6 41.2 0.71 (0.22, 2.23)
Cycling mountainbike 43 34.9 25.6 23.3 16.3 2.13 (0.87, 5.22) 11 36.4 9.1 27.3 27.3 1.32 (0.30, 5.91)
Total 4796 28.0 26.7 25.1 20.2 1.38 (1.26, 1.51) 746 26.5 27.3 26.4 19.7 1.34 (1.08, 1.65)
Notes: Q1 to Q4 = Quartile 1 to 4; OR = Odds ratio; 95% CI = 95% Confidence Interval.
4M. ROMANN ET AL.
In line with existing evidence (Albuquerque et al., 2012;
Delorme, 2014), no relevant RAE were found in sports with
weight categories for male athletes in the STDP. Similar evi-
dence has already been revealed in combat sports (Delorme,
2014). This phenomenon might be explained by a “strategic
adaptation”, which is a voluntary shift of children to another
sports where their physical capacities will be less determining
for performance. Albuquerque et al. (2012) also did not find
RAE within Olympic taekwondo athletes over 12 years of age.
He assumed that RAE are reduced or diminished in taekwondo
due to its competitive categories (weight categories and belt
level), which are determined by the level of technical skills.
Thus, the younger athletes are less disadvantaged because of
lower physical capacities. This system of categorization may
act as a key against drop out of younger athletes.
Methodological considerations
The obvious strength of the present study is that it ana-
lysed a complete countrywide data pool comprised of all
selected youth athletes in Switzerland in 2014. Therefore,
comparisons between sports are not affected by possible
differences between countries in, for instance, the talent
identification, selection and development systems or the
popularity of particular sports. However, Switzerland is a
small country with small samples available for analyses in
several sports. Thus, the results in small sports should be
interpreted with care and the generalizability to other
countries is limited. Swiss sports federations are aware of
the problems associated with RAE, and coaches are
educated in this regard. Additionally the current approach
to counter RAE on the national level, which was introduced
in 2008, did not reduce the effect. In this approach, the
RAE phenomenon was educated to coaches and the fed-
erations had to integrate bonus points for RAE disadvan-
taged athletes in the PISTE-selection process (Fuchslocher
et al., 2013). In international professional football, it has
been shown that awareness of the problem and 10 years
of research did not solve the problem as well, indicating
that RAE is a robust phenomenon and cannot easily be
changed (Helsen et al., 2012).
Withinthepresentinvestigationwewerenotableto
investigate possible causal factors of the RAE (e.g., biolo-
gical maturity status or physical performance). Additionally,
the evolution of RAE in different age categories have not
been analysed. Therefore, future research should focus on
the underlying mechanisms behind the RAE and investi-
gate differences of RAE between age categories and spe-
cific types of sports.
Conclusions and practical implications
RAE are present in several sports in the talent develop-
ment system of Switzerland, particularly in male youth
athletes in Olympic sports on the highest selection level.
This implies that talent and resources are being wasted,
which is especially problematic for a small country like
Switzerland. The present data support the detection of
high-risk sports or groups of sports and, thus, a tailored
implementation of preventive measures. Such measures,
Table 2. Nationwide data of relative age effects in male youth athletes in the Swiss talent development programme (STDP) and in the national level selection (NLS)
in the year 2014 for the 25 largest sports.
STDP NLS
Sport N Q1 (%) Q2 (%) Q3 (%) Q4 (%)
OR Q1 vs. Q4
(95% CI) N Q1 (%) Q2 (%) Q3 (%) Q4 (%)
OR Q1 vs. Q4
(95% CI)
Soccer 6010 35.9 27.8 20.9 15.3 2.33 (2.13, 2.54) 217 48.8 23.0 17.5 10.6 4.57 (2.91, 7.18)
Ice hockey 1501 33.3 26.9 23.3 16.5 2.01 (1.71, 2.35) 226 38.9 28.8 23.0 9.3 4.16 (2.58, 6.69)
Handball 601 29.8 25.6 24.5 20.1 1.47 (1.16, 1.85) 66 36.4 22.7 24.2 16.7 2.16 (1.06, 4.42)
Alpine skiing 580 30.7 27.4 24.3 17.6 1.73 (1.35, 2.22) 74 40.5 27.0 23.0 9.5 4.25 (1.87, 9.68)
Tennis 432 32.4 24.5 22.9 20.1 1.60 (1.22, 2.09) 39 53.8 23.1 17.9 5.1 10.41 (2.44, 44.4)
Gymnastics artistic 339 25.1 27.1 26.0 21.8 1.14 (0.83, 1.56) 67 23.9 35.8 19.4 20.9 1.13 (0.55, 2.32)
Swimming 335 30.1 25.7 23.9 20.3 1.47 (1.08, 2.01) 21 28.6 14.3 14.3 42.9 0.66 (0.24, 1.86)
Floorball 299 28.4 25.4 26.4 19.7 1.43 (1.02, 2.00) 24 29.2 33.3 25.0 12.5 2.31 (0.60, 8.95)
Basketball 273 31.9 28.9 23.8 15.4 2.05 (1.42, 2.98) 60 38.3 33.3 13.3 15.0 2.53 (1.17, 5.48)
Judo 250 20.4 24.4 29.6 25.6 0.79 (0.55, 1.14) 23 30.4 30.4 26.1 13.0 2.31 (0.60, 8.95)
Volleyball 223 29.6 30.0 24.7 15.7 1.87 (1.24, 2.82) 34 32.4 32.4 17.6 17.6 1.82 (0.67, 4.92)
Athletics (track and field) 179 35.8 25.7 24.0 14.5 2.44 (1.54, 3.86) 34 26.5 35.3 23.5 14.7 1.79 (0.60, 5.33)
Water polo 174 31.0 23.0 20.7 25.3 1.22 (0.82, 1.82) 72 33.3 26.4 18.1 22.2 1.49 (0.79, 2.80)
Orienteering (running) 164 22.0 25.0 25.6 27.4 0.79 (0.51, 1.23) 12 0.0 16.7 41.7 41.7
Snowboarding 163 27.0 20.2 25.8 27.0 0.99 (0.65, 1.51) 20 35.0 15.0 40.0 10.0 3.47 (0.72, 16.71)
Badminton 162 25.9 25.9 30.9 17.3 1.49 (0.92, 2.40) 26 23.1 23.1 46.2 7.7 2.98 (0.60, 14.74)
Cross-country skiing 160 28.8 31.9 23.1 16.3 1.75 (1.08, 2.84) 22 36.4 27.3 22.7 13.6 2.64 (0.70, 9.97)
Cycling mountainbike 109 27.5 29.4 25.7 17.4 1.57 (0.88, 2.79) 29 37.9 20.7 20.7 20.7 1.82 (0.67, 4.92)
Sports climbing 106 26.4 23.6 34.0 16.0 1.63 (0.89, 2.99) 11 27.3 27.3 45.5 0.0
Shooting 97 30.9 27.8 27.8 13.4 2.29 (1.19, 4.39) 8 12.5 37.5 50.0 0.0
Field hockey 95 28.4 26.3 30.5 14.7 1.91 (1.00, 3.65) 49 28.6 24.5 22.4 24.5 1.16 (0.54, 2.50)
Karate 87 21.8 26.4 16.1 35.6 0.61 (0.34, 1.08) 22 31.8 27.3 13.6 27.3 1.16 (0.39, 3.44)
Table tennis 86 22.1 36.0 20.9 20.9 1.05 (0.55, 2.00) 22 22.7 36.4 13.6 27.3 0.83 (0.25, 2.71)
Rowing 85 27.1 27.1 28.2 17.6 1.52 (0.79, 2.92) 24 45.8 25.0 20.8 8.3 5.46 (1.21, 24.6)
Ski jumping 74 33.8 13.5 24.3 28.4 1.18 (0.66, 2.11) 10 20.0 10.0 30.0 40.0 0.50 (0.09, 2.71)
Total 12,584 32.7 27.1 22.9 17.3 1.87 (1.76, 2.00) 1212 36.9 26.8 21.7 14.6 2.50 (2.10, 2.98)
Notes: Q1 to Q4 = Quartile 1 to 4; OR = Odds ratio; 95% CI = 95% Confidence Interval.
JOURNAL OF SPORTS SCIENCES 5
which have been already described and discussed in lit-
erature, might include (a) improving coach education in
sports at high risk of RAE, (b) building categories by bio-
logical age, weight or height (Musch & Grondin), (c) quotas
of same RAE quarters, (d) rotating cut-off dates (Cobley
et al., 2009) and (e) RAE correction factors in centimetre-
gram-second sports (Romann & Cobley, 2015). As every
selection increases RAE, selection as late as possible, at
best after age at peak height velocity, may also contribute
to reducing RAE. In Switzerland, talent identification, selec-
tion and development should be considered as a long-
term process. Moreover, reducing RAE in the Swiss sport
system would make long-term athlete development more
legitimate and effective.
Disclosure statement
No potential conflict of interest was reported by the authors.
ORCID
Michael Romann http://orcid.org/0000-0003-4139-2955
Roland Rössler http://orcid.org/0000-0002-6763-0694
Marie Javet http://orcid.org/0000-0002-6493-4764
References
Albuquerque, M. R., Lage, G. M., Da Costa, V. T., Ferreira, R. M., Penna, E. M.,
Moraes, L. C., & Malloy-Diniz, L. F. (2012). Relative age effect in Olympic
Figure 2. Relative age effects (RAE) in the Swiss talent development programme (STDP; regional to national level; open squares and circles) and the national level
selection (NLS; national to international level; filled squares and circles; squares indicate Olympic sports, circles non-Olympic sports). Odds ratios (OR) for the
frequency of athletes born in the first quarter of the year relative to athletes born in the last quarter of the year (females, upper panel; males, lower panel; the
coloured bars indicate a large (red), medium (orange) and small (green) RAE).
6M. ROMANN ET AL.
Taekwondo athletes. Perceptual and Motor Skills,114(2), 461–468.
doi:10.2466/05.25.PMS.114.2.461-468
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. doi:10.2165/00007256-200939030-
00005
Delorme, N. (2014). Do weight categories prevent athletes from relative age
effect? Journal of Sports Sciences,32(1), 16–21. doi:10.1080/
02640414.2013.809470
Delorme, N., Boiché, J., & Raspaud, M. (2010). Relative age and dropout in
French male soccer. Journal of Sports Sciences,28(7), 717–722.
doi:10.1080/02640411003663276
Delorme,N.,Chalabaev,A.,&Raspaud,M.(2011). Relative age is associated with
sport dropout: Evidence from youth categories of French basketball.
Scandinavian Journal of Medicine and Science in Sports,21(1), 120–128.
doi:10.1111/j.1600-0838.2009.01060.x
Delorme, N., & Raspaud, M. (2009). Is there an influence of relative age on
participation in non-physical sports activities? The example of shooting
sports. Journal of Sports Sciences,27(10), 1035-1042. doi:10.1080/
02640410902926438
Fuchslocher,J.,Romann,M.,&Gulbin,J.(2013). Strategies to Support
Developing Talent. Schweizerische Zeitschrift Für Sportmedizin Und
Sporttraumatologie,61,4.
Gibbs, B. G., Shafer, K., & Dufur, M. J. (2012). Why infer? The use and misuse
of population data in sport research. International Review for the
Sociology of Sport. doi:10.1177/1012690212469019
Gil, S. M., Badiola, A., Bidaurrazaga-Letona, I., Zabala-Lili, J., Gravina, L., Santos-
Concejero, J., . .. Granados, C. (2014). Relationship between the relative age
effect and anthropometry, maturity and performance in young soccer
players. Journal of Sports Sciences,32(5), 479–486. doi:10.1080/
02640414.2013.832355
Goldschmied, N. (2011). No evidence for the relative age effect in professional
women’ssports.Sports Medicine,41(1), 87–91. doi:10.2165/11586780
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. doi:10.1080/
17461391.2013.775352
Helsen, W., Baker, J., Michiels, S., Schorer, J., Van Winckel, J., &
Williams, A. M. (2012).TherelativeageeffectinEuropeanprofes-
sional soccer: Did ten years of research make any difference?
Journal of Sports Sciences,30(15), 1665–1671. doi:10.1080/
02640414.2012.721929
Müller, L., Hildebrandt, C., & Raschner, C. (2015). The relative age effect
and the influence on performance in youth alpine ski racing. Journal
of Sports Science and Medicine,14,16–22.
Musch,J.,&Grondin,S.(2001). Unequal competition as an impedi-
ment to personal development: A review of the relative age effect
in sport. Developmental Review,21(2), 147–167. doi:10.1006/
drev.2000.0516
Olivier, J., & Bell, M. L. (2013).Effectsizesfor2×2contingencytables.
PLoS One,8(3), e58777. doi:10.1371/journal.pone.0058777
Raschner, C., Müller, L., & Hildebrandt, C. (2012).Theroleofarelative
age effect in the first winter Youth Olympic Games in 2012. British
Journal of Sports Medicine,46(15), 1038–1043. doi:10.1136/bjsports-
2012-091535
Reed, K. E., Parry, D., & Sandercock, G. (2016). Maturational and social
factors contributing to relative age effects in school sports: Data from
the London Youth Games. Scandinavian Journal of Medicine and Science
in Sports. doi:10.1111/sms.12815
Romann,M.,&Cobley,S.(2015). Relative age effects in athletic sprint-
ing and corrective adjustments as a solution for their removal. PLoS
One,20(3). doi:10.1371/journal.pone.0122988
Romann, M., & Fuchslocher, J. (2013). Influences of player nationality,
playing position, and height on relative age effects at women’s
under-17 FIFA world cup. Journal of Sports Sciences,31(1), 32–40.
doi:10.1080/02640414.2012.718442
Schorer, J., Cobley, S., Busch, D., Brautigam, H., & Baker, J. (2009).
Influences of competition level, gender, player nationality, career
stage and playing position on relative age effects. Scandinavian
Journal of Medicine and Science in Sports,19(5), 720–730.
doi:10.1111/j.1600-0838.2008.00838.x
Skorski, S., Skorski, S., Faude, O., Hammes, D., & Meyer, T. (2016). The
relative age effect in elite German youth soccer: Implications for a
successful career. International Journal of Sports Physiology &
Performance,11(3). doi:10.1123/ijspp.2015-0071
Swiss Federal Office of Sport (2015). Jugend + Sport. Retrieved February 3,
2015, from http://www.jugendundsport.ch/de/ueber-j-s/statistik/2015.
html
Swiss Olympic Association (2017). Richtlinien Beiträge an die
Mitglieder von Swiss Olympic. Retrieved Dezember 4, 2017, from
https://www.swissolympic.ch/dam/jcr:3b1e35c8-21a6-4983-96ec-
dfddf721e916/Richtlinien_Verbandsbeitraege_170120_DE.PDF
Vincent, J., & Glamser, F. D. (2006). Gender differences in the relative age
effect among US olympic development program youth soccer players.
Journal of Sports Sciences,24(4), 405–413. doi:10.1080/
02640410500244655
Wattie,N.,Schorer,J.,&Baker,J.(2015). The relative age effect in
sport: A developmental systems model. Sports Medicine,45(1), 83–
94. doi:10.1007/s40279-014-0248-9
JOURNAL OF SPORTS SCIENCES 7
