See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/267839188

8 — Controlling for Maturation in Pediatric Exercise

Science

Article  in  Pediatric exercise science · November 2004

DOI: 10.1123/pes.17.1.18

CITATIONS READS

137 3,651

3 authors:

Adam Dominic George Baxter-Jones Joey C Eisenmann

University of Saskatchewan 290 PUBLICATIONS   11,163 CITATIONS   
237 PUBLICATIONS   7,864 CITATIONS   
SEE PROFILE
SEE PROFILE

Lauren Sherar
Loughborough University
116 PUBLICATIONS   3,365 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Volt Sport Science View project

Saskatchewan Growth and Development Studyn View project

All content following this page was uploaded by Adam Dominic George Baxter-Jones on 26 November 2014.

The user has requested enhancement of the downloaded file.

18 — Baxter-Jones, Eisenmann, and Sherar

Pediatric Exercise Science, 2005, 17, 18-30

© 2005 Human Kinetics Publishers, Inc.

Controlling for Maturation

in Pediatric Exercise Science

Adam D. G. Baxter-Jones, Joey C. Eisenmann,

and Lauren B. Sherar

The process of maturation is continuous throughout childhood and adoles-

cence. In a biological context, the effects of a child’s maturation might mask
or be greater than the effects associated with exposure to exercise. Pediatric
exercise scientists must therefore include an assessment of biological age in
study designs so that the confounding effects of maturation can be controlled
for. In order to understand how maturation can be assessed, it is important to
appreciate that 1 year of chronological time is not equivalent to 1 year of
biological time. Sex- and age-associated variations in the timing and tempo of
biological maturation have long been recognized. This paper reviews some of
the possible biological maturity indicators that the pediatric exercise scientist
can use. As a result, we recommend that any of the methods discussed could
be used for gender-specific comparisons. Gender-comparison studies should
either use skeletal age or some form of somatic index.

Introduction
Pediatric exercise science examines the acute and chronic responses of chil-
dren and adolescents to exercise and/or physical activity. Of primary interest are
physiological changes, physical activity, and health-related outcomes, and the dif-
ferences in these changes and outcomes between genders and between children
and adults. Morphological parameters and physiological functions such as heart
volume, lung function, aerobic power, and muscular strength develop with increas-
ing age and body size. Furthermore, physical fitness (e.g., muscular, motor, and
cardiorespiratory fitness) also changes with growth and maturation. Given the in-
creasing interest in pediatric exercise science, there is a need for greater under-
standing of the general principals of human growth, maturation, and development,
generally referred to as auxology.
The question facing pediatric exercise scientists is: How do you separate the
independent effects of growth and maturation from those of other environmental

Baxter-Jones and Sherar are with the University of Saskatchewan, College of Kine-
siology, 87 Campus Drive, Saskatoon SK, S7N 5B2, Canada. Eisenmann is with Iowa State
University, Department of Health and Human Performance, 255 Forker Building, Ames, IA
50011.

18
 Controlling for Maturation — 19

factors, such as physical activity or systematic training? This is important because

often the effects of growth and maturation mask or are greater than the effects
associated with such environmental exposures. In part, the confounding effects of
physical growth have been controlled by creating size-free variables that allow for
within- and between-gender comparisons. Once the confounding effects of size
have been controlled for, however, it has been shown that an individual’s maturity
status (biological age) can also have an additional and significant independent
effect on many physiological outcomes (2,5,18).
Sex- and age-associated variations in the timing and tempo of biological
maturation have long been recognized (8,28,29,30). During adolescence, children
of the same chronological age differ considerably in biological maturity. The im-
plication for pediatric exercise science is that, in addition to controlling for body
size, researchers also need to control for maturational differences. This article re-
views some of the possible biological maturity indicators that pediatric exercise
scientists can use in research and practice.

General Principles of Human Growth

and Maturation
Before discussing the assessment of maturity, we must first have an under-
standing of the processes of normal growth and maturation. Today’s concepts of
growth and maturation rely on the early work of many eminent auxologists (29).
The terms growth, biological maturation, and development are often used synony-
mously. Although interrelated, the concepts have fundamental and semantic dif-
ferences. Growth refers to a change in size, either of the entire individual (e.g.,
stature) or of various parts (e.g., heart, lung, etc). Maturation implies progress
towards the adult state, which varies in timing and tempo. Development can be
defined as a psychosocial or behavioral construct. It is important to recognize that
the three processes both occur simultaneously and interact.

Growth
As children grow, they become taller and heavier, the amount of lean and fat
tissue increases, their organs increase in size, and so on. Different parts of the body
grow at different rates and at different times. The seminal work by Scammon (24),
produced in the 1920s and 30s, used an incremental measuring system and deter-
mined the relative size attained for general (e.g., stature, heart size), neurological
(e.g., brain and head), and genital (reproductive organs) growth at various growth
stages as a percentage of the size attained at full maturity (see Figure 1). From
birth, brain and head growth are the most rapid. By 2 years of age, the brain and
head have reached nearly 50% of their adult size, and full adult size is reached by
8 to 10 years of age. In contrast, the growth of reproductive organs is much slower;
by 10 to 12 years of age, reproductive organs are only 10% of their adult size. A
very rapid growth spurt then occurs as children enter adolescence (puberty). At the
same time, there is also an increase in the rate of growth observed in the general
growth curve, which represents most parameters associated with general body size
(height, weight, heart size, etc.). For example, at 10–12 years of age children are
roughly 84% of their adult height.
20 — Baxter-Jones, Eisenmann, and Sherar

Figure 1 — Scammon growth curves showing different parts and tissues of the body.
All curves are of size attained plotted as a percentage of total gain from birth to 20
years. Size at 20 years is 100% on the vertical scale. (Reprinted, by permission, from
R.E. Scammon, 1930, The Measurement of Man, edited by J.A. Harris et al., Minne-
apolis, MN: University of Minnesota Press).

Adolescence is a period of dynamic physical and biological change. During

the pubertal period, a sudden increase in the velocity of growth occurs (e.g., ado-
lescent growth spurt), second only to the growth rate observed in the first year of
life. This change in velocity affects all parts of the body, including long bones,
vertebrae, heart, lungs, viscera, and muscle mass. During puberty there are notice-
able changes in body size and shape, in the relative proportions of muscle, fat, and
bone, and in a variety of physiological functions (30).

Biological Maturation
Biological maturation refers to progress towards the biologically mature state
and is not absolutely linked to time in a chronological sense. During childhood
and adolescence, individuals of the same chronological age can differ dramati-
cally in the degree of biological maturity. The process of maturing has two compo-
nents: timing and tempo. The former refers to the age at which specific matura-
tional events occur (e.g., age when menarche is attained, age at the beginning of
breast development, age at the appearance of pubic hair, or age at maximum growth
 Controlling for Maturation — 21

during the adolescent growth spurt.) Tempo refers to the rate at which maturation
progresses (i.e., how quickly or slowly an individual passes from the initial stages
of sexual maturation to the mature state).

Development
Development refers to the acquisition of behavioral competence, (i.e., learn-
ing the appropriate behaviors expected by society). Development is culturally spe-
cific. As children experience home, school, church, sports, recreation, and other
community activities, they develop cognitively, socially, emotionally, morally, and
so on. Development refers to the process by which children and adolescents learn
to behave in culturally appropriate ways.

Controlling for the Confounding Effects of Maturity

More often than not, pediatric exercise scientists are interested in examining
the trainability of children or, in other words, the association between physical
activity and health outcomes. Interpretation of outcomes, however, must consider
the process of normal growth and maturation before any definitive conclusion can
be reached. Unless body size and biological maturity are considered, one cannot
definitively identify the independent effects of physical activity or training on the
outcome.

Accounting for Body Size

The problem of creating a “size-free” variable has been given considerable
attention in the literature (1,10,23,26,36) and will not be discussed in detail in the
present review. In brief, the traditional (and current) convention is to express physi-
ological measurements as a ratio standard, adjusting the function for body size
(e.g., VO2max expressed relative to body mass as ml·kg-1·min-1). The use of ratio
standards was first criticized by Tanner (27) in the late 40’s and more recently by
Winter et al. (36) and others (1,10). The criticisms result from the fact that spuri-
ous correlations, misinterpretation of the data, and indirect conclusions can follow
from use of the method. Allometry and regression analysis are suggested alterna-
tive techniques that can be used for body size adjustments (1,23).

Research Design
Biological maturity can be controlled for by aligning individuals of similar
maturity status (or biological age), which requires an assessment of maturity. Such
measures of maturity can either be incorporated into the research design and/or the
statistical analysis. Longitudinal research designs with sophisticated statistical
analyses, such as repeated measures within individuals over time and multilevel
modeling (6), have been employed to distinguish the independent effects of envi-
ronmental exposure (e.g., exercise) from those of normal growth and biological
maturation (2,5,6,18). The advantage of longitudinal over cross-sectional designs
is that within-individual variance can be obtained and thus the timing and tempo
of an individual’s pattern of growth identified.
When conducting longitudinal research, it is important to remember that
two measures separated by a time period do not constitute longitudinal data; true
22 — Baxter-Jones, Eisenmann, and Sherar

longitudinal data has at least three measures and, thus, two velocities. A pure lon-
gitudinal design is where a cohort of children born in the same year is followed
continuously for 3 or more years. A mixed-longitudinal design is where a number
of relatively short longitudinal studies are interlocked covering a whole age range
(e.g., 8–10 years, 9–11 years, 10–13 years, etc). Unfortunately, longitudinal re-
search is often impractical for pediatric exercise research because the process is
laborious, expensive, and time consuming for both the participants and investiga-
tors. The reality is that the primary objective of the research is not usually the
documentation of growth and biological maturation. Thus, spending the money,
time, and effort on longitudinal research in order to account for maturity is not
feasible. This means that most knowledge in pediatric exercise science is based on
cross-sectional studies because they are cheaper, can be carried out quickly, and
can include larger numbers of children.

Indicators of Biological Maturity

Maturity can only be adequately controlled for by incorporating a maturity
indicator in the research methodology. The maturity indicator chosen can be any
definable and sequential change in any part of the body that is characteristic of the
progression of the body from immaturity to maturity (8). The most commonly
used methods involve an assessment of skeletal age, secondary sex characteristics,
menarcheal status, and/or somatic characteristics. Each method will be briefly re-
viewed.

Skeletal Age
A skeletal age assessment requires an X-ray, usually of the hand and wrist or
knee, and is the only method that spans the entire growth period. The assessment
of skeletal maturity is based on the observation that a person more advanced in
maturity will have greater bone development and a smaller amount of cartilage
than a less mature person. Skeletal ages ranging from 9 to 16 years have been
demonstrated in a group of 13- and 14-year olds, thus illustrating the wide varia-
tion in skeletal age evident in children of a similar chronological age (11). This
variation emphasizes why using a common chronological age as a pubertal cut-off
point, for example all children less than 12 years of age classified as prepubertal, is
not tenable. Although the assessment of skeletal age is considered the best matura-
tional index, it is costly, requires specialized equipment and interpretation, and
incurs radiation safety issues. Furthermore, discrepancies of 1 or more years in
skeletal age between X-rays of the knee and hand/wrist have been documented in
individual youths (22). This finding brings into question whether the skeletal ma-
turity of the hand and wrist represents the maturity of the whole skeleton.

Age at Peak Height Velocity

The identification of landmarks on the human growth curve can be used for
comparisons among individuals or groups. The most commonly used somatic mile-
stone in longitudinal studies of childhood growth is the age at peak height velocity
(PHV). In order to obtain age at PHV, whole year height-velocity (cm/year) incre-
ments are plotted and mathematical curve-fitting procedures are used to identify
the age when the maximum velocity in statural growth occurs (Figure 2). The
 Controlling for Maturation — 23

Figure 2 — Whole year height velocities are plotted against chronological age. Peak
height velocity was determined with a cubic spline fitting procedure. Data was taken
from the Saskatchewan Pediatric Bone Mineral Accrual Study (3).

timing of this event in relation to chronological age shows great variance. The
average age for girls is 12 years (range 9.5 to 14.5) and for boys, 14 years (range
10.5 to 17.5) (29). Once age at PHV has been determined, individuals can be aligned
by biological age (years from age at PHV) rather than by chronological age. At age
of PHV, an individual has a biological age equal to 0.0 years from PHV. At 11.8
years, an individual who reaches PHV at 13.8 years will have a biological age of -
2.0 years from PHV. Alternatively, individuals can be characterized as early, aver-
age, or late maturing depending on the age at which PHV is attained. Early maturers
are those individuals whose age at PHV is 1 year (or more) less than the mean age
of PHV, whereas late maturers have an age at PHV of 1 year or more than the mean
age of PHV; the remainder are is classified as average matures.
24 — Baxter-Jones, Eisenmann, and Sherar

Predicting Age at PHV

In order to obtain the maturity offset age, that is, the years from age at PHV,
serial data is required, and therefore, this indicator of maturity has previously been
limited to longitudinal studies. Recently, Mirwald et al. (17) have developed gen-
der-specific multiple-regression equations based on segmental growth patterns that
predict the maturity offset age parameter. The prediction equations require mea-
sures of stature, trunk length, and leg length, as well as body mass and chronologi-
cal age. Using these growth indicators, age from PHV was predicted within ± 1
year in 95% of cases. The authors suggest that maturity offset could be used as a
categorical (pre- or post-PHV), as well as a continuous measure of maturity. This
predicted maturity offset age is quick, cheap, noninvasive to administer, and can
be used in cross-sectional studies. The added advantage for this technique is that it
predicts a maturity benchmark that exists in both boys and girls. Therefore, it al-
lows for between-gender comparisons.

Predictions Based on Morphological Age

The height attained at any given chronological age can be compared with
reference norms in order to assess maturity. An individual is assigned a morpho-
logical age based on height and age classifications. The major disadvantage of this
method is that it does not take into account the variability of height. Statural growth
of three males measured at 7, 14, and 40 years of age are presented in Figure 3;
these are males who participated in the Saskatchewan Growth and Development
Study (SGDS), which ran from 1964 through 1973 (16), and who were then

Figure 3 — Stature and percentage of adult stature for three males at 7, 14 and 40
years of age. Data was taken from three individuals who participated in the
Saskatchewan Growth and Development Study (16).
 Controlling for Maturation — 25

reassessed in 1999. At 7 years of age, Boys (a) and (c) are about the same height,
whereas Boy (b) is 10 cm shorter. By 14 years of age, Boys (a) and (c) are still
about the same height, but Boy (c) is nearly 18 cm shorter. Using a morphological
age scale, Boy (b) at 14 years of age would be identified as a late maturer.
Another method of using somatic growth is to express measured height in
terms of the percentage of final adult height (4,7,22,28,30,31,32,34). As shown in
Figure 3, although Boy (b) appears to be small for his age in absolute terms, when
presented as a percentage of final adult height there is no difference between Boys
(b) and (c) at 7 and 14 years of age. This is because at 40 years of age Boy (a) and
(b) are the same height and Boy (c) is 15 cm taller. Because roughly 92% of adult
stature is reached at PHV (28), individuals could be classified into two maturity
groups, pre- or post-PHV. Because the average age of PHV in boys is 14 years,
Boy (a) at 14 years of age would be classified as early maturing (percentage adult
stature > 92%) and Boys (b) and (c) as average maturers (percentage adult stature
< 92%). The disadvantage of this technique is that an adult value is required and,
consequently, maturity status can only be applied retrospectively.
Expressing current height as a percentage of adult height can, however, be
used in cross-sectional studies if adult height is predicted. Many equations have
been developed to predict adult height. The most commonly used methods are
those of Bayley and Pinneau (4), Roche et al. (22), and Tanner et al. (31,32). These
methods all require an assessment of skeletal age and are, therefore, not practical.
Recently, however, predictive equations have been developed that do not require a
measure of skeletal age (7,34) and, thus, have potential for use in pediatric studies.

Secondary Sex Characteristics

One of the most recognizable names in auxology is James M. Tanner. Dr
Tanner was a UK pediatrician who conducted the Harpenden Growth Study be-
tween 1948 and 1971 (29). He is best known for the classic texts, Growth at Ado-
lescence (28) and Foetus into Man (30). His most influential work related to the
staging of secondary sex characteristics and this has played a prominent part in
controlling for maturation in the pediatric exercise literature.
Sexual maturation can be assessed using indices of secondary sexual char-
acteristics (8,12). Secondary sex staging divides the process of breast develop-
ment in girls, genitalia development in boys, and pubic hair development in both
sexes, into 5 stages. These secondary sex stages are commonly referred to as Tan-
ner Stages; the technique was first documented by Reynolds and Wines in the
1950’s (20,21) and later described in more detail by Tanner (28). Facial hair, axil-
lary hair, voice change, body odor, body shape, and menarche are other aspects of
pubertal change that have been indexed (8). In addition to staging techniques, genital
maturity in males can be determined through the assessment of testicular volume.
The first overt signs of pubertal development are testicular enlargement in boys
and breast development in girls.
Traditionally, determination of sexual maturity has been obtained through
direct visual observation. This approach is appropriate for clinical settings, but it
poses problems for the assessment of children in a nonclinical setting because
many adolescents (and their parents) feel uncomfortable with this method of as-
sessment. In order to address these concerns, techniques have been developed based
on self-assessment (from photographs and drawings), and it has been demonstrated
that children and adolescents can rate their own sexual development accurately
26 — Baxter-Jones, Eisenmann, and Sherar

and reliably (9,15,19,25,33,35). Nonetheless, there are still concerns that youths
overestimate early stages and underestimate later stages of sexual development (8).
Individual variation in the timing and tempo of somatic and sexual matura-
tional indices has long been recognized (13,14). Figure 4 illustrates various tim-
ings of these pubertal events using data from the Saskatchewan Pediatric Bone
Mineral Accrual Study ( 3). Figure 4 illustrates that several pubertal events occur
at the same time under the control of various (not necessarily the same) endocrine
systems and are ultimately controlled by genetic expression. The timing of puber-
tal events, however, varies between individuals of the same sex. For example, the
average age of breast development in girls is 10.5 years, but it is not unusual for
this event to start as early as 8 years or as late as 13 years. On average, breast
development reaches the mature state by 14.5 years, although early maturing girls

Figure 4 — Average age of attainment of pubic hair (PH) Stages 3-5 and peak height
velocity (PHV). Axillary hair and facial hair growth are shown in boys only; menarche
is shown in girls only. Values are means (circles) and two standard deviations (bars).
Data was taken from the Saskatchewan Pediatric Bone Mineral Accrual Study (3).
 Controlling for Maturation — 27

can reach this stage of development by 12 years of age and the late maturing girls
by 18 years of age.
As well as individual variation, there is also a marked sex difference in the
timing of somatic and sexual maturation. Girls enter and end puberty approxi-
mately 2 years before boys, and pubertal events do not occur in the same sequence
between the sexes. For example, when comparing pubic hair growth to statural
growth, PHV is a relatively early event in girls and a relatively late event in boys
(Figure 4). Boy’s PHV occurs, on average, during pubic hair Stage 4 and 5, whereas
girls’ PHV usually occurs during pubic hair Stage 3 and 4. This suggests that the
timing of sexual and somatic maturation is not the same in girls and boys. It should
also be noted that secondary sex stages are superimposed on one another and are a
continuous process.
A common misuse of secondary sex characteristics in controlling for matu-
rity is analyzing pubic hair categorizes as if they were continuous variables. An
individual in the early phase of the third stage of pubic hair development is rated
the same as an individual in the late phase of this stage. It is rare that the point in
time at which the change from one stage to another is ever measured; what is
actually being reported is the interval between two stages. This provides even less
information when you consider that one individual might pass through a stage in 2
years when another might take 5 years to pass through the same stage. In addition,
there is no relationship between the age at which a secondary sex characteristic
begins and the length of time that required to pass through the stage (13,14).
Aligning individuals by secondary sex characteristics is used frequently in
pediatric exercise science literature because it does not require longitudinal obser-
vations, and it is easy to administer, cost effective, and noninvasive (when physi-
cian assessment is replaced with self-assessment). The concern with the use of
secondary sex staging, however, relates to possible mistakes in the alignment of
individuals. Many pediatric studies align boys and girls on the same secondary sex
characteristics, on different secondary sex characteristic, or on more than one sec-
ondary sex characteristic in order to develop a composite score of sexual develop-
ment. The assumption behind these strategies is that the order and timing of the
appearance of the same secondary sex characteristic and/or different sex charac-
teristics are identical in both sexes. It further presumes that the sequence of the
appearance of secondary sex characteristics between genders with other maturity
indicators is also identical. As previously described, however, there is consider-
able difference in timing and tempo of somatic and sexual development between
sexes during adolescence. This means that all three of these alignment strategies
are inappropriate when making comparisons between boys and girls.
Furthermore, as Figure 4 demonstrates, different maturity events occur at
different times during adolescence. For example, genitalia development and breast
development occur early in adolescence, whereas menarche in girls, and axillary
and facial hair in boys, occurs late in adolescence. The current standards for sec-
ondary sex staging ignore this difference in timing of secondary sex characteris-
tics. An individual who is at Stage 3 for breast development will not necessarily be
at Stage 3 for pubic hair development. Likewise, a boy at Stage 3 for genital devel-
opment is not necessarily of the same biological age as a girl that is at Stage 3 for
breast development. Hence, it is inaccurate to make comparisons between indi-
viduals using different secondary sex characteristics. It is also important for re-
searchers to detail which secondary sex characteristic is being used as the maturity
indicator.
28 — Baxter-Jones, Eisenmann, and Sherar

When controlling for the confounding effects of maturity between genders,

boys and girls are most often aligned using pubic hair stages because this is the
only sex characteristic that is present in both boys and girls (apart from axillary
hair growth, which proves to be problematic if girls remove underarm hair). In
addition to the other cautions previously outlined, one should be aware that pubic
hair development represents the onset of adrenarche (an increased secretion of sex
hormones by the adrenal cortex) and not necessarily the onset of true pubertal
development (28,30). As stated previously, the onset of breast development in girls
and testicular volume in boys is the first true sign of centrally mediated puberty.
Thus, if individuals are aligned on pubic hair development only, caution should be
taken when classifying individuals in the early stages.

Menarcheal Status
Age at menarche (the first menstrual period) is the most commonly reported
developmental milestone of female adolescence in both cross-sectional and longi-
tudinal studies. Three methods (prospective, status quo, and recall) are commonly
used to establish age at menarche. The best and most reliable is to follow individu-
als and note the date at which menarche occurs. This method is limited, however,
because longitudinal data is required. Alternatively, normative values can be es-
tablished by the status-quo method. This involves asking a large number of girls
(usually aged between 8 and 18) when they were born and whether they have
started their menstrual flow. From their ages and their answers (yes or no), it is
possible to calculate mean and standard deviation values for age of menarche. The
third method is the recall method. A simple questionnaire is used to establish if an
individual has experienced menarche; if the answer is yes, they are asked to indi-
cate the date or month and year. The retrospective method is useful for individuals
after 17 years of age, when almost all girls have attained menarche.
Although age at menarche is a widely used maturity indicator in studies of
females, its use is limited to later adolescence because menarche usually occurs
after PHV (10). Most studies, especially in athletes, use the recall method, which
has the limitation of recall error. Estimated mean ages are biased because all sub-
jects have not yet reached menarche. Furthermore, age of menarche has little use
in gender comparison studies, as no corresponding maturity indicator exists in
males.

Relationship Between Indicators

As discussed earlier, the biological age—maturity—of an adolescent can be
assessed through a variety of techniques. Although the methods vary considerably,
they all express the estimated biological age relative to chronological age. For
example, youths are classified as early maturers if their biological age is greater
than their chronological age by one year or more. If an adolescent’s biological age
is within a year of their chronological age, they are classified as an average ma-
turer. Finally, if a youth’s estimated biological age is less than their chronological
age by 1 year or more, they are classified as a late maturer.
The correlations among the indicators are generally moderate to high but
tend to be lower for boys than for girls. One study found the correlation coefficient
between skeletal age and age at PHV to be 0.83 (12). The direction and strength of
 Controlling for Maturation — 29

this relationship indicates a maturity commonality between these two methods.

Although a general maturity factor underlies the tempo of growth and maturation
during adolescence in both genders, there is sufficient variation to suggest that no
single system (i.e., sexual, skeletal, or somatic) provides a complete description of
the tempo of maturation during adolescence. The apparent discord among the afore-
mentioned indicators reflects individual variation in the timing and tempo of sexual
and somatic maturity, and the methodological concerns that have been previously
outlined.

Conclusions
In this communication, we have reviewed important auxological principles
that should be applied to the study of pediatric exercise science. When researching
children, the investigator must always give consideration to the underlying effect
of growth and maturation, which could mask or be greater than the effects of train-
ing or physical activity. We recommend that, for gender specific comparisons, any
of the discussed methods could be used. For gender comparisons, however, stud-
ies should either use skeletal age or some form of somatic indices.

Reference List
1. Armstrong, N., and J. Welsman. Young people and physical activity. Oxford: Oxford
University Press, 1997.
2. Armstrong, N., J. Welsman, A.M. Nevill, and B. Kirby. Modeling growth and matura-
tion changes in peak oxygen uptake in 11–13 yr olds. J. Appl. Physiol. 87:2230-2236,
1999.
3. Bailey, D.A. The Saskatchewan pediatric bone mineral accrual study: Bone mineral
acquisition in the growing years. Int. J. Sports Med. 18:191-194, 1997.
4. Bayley, N., and S.R. Pinneau. Tables for predicting adult height from skeletal age:
Revised from and for use with the Greulich-Pyle hand standards. J. Pediatr. 40:423-
441, 1952.
5. Baxter-Jones, A., H. Goldstein, and P. Helms. The development of aerobic power in
young athletes. J. Appl. Physiol. 75:1160-1167, 1993.
6. Baxter-Jones, A., and R. Mirwald, Multilevel modelling. In: Methods in Human Growth
Research. R.C. Hauspie, N. Cameron, and L. Molinari (Eds.). Cambridge: Cambridge
University Press, 2004, pp. 306-330.
7. Beunen, G.P., R.M. Malina, J. Lefevre, A.L. Claessens, R. Renson, and J. Simons.
Prediction of adult stature and noninvasive assessment of biological maturation. Med.
Sci. Sports Exerc. 29:225-230, 1997.
8. Cameron, N. Assessment of maturation. In: Human Growth and Development. N.
Cameron (Ed.). San Diego: Academic Press, 2002. pp. 363-382.
9. Duke, P.M., I.F. Litt, and R.T. Gross. Adolescents’ self-assessment of sexual matura-
tion. Pediatrics. 66:918-920, 1980.
10. Eisenmann, J.C., J.M. Pivarnik, and R.M. Malina. Scaling peak VO2 to body mass in
young male and female distance runners. J. Appl. Physiol. 90:2172-2180, 2001.
11. Kemper, H.C.G., and R. Verschuur. Maximal aerobic power in 13- and 14-year-old
teenagers in relation to biological age. Int. J. Sports Med. 2:97-100, 1981.
12. Malina, R.M., C. Bouchard, and O. Bar-Or. Growth, Maturation, and Physical Activ-
ity. 2nd ed. Champaign, IL: Human Kinetics, 2004.
 30 — Baxter-Jones, Eisenmann, and Sherar

13. Marshall, W.A., and J.M. Tanner. Variations in the pattern of pubertal changes in girls.
Arch. Dis. Child. 44:291-303, 1969.
14. Marshall, W.A., and J.M. Tanner. Variations in the pattern of pubertal changes in boys.
Arch. Dis. Child. 45:13-23, 1970.
15. Matsudo, S.M.M., and V.K.R. Matsudo. Self-assessment and physician assessment of
sexual maturation in Brazilian boys and girls: concordance and reproducibility. Am. J.
Hum. Biology. 6:451-455, 1994.
16. Mirwald, R.L. Saskatchewan growth and development study. In: Kinanthropometry II.
M. Ostyn, G. Beunen, and J. Simons (Eds.). Baltimore: University Park Press, 1980,
pp.289-305.
17. Mirwald, R.L., A.D. Baxter-Jones, D.A. Bailey, and G.P. Beunen. An assessment of
maturity from anthropometric measurements. Med. Sci. Sports Exerc. 34:689-94, 2002.
18. Nevill, A.M., R.L. Holder, A. Baxter-Jones, J.M. Round, and D.A. Jones. Modeling
developmental changes in strength and aerobic power in children. J. Appl. Physiol.
84:963-970, 1998.
19. Petersen, A.C., L. Crockett, M. Richards, and A. Boxer. A self-report measure of pu-
bertal status: reliability, validity, and initial norms. J. Youth Adoles. 17:117-33, 1988.
20. Reynolds, E.L., and J.V. Wines. Individual differences in physical changes associated
with adolescence in girls. Am. J. Dis. Child. 75:329-350, 1948.
21. Reynolds, E.L., and J.V. Wines. Physical changes associated with adolescence in boys.
Am. J. Dis. Child. 82:529-547, 1951.
22. Roche, A.F., H. Wainer, and D. Thissen. The RWT method for the prediction of adult
stature. Pediatrics. 56:1026-1033, 1975.
23. Rowland, T. Developmental Physiology. Champaign, IL: Human Kinetics, 1996.
24. Scammon, R.E. The Measurement of Man. Minneapolis: University of Minnesota Press,
1930.
25. Sclosserberger, N.M., R.A. Turner, and C.E. Irwin. Validity of self-report pubertal
maturation in early adolescence. J. Adol. Health. 13:109-113, 1992.
26. Schmidt-Nielson, K. Scaling: Why is Animal Size so Important? Cambridge: Cam-
bridge University Press, 1984.
27. Tanner, J.M. Fallacy of per-weight and per-surface area standards, and their relation to
spurious correlation. J. Appl. Physiol. 2:1-15, 1949.
28. Tanner, J.M. Growth at Adolescence. Oxford: Blackwell Scientific Publications, 1962.
29. Tanner, J.M. A History of the Study of Human Growth. Cambridge: Cambridge Uni-
versity Press, 1981.
30. Tanner, J.M. Foetus Into Man. London: Castlemead, 1989.
31. Tanner, J.M., M.J.R. Healy, H. Goldstein, and N. Cameron. Assessment of Skeletal
Maturity and Prediction of adult height (TW3 method), 3rd ed. London: Saunders,
2001.
32. Tanner, J.M., R.H. Whitehouse, N. Cameron, W.A. Marshall, M.J.R. Healy, and H.
Goldstein. Assessment of Skeletal Maturity and Prediction of Adult Height. New York:
Academic, 1983.
33. Wacharasindhu, S., P. Pri-ngam, and T. Kongchonrak. Self-assessment of sexual matu-
ration in Thai children by Tanner photograph. J. Med. Ass. Thailand. 85:308-319, 2002.
34. Wainer H., A.F. Roche, and S. Bell. Predicting adult stature without skeletal age and
without parental data. Pediatrics. 61:569-572, 1978.
35. Williams R.L., K.L. Cheyne, L.K. Houtkooper, and T.G. Lohman. Adolescent self-
assessment of sexual maturation: effects of fatness classification and actual matura-
tion stage. J. Adol. Health Care. 9:480-482, 2003.
36. Winter E.M., F.B.C. Brookes, and E.J. Hamley. Maximal exercise performance and
lean leg volume in men and women. J. Spt. Sci. 9:3-13, 1991.

View publication stats