5 Bala, G. (2001). Dependence of the motor dimension difinition on the mode of result registration procedure… KinSI 7(1–2), 5–12
DEPENDENCE OF THE MOTOR DIMENSION
DEFINITION ON THE MODE OF RESULT
REGISTRATION PROCEDURE OF MOTOR
TEST PERFORMANCE
ODVISNOST OPREDELITVE MOTORI^NIH
RAZSE@NOSTI OD NA^INA IZRA^UNA
REZULTATA PRI MOTORI^NIH TESTIH
Gustav Bala
Abstract
On a sample of 260 first and second year students of
the Faculty of Physical Education in Novi Sad - Yugo-
slavia, four factor analyses were applied on the cor-
relation matrices with 21 variables, obtained on the
basis of different result registration procedures of mo-
tor test performance. It was concluded that the best-
fitted and most parsimonious solution was the case
when the test results were the factor scores of first
principal component of the correlation matrix of ev-
ery repetition result in every motor test.
Keywords: result registration, motor test, motor abili-
ties, factor analysis, metric characteristics
Izvle~ek
Na vzorcu 260 {tudentov prvega in drugega letnika
Fakultete za telesno vzgojo v Novem Sadu (Jugoslavi-
ja), je bila izvedena primerjava {tirih faktorskih analiz
korelacijskih matrik 21 kompozitnih motori~nih te-
stov, dobljenih z razli~nimi na~ini izra~una kon~nih
rezultatov merjencev iz posameznih ponovitev mer-
jenja. Ugotovljeno je bilo, da je bila dobljena najbolj
racionalna in ustrezna re{itev, ko so bili kon~ni rezul-
tati merjencev faktorske vrednosti prve glavne kom-
ponente korelacijske matrike vseh ponovitev vsakega
testa.
Klju~ne besede: motori~ni test, na~in izra~una rezul-
tata, motori~ne sposobnosti, faktorska analiza, merske
zna~ilnosti
Contact address
Gustav Bala
University of Novi Sad – Faculty of Physical Education
Lov}enska 16
YU-21000 Novi Sad
Yugoslavia
E-mail: BalaG@eunet.yu
(Received: 19.02.2001 – Accepted: 4.10.2001)

INTRODUCTION
Motor tests measure very complex human abilities.
These abilities, which are called »motor abilities«,
can’t be measured directly, but we must make conc-
lusions about them on the basis of a larger number of
indicators (motor tests), which never show the exact
level of the measured abilities, with equal reliability
and validity. For these reasons, motor tests should be
constructed from a larger number of items; usually
they are the same motor tasks, which the subjects
need to perform several times successively, without
or with a short pause. Such realised motor tests with
several items are called »composite motor tests«.
With the composite motor test, there are problems
about: the definition of internal metric characteristics
of the motor test as a whole, as well as the definition
of metric characteristics of items, from which the mo-
tor test is composed. Beside those problems, it is ne-
cessary to define the result registration procedures of
motor test performance, with the intention to get
good measure exactness. Solving all these problems,
we can get information about the reliability, represen-
tativeness, homogeneity, and validity of the used
items, as well as about the reliability, representative-
ness and homogeneity of the total result in the motor
test. Validity of the total result in the motor test should
be defined on the basis of external metric characteri-
stics (factor and pragmatic validity). The reader can
find more on measurement instruments in Momiro-
vi}, Wolf and Popovi} (1999).
The aim of this paper is to analyse the factor validity
of the motor tests, with regard to the pattern and
structure of entire analysed motor space, on the ba-
sis of different testing results of the same motor test,
being repeated several times. 
METHODS
Subject sample
The sample of subjects consisted of 260 first and se-
cond year students of the Faculty of Physical Educa-
tion in Novi Sad – Yugoslavia, 18-22 years of age,
which were selected according to their biological,
health, motor and psychological development; and
were tested at the entrance examination and during
the school year.
Motor test sample
The sample of motor variables was obtained on the
basis of the motor model according to Kureli} et al.
(1975); and Gredelj, Metiko{, Ho{ek and Momirovi}
(1975):
A) MECHANISM FOR MOVEMENT STRUCTURING
I Functional co-ordination of primary motor abi-
lities
1 The test »Agility on the floor« (CAGFLOOR)
2 The test »Dragging and jumping over«
(CDRAJUMP)
3 The test »Co-ordination with stick« (CCO-
STICK)
4 The test »Figure-8 duck« (CFI8DUCK)
B) MECHANISM FOR TONUS AND SYNERGETIC
REGULATION
II Balance
5 The test »One foot cross balance-eyes open«
(B1CROPEN)
6 The test »One foot length-wise balance-eyes
open (B1LEOPEN)
7 The test »Flamingo« (BFLAMING)
III Frequency of simple movements
8 The test »Two foot tapping« (F2FOOTTA)
9 The test »Arm plate-tapping« (FTAPPING)
10 The test »One foot-tapping« (F1FOOTTA)
III Flexibility
11 The test »Toe touching-sitting straddle«
(FLTOESIT)
12 The test »Toe touching-standing« (FLTOESTA)
13 The test »Push off one leg-lying on the side«
(FLPUSH1L)
C) MECHANISM FOR EXCITATION INTENSITY REG-
ULATION
IV Explosive strength
14 The test »Standing broad jump« (ESTANJUM)
15 The test »20m dash« (E20MDASH)
16 The test »Spring forward from front support
on the floor« (EREFLOOR)
D)MECHANISM FOR EXCITATION DURATION REG-
ULATION
V General strength
17 The test »Bent arm hang« (SARMHANG)
18 The test »Pull-ups« (SPULLUPS)
19 The test »Horizontal hold laying on the back«
(SHORHOLD)
20 The test »Sit-ups« (SSIT-UPS)
21 The test »Hand grip« (SHANDGRI).
The coding of the motor tests was as follows: the first
letter in the code name was according to the hypot-
hetical motor factor and the others according to the
names of the tests. The reason for that was to make
interpretation of analysed factors easier. This means
that the first letter in the code name C was for co-or-
dination, B for balance, F for frequency of simple mo-
vements, FL for flexibility, E for explosive strength and
S for general strength.
6 Bala, G. (2001). Dependence of the motor dimension difinition on the mode of result registration procedure… KinSI 7(1–2), 5–12

Table 2. EIGENVALUES ( λ), % OF VARIANCE AND CUMULATIVE % OF PRINCIPAL COMPONENTS OF
CORRELATION MATRICES OBTAINED ON THE BASIS OF DIFFERENT RESULT REGISTRATION PRO-
CEDURES
The conditions and techniques of measuring for most
of the tests were according to Metiko{, Prot, Hofman,
Pintar and Oreb (1989); except the tests »Flamingo«
(Moravec, 1996),  »Push off one leg-lying on the side«
and »Spring forward from front support on the floor«
(Madi}, 1995; Madi}, 2000). Every test was perfor-
med three times, so each of them was a composite
test of three items.
The entire analysis of motor space were made on the
basis of these procedures of result registration in mo-
tor testing of the same test: 
1 in every test the final result was the first one
of three repetitions; 
2 the best result of the three repetitions; 
3 the mean (or sum) of the results of the three
repetitions; 
4 the factor scores of the first principal compo-
nent of the correlation matrix of every repeti-
tion result.
Data analysis
In all four analyses, promax transformation of the prin-
cipal components of the corresponding correlation
matrix was applied. The number of statistical signifi-
cant principal components was obtained on the ba-
sis of the Guttman-Kaiser criterion.
RESULTS
The reliabilities of these tests were computed as Cron-
bach’s α-coefficient ( α). The values of the reliabilities
are shown in Table 1.
It is obvious that all the motor tests have quite good
reliabilities, except the tests SARMHANG (Bent-arm
hang) and B1CROPEN (One foot cross balance-eyes
open).
In the analysis of the complete motor space of 21 va-
riables, the most important results of the promax
transformation of principal components of the four
inter-correlation matrices were used. The obtained
information is given in the tables that follow. The com-
munalities and the principal components are not
shown due to the restricted length of papers.
It is obvious from Table 2, that the first three proce-
dures produced seven motor factors and when the
test result was the factor score of the first principal
component of the correlation matrix of every repeti-
tion result, the significant number of motor factors
was reduced to six. It seams that this procedure pro-
duced a simpler solution than the others, especially
the procedures with only one repetition (the first and
the best repetition). Such information can be seen
from the proportion of variance of every isolated prin-
cipal component and the cumulative percentage of
all isolated principal components.
When the factor analysis of motor space was made on
the basis of procedures of result registration taking in
7 Bala, G. (2001). Dependence of the motor dimension difinition on the mode of result registration procedure… KinSI 7(1–2), 5–12
TEST α TEST α
1. CAGFLOOR .92 12. FLTOESTA .94
2. CDRAJUMP .93 13. FLPUSH1L .99
3. CCOSTICK .88 14. ESTANJUM .94
4. CFI8DUCK .89 15. E20MDASH .91
5. B1CROPEN .76 16. EREFLOOR .95
6. B1LEOPEN .90 17. SARMHANG .62
7. BFLAMING .92 18. SPULLUPS .95
8. F2FOOTTA .92 19. SHORHOLD .97
9. FTAPPING .85 20. SSIT-UPS .92
10. F1FOOTTA .87 21. SHANDGRI .95
11. FLTOESIT .93
λ % of Variance Cumulative %
ABCDABCDABCD
1 4.08 4.21 4.60 4.66 19.42 20.05 21.94 22.19 19.42 20.05 21.94 22.19
2 1.67 1.78 1.71 1.73 7.95 8.48 8.15 8.26 27.38 28.54 30.09 30.46
3 1.53 1.58 1.58 1.57 7.30 7.52 7.54 7.48 34.68 36.06 37.63 37.94
4 1.45 1.41 1.40 1.41 6.92 6.75 6.69 6.75 41.61 42.82 44.33 44.70
5 1.22 1.25 1.21 1.19 5.84 5.98 5.78 5.67 47.45 48.80 50.11 50.38
6 1.08 1.12 1.13 1.12 5.15 5.36 5.41 5.35 52.61 54.16 55.53 55.73
7 1.05 1.02 1.01 5.04 4.87 4.84 57.66 59.04 60.37
A – the first one of three repetitions; 
B – the best result of three repetitions; 
C – the mean (or sum) of the results of three repetitions; 
D – the factor scores of first principal component of the correlation matrix of every repetition result
Table 1. RELIABILITIES OF THE MOTOR TESTS

every test the result of first of three repetitions, the fi-
nal solutions are in Table 3 and Table 4.
Table 3 shows the pattern and structure matrices af-
ter promax rotation of the principal components, with
grey boxes for the values of the variables, which were
involved significantly in defining the motor factors.
The pattern coefficients are the validities of applied
motor tests in this research. According to these coef-
ficients, as well as structure coefficients and correla-
tion coefficients between the factors, in this analysis,
the obtained motor factors were interpreted as:
1 Frequency of simple movements (FREQUENCY)
2 Flexibility (FLEXIBILITY)
3 The speed of solving complex motor problems
(COORDINATION)
4 Continuous regulation of muscle force (FORCE-
REGUL)
5 Balance (BALANCE)
6 Dual factor, defined by explosive strength (EXPLO-
SIVE)
7 Single factor, defined by variable of absolute isome-
tric strength of upper extremities (ISOMETRIC).
All correlation coefficients, which are greater than
0.11, are statistically significant with a 1% error level
and those greater than 0.15 with a 5% error level.
These point to a real relationship of motor factors (abi-
lities) (grey boxes in Table 4).
The final solution of factor analysis in the case, when
the best result of three repetitions was the test score,
is given in Table 5 and Table 6. 
1 The speed of solving complex motor problems
(COORDINATION)
2 Frequency of simple movements (FREQUENCY)
3 Flexibility (FLEXIBILITY)
4 Absolute strength of upper extremities and trunk
(STRENGTH)
8 Bala, G. (2001). Dependence of the motor dimension difinition on the mode of result registration procedure… KinSI 7(1–2), 5–12
Table 3. PATTERN AND STRUCTURE MATRICES OF FIRST REPETITION
Pattern Matrix Structure Matrix
VARIABLES 1 2 3 4 5 6 71234567
CAGFLOOR .13 -.10 .80 .13 -.04 -.01 -.08 -.19 -.13 .71 -.15 -.13 .09 -.12
CDRAJUMP -.38 -.01 .27 .05 -.03 .01 -.24 -.52 -.17 .41 -.14 -.19 .11 -.32
CCOSTICK -.04 -.11 .83 .19 -.06 -.01 .07 -.34 -.16 .78 -.15 -.15 .12 .00
CFI8DUCK -.26 .08 .38 -.05 .30 .29 -.07 -.39 -.03 .49 -.22 .18 .41 -.08
B1CROPEN .48 -.05 .22 .12 .44 .03 -.04 .50 .10 -.07 .23 .53 .00 .09
B1LEOPEN .02 .02 -.05 -.18 .87 -.07 -.18 .18 .09 -.13 .05 .81 -.02 -.02
BFLAMING .04 -.02 .09 -.09 -.69 .05 -.16 -.20 -.14 .24 -.28 -.75 .06 -.28
F2FOOTTA .70 -.06 -.08 -.04 -.06 -.07 -.07 .70 .10 -.32 .13 .07 -.18 .02
FTAPPING .74 .02 .04 .03 .03 -.05 -.06 .74 .20 -.26 .19 .18 -.16 .06
F1FOOTTA .54 .13 .14 .03 .07 -.06 .22 .58 .31 -.11 .13 .22 -.15 .33
FLTOESIT -.05 .87 -.06 .12 -.02 -.12 -.06 .21 .87 -.16 .22 .09 -.24 .08
FLTOESTA -.04 .85 -.16 .00 .08 .02 -.02 .24 .85 -.22 .15 .19 -.10 .13
FLPUSH1L .40 .55 .04 -.02 -.08 .22 .07 .48 .62 -.11 .06 .08 .09 .20
ESTANJUM -.12 .10 -.02 .23 .06 -.78 .02 .09 .19 -.20 .34 .07 -.80 .07
E20MDASH -.27 .08 -.03 .14 -.08 .76 .08 -.33 -.06 .15 -.02 -.05 .76 .00
EREFLOOR .00 .04 .14 .49 -.02 -.09 .13 .09 .10 -.07 .45 .09 -.15 .12
SARMHANG .04 -.04 -.05 .06 -.06 .05 .88 .17 .11 -.10 .06 .11 -.01 .87
SPULLUPS -.06 -.16 -.31 .53 .21 .07 .21 .20 -.05 -.50 .65 .38 -.03 .21
SHORHOLD .15 .02 .10 .69 -.06 .08 -.03 .24 .09 -.19 .66 .11 -.02 -.03
SSIT-UPS .15 -.05 -.34 .45 -.02 .11 -.41 .28 -.03 -.53 .59 .09 .00 -.40
SHANDGRI -.14 .13 .23 .71 -.08 -.12 -.05 -.05 .14 .01 .60 .01 -.18 -.07
Table 4. CORRELATION MATRIX OF FACTORS OF
FIRST REPETITION
FACTORS 1234567
1. FREQUENCY 1.00 .24 -.38 .22 .22 -.15 .15
2. FLEXIBILITY .24 1.00 -.07 .09 .11 -.12 .17
3. COORDINATION -.38 -.07 1.00 -.38 -.15 .16 -.04
4. FORCE-REGUL .22 .09 -.38 1.00 .22 -.14 -.01
5. BALANCE .22 .11 -.15 .22 1.00 .04 .17
6. EXPLOSIVE -.15 -.12 .16 -.14 .04 1.00 -.05
7. ISOMETRIC .15 .17 -.04 -.01 .17 -.05 1.00

5 Balance (BALANCE)
6 Dual factor, defined by tests of explosive strength
(EXPLOSIVE)
7 Dual factor, defined by variables of trunk strength
and maximal force of attempted movements of
hand (STRE-FORCE)
When the mean (or sum) of the results of three repe-
titions was the result registration procedure of motor
test performance, the factor analysis solution is shown
in Table 7 and Table 8.
1 The speed of solving complex motor problems
(COORDINATION)
2 Flexibility (FLEXIBILITY)
3 Frequency of simple movements (FREQUENCY)
4 Continuous regulation of muscle force (FORCE-
REGUL)
5 Balance (BALANCE)
6 Dual factor, defined by two tests of explosive
strength (EXPLOSIVE)
7 Single factor, defined by variable of absolute isome-
tric strength of upper extremities (ISOMETRIC).
And finally, when the factor analysis of motor space
was made on the basis of procedures of result registra-
tion taking in analysis of the factor scores of first prin-
cipal component of the correlation matrix of every re-
petition result, the final solution is in Table 9 and Table
10.
1 Balance (BALANCE)
2 The speed of solving complex motor problems
(COORDINATION)
3 Flexibility (FLEXIBILITY)
4 Continuous regulation of muscle force (FORCE-
REGUL)
5 Frequency of simple movements (FREQUENCY)
6 Dual factor, defined by two variables of explosive
strength (EXPLOSIVE).
9 Bala, G. (2001). Dependence of the motor dimension difinition on the mode of result registration procedure… KinSI 7(1–2), 5–12
Table 5. PATTERN AND STRUCTURE MATRICES OF BEST REPETITON
Table 6. CORRELATION MATRIX OF FACTORS OF
BEST REPETITION
VARIABLES Pattern Matrix Structure Matrix
12345671234567
CAGFLOOR .78 .18 -.06 .07 .01 -.08 -.09 .69 -.11 -.13 -.18 -.13 -.15 -.08
CDRAJUMP .52 -.19 -.04 -.10 .02 .12 .02 .62 -.44 -.15 -.34 -.16 .05 .12
CCOSTICK .81 -.01 -.10 .31 -.09 -.13 -.12 .73 -.26 -.21 -.06 -.22 -.17 -.10
CFI8DUCK .43 -.15 .04 -.31 .37 -.10 .21 .54 -.38 -.06 -.46 .17 -.16 .28
B1CROPEN .08 .30 -.04 .11 .57 -.12 -.06 -.21 .42 .08 .30 .63 -.09 -.12
B1LEOPEN .03 .04 -.01 -.09 .81 .01 .00 -.14 .15 .08 .12 .79 .01 .02
BFLAMING .10 .07 -.07 -.11 -.65 -.14 .02 .28 -.16 -.17 -.32 -.69 -.20 .00
F2FOOTTA -.04 .88 -.10 -.27 .00 .10 .22 -.27 .74 .10 .04 .11 .09 .09
FTAPPING .08 .76 .05 .05 .06 .10 .12 -.26 .75 .25 .31 .21 .15 .00
F1FOOTTA .11 .57 .06 .16 .12 .08 -.33 -.25 .68 .22 .39 .25 .10 -.42
FLTOESIT .04 .00 .85 .11 -.04 .10 .06 -.11 .21 .87 .20 .08 .24 .08
FLTOESTA -.16 -.13 .84 .02 .08 .00 -.05 -.25 .17 .85 .17 .19 .13 -.03
FLPUSH1L -.10 .35 .57 -.08 -.04 -.27 .03 -.25 .48 .62 .08 .08 -.18 -.06
ESTANJUM -.07 .01 .07 .03 .05 .80 .10 -.17 .09 .20 .24 .11 .84 .19
E20MDASH .09 -.24 .09 .07 .01 -.72 .13 .22 -.28 -.06 -.20 -.05 -.69 .09
EREFLOOR .12 -.09 .04 .58 -.01 .02 .13 -.06 .05 .08 .50 .10 .15 .11
SARMHANG -.13 -.15 .01 .63 .22 .02 -.16 -.38 .19 .09 .70 .37 .14 -.19
SPULLUPS -.35 -.01 -.21 .46 .14 -.06 .37 -.49 .19 -.10 .56 .31 .09 .30
SHORHOLD .19 .04 .07 .75 -.07 -.07 .30 -.08 .17 .14 .65 .10 .12 .24
SSIT-UPS -.29 .14 -.03 .16 .01 -.13 .75 -.33 .17 .06 .23 .15 .02 .67
SHANDGRI .22 .06 .11 .33 -.09 .24 .57 .10 .00 .16 .26 -.02 .38 .58
FACTORS 1234567
1. COORDINATION 1.00 -.40 -.14 -.39 -.23 -.09 .08
2. FREQUENCY -.40 1.00 .24 .34 .19 .03 -.18
3. FLEXIBILITY -.14 .24 1.00 .12 .11 .13 .01
4. STRENGHT -.39 .34 .12 1.00 .25 .21 -.08
5. BALANCE -.23 .19 .11 .25 1.00 .03 .02
6. EXPLOSIVE -.09 .03 .13 .21 .03 1.00 .12
7. STRE-FORCE .08 -.18 .01 -.08 .02 .12 1.00

DISCUSSION
The obtained pattern and structure matrices showed
that in every case, there were six factors named in the
same way, and their patterns and structures are simi-
lar, but not the same. The first three cases produced
seven factors, and the seventh factor in every case
was a single or dual factor. In the case, where a mo-
tor test result was the score on the first principal com-
ponent, the factor analysis produced six quite well
defined factors. At first sight it is obviously that in this
case the applied factor analysis produced a more par-
simonious solution than in the other cases. Advanta-
ge of the last result registration procedure was obtai-
ned also in the research with pre-school children (Bala
1999).
Frequency of simple movements was best defined,
when all items are considered in the test, which real-
ly measure this ability. Even one item is good, but it
must be the best test score of the three performed
items. But as the pattern and the structure coefficients
of this factor showed in every case (Table 3, 5, 7 and
9), the motor test »One foot tapping« is a poor esti-
mator of frequency of simple movements. This means
that this test had very poor validity. This group of va-
riables (Two-foot tapping – F2FOOTTA, Arm plate tap-
ping – FTAPPING and One foot tapping – F1FOOT-
TA) had the lowest communalities in the entire motor
space, and »One foot tapping« had the lowest value
in this group.
Flexibility is quite a well-defined factor in every analy-
sed case. Similarity of performance of the test »Toe
touching-sitting straddle« (FLTOESIT) and »Toe touc-
hing-standing« is obvious, so their variables had lite-
rally the same projections on this factor. It is very in-
teresting that this factor had no statistically significant
correlations with the other four motor factors, in the
case when only one item – first or best repetition – is
taken in the analysis. These two tests had very good
validities of obtained factors named »flexibility«. The
10 Bala, G. (2001). Dependence of the motor dimension difinition on the mode of result registration procedure… KinSI 7(1–2), 5–12
Table 7. PATTERN AND STRUCTURE MATRICES OF MEANS OF REPETITIONS
VARIABLES Pattern Matrix Structure Matrix
12345671234567
CAGFLOOR .79 -.07 .12 .06 .02 -.03 -.04 .74 -.14 -.19 -.17 -.17 -.17 -.21
CDRAJUMP .40 .00 -.15 .05 .03 .00 -.37 .52 -.16 -.41 -.08 -.20 -.16 -.51
CCOSTICK .81 -.13 .00 .18 -.10 -.05 .11 .79 -.22 -.31 -.10 -.27 -.18 -.14
CFI8DUCK .39 .03 -.12 -.03 .23 -.32 -.32 .51 -.14 -.36 -.16 -.02 -.46 -.45
B1CROPEN .10 -.13 .22 -.09 .63 .07 .13 -.15 .04 .35 .06 .66 .15 .34
B1LEOPEN -.05 .02 .05 -.09 .85 .00 -.15 -.23 .12 .23 .14 .82 .07 .12
BFLAMING .09 -.09 .14 -.10 -.65 .04 -.17 .29 -.20 -.13 -.28 -.72 -.09 -.34
F2FOOTTA -.01 -.07 .85 -.02 .04 .11 -.14 -.30 .13 .81 .06 .20 .21 .16
FTAPPING .06 .05 .77 .17 .03 .00 -.01 -.29 .26 .78 .22 .25 .16 .26
F1FOOTTA .14 .17 .41 -.06 .14 .01 .42 -.15 .36 .57 -.01 .33 .17 .60
FLTOESIT .01 .83 -.03 .12 -.03 .13 -.02 -.14 .85 .20 .22 .12 .28 .16
FLTOESTA -.19 .84 -.10 -.04 .03 -.01 .01 -.26 .84 .19 .10 .17 .14 .20
FLPUSH1L -.02 .57 .34 -.02 -.02 -.21 .07 -.18 .63 .47 .02 .14 -.06 .25
ESTANJUM -.04 .10 -.08 .13 .06 .80 -.09 -.21 .22 .09 .32 .16 .83 .09
E20MDASH .06 .09 -.18 .11 .01 -.75 -.02 .23 -.07 -.30 -.06 -.10 -.76 -.22
EREFLOOR .10 .07 -.23 .46 .07 .07 .29 -.06 .13 -.09 .45 .20 .19 .24
SARMHANG .02 -.01 -.13 .17 .02 -.06 .83 -.18 .14 .14 .16 .25 .12 .78
SPULLUPS -.28 -.22 -.04 .47 .21 .00 .24 -.49 -.06 .17 .58 .42 .18 .31
SHORHOLD .13 .02 .14 .80 -.15 -.12 .21 -.15 .12 .16 .71 .09 .07 .18
SSIT-UPS -.43 -.04 .20 .55 -.01 -.14 -.32 -.55 .03 .25 .65 .17 .02 -.18
SHANDGRI .27 .11 .06 .59 .03 .29 -.19 .04 .16 .02 .59 .11 .35 -.14
Table 8. CORRELATION MATRIX OF FACTORS OF
MEANS OF REPETITIONS
FACTORS 1234567
1. COORDINATION 1.00 -.13 -.37 -.30 -.26 -.19 -.24
2. FLEXIBILITY -.13 1.00 .25 .10 .15 .16 .21
3. FREQUENCY -.37 .25 1.00 .07 .25 .17 .33
4. FORCE-REGUL -.30 .10 .07 1.00 .24 .21 .01
5. BALANCE -.26 .15 .25 .24 1.00 .11 .28
6. EXPLOSIVE -.19 .16 .17 .21 .11 1.00 .20
7. ISOMETRIC -.24 .21 .33 .01 .28 .20 1.00

third one had poor validity and should not be used to
measure general flexibility (maybe specific flexibility
for some specific kinesiological activities, for instance:
artistic and rhythmic gymnastics, artistic skating, kara-
te, kick-boxing etc.).
Independence of the flexibility factor from most of
the other motor factors pointed to the existence of
the problem linked to the nature of flexibility.
The speed of solving complex motor problems (co-
ordination) is the factor, which had very high projec-
tions of two variables (test »Agility on the floor« –
CAGFLOOR and test »Co-ordination with stick« –
CCOSTICK), and much smaller projections of the va-
riables »Figure-8 duck« (CFI8DUCK) and »Dragging
and jumping over« (CDRAJUMP). The best result was
obtained in the case, when a motor test result was the
factor score of the first principal component of the
correlation matrix of every repetition (item) in the cor-
responding test. In that case, the correlation coeffi-
cients between this factor and all other motor factors
were statistically significant. Validity of the applied
motor tests for such co-ordination was very good for
»Agility on the floor« and »Co-ordination with stick«,
but very poor for the other two tests.
Continuous regulation of muscle force is the factor
which had a very similar pattern and structure in all
cases of the analysed result registration procedures,
except when the result was the score of the best re-
petition. The hypothetical sample of motor variables
which should estimate general strength, was divided
into two groups and defined two factors: 1) continu-
ous regulation of muscle force, and 2) dual factor, de-
fined by variables of trunk strength and maximal for-
ce of attempted movement of hand – when the best
repetition of every test was analysed.
In other analyses, hypothetical general strength was
defined by almost all the variables mentioned above,
11 Bala, G. (2001). Dependence of the motor dimension difinition on the mode of result registration procedure… KinSI 7(1–2), 5–12
Table 9. PATTERN AND STRUCTURE MATRICES OF FIRST PRINCIPAL COMPONENT OF THE CORRE-
LATION MATRIX OF EVERY REPETITION RESULT
Table 10. CORRELATION MATRIX OF FACTORS
OF FIRST PRINCIPAL COMPONENT OF THE
CORRELATION MATRIX OF EVERY REPETITION
RESULT
VARIABLES Pattern Matrix Structure Matrix
123456123456
CAGFLOOR .06 .80 -.06 .06 .09 -.05 -.23 .75 -.15 -.15 -.16 -.20
CDRAJUMP -.25 .40 -.02 .19 -.12 -.14 -.44 .52 -.21 .02 -.35 -.29
CCOSTICK .04 .80 -.11 .11 -.03 .01 -.27 .78 -.23 -.09 -.29 -.17
CFI8DUCK .02 .42 -.01 .08 -.08 -.45 -.28 .52 -.20 -.10 -.26 -.54
B1CROPEN .68 .12 -.07 -.04 .22 -.07 .65 -.16 .12 .03 .36 .09
B1LEOPEN .72 .06 -.06 .02 .12 -.16 .67 -.20 .10 .11 .27 .03
BFLAMING -.72 .04 -.06 -.08 .10 .08 -.71 .28 -.20 -.22 -.12 -.14
F2FOOTTA -.04 -.04 -.09 .05 .82 .14 .24 -.30 .15 .04 .81 .21
FTAPPING .04 .03 .07 .19 .74 .06 .32 -.28 .29 .18 .76 .20
F1FOOTTA .37 .10 .21 -.16 .37 .14 .51 -.17 .41 -.10 .52 .29
FLTOESIT -.07 .01 .85 .17 -.04 .08 .18 -.14 .85 .20 .15 .30
FLTOESTA .00 -.17 .86 .00 -.11 -.06 .24 -.26 .84 .07 .16 .18
FLPUSH1L .04 -.04 .60 -.03 .29 -.19 .24 -.19 .65 -.03 .45 .01
ESTANJUM -.09 -.01 .04 .23 -.02 .75 .17 -.20 .21 .34 .05 .77
E20MDASH .07 .03 .13 .06 -.21 -.78 -.18 .20 -.11 -.05 -.26 -.75
EREFLOOR .29 .12 .09 .33 -.22 .16 .31 -.05 .14 .39 -.15 .26
SARMHANG .57 -.10 .03 .02 -.23 .17 .60 -.29 .18 .19 -.01 .34
SPULLUPS .39 -.29 -.20 .39 -.05 .04 .52 -.51 -.04 .54 .10 .22
SHORHOLD .15 .11 .06 .68 .08 -.02 .27 -.15 .13 .67 .08 .13
SSIT-UPS -.11 -.41 -.05 .62 .20 -.21 .14 -.54 .00 .66 .23 -.04
SHANDGRI -.14 .24 .11 .69 .10 .20 .01 .03 .14 .63 .01 .26
FACTORS 123456
1. BALANCE 1.00 –.37 .26 .18 .29 .28
2. COORDINATION –.37 1.00 –.15 –.26 –.30 –.22
3. FLEXIBILITY .26 –.15 1.00 .04 .26 .26
4. FORCE-REGUL .18 –.26 .04 1.00 –.04 .16
5. FREQUENCY .29 –.30 .26 –.04 1.00 .12
6. EXPLOSIVE .28 –.22 .26 .16 .12 1.00

which were named as continuous regulation of musc-
le force. The exception was the variable SARMHANG
(Bend arm hang), which made a »single factor, defi-
ned by the variable of absolute isometric strength of
upper extremities«.
Balance is a motor factor, which was defined well in
all four cases, but the most homogenous solution was
obtained, when the motor test result was considered
as a first principal component of the correlation ma-
trix of every repetition. This factor had statistically sig-
nificant correlations with all other factors in the men-
tioned case and when the result of test was the
average (or sum) of repetitions in every motor test. In
the other two cases, there were no significant corre-
lations with flexibility and explosive strength (Table 4)
and flexibility, explosive strength and dual factor, de-
fined by variables of trunk strength and maximal for-
ce of attempted movements of hand (Table 6). The
test »One foot length-wise balance-eyes open«
(B1LEOPEN) had the best validity and the test »Fla-
mingo« (BFLAMING) a lower one. Balance test with
the lowest validity was »One foot cross balance-eyes
open« (B1CROPEN).
Explosive strength was defined very well only by two
variables (»Standing broad jump« – ESTANJUM and
»20m dash« – E20MDASH), but the third hypotheti-
cal explosive variable (»Spring forward from front sup-
port on the floor« – EREFLOOR) had validity too low
to be taken as a measure of this motor ability.
Isometric strength was obtained in two cases as a sin-
gle factor, with very high validity of the variable »Bent
arm hang« (SARMHANG), or in combination with the
variable to estimate the absolute strength of upper ex-
tremities and trunk, or even to estimate balance,
which has in common excitation duration regulation
(tonus and synergetic, but also strength regulation).
CONCLUSIONS
When the test results are used for scientific purposes,
from a strictly theoretical point of view, in order to
obtain good motor test results, the researchers should
use composite motor tests with several items in every
test. The items are the same motor tasks, which the
subject needs to perform several times successively,
without or with a short pause. Of course, standardi-
sed performance instructions of every test must be
fully respected. Preference should be given to the pro-
cedure that assumes computing the factor scores of
the first principal component of the correlation matrix
of every performance of the same test. In that way
the metric characteristics of motor tests should be
adequate.
Certain preference should be given also to the sum-
results registration procedure of all repetitions of the
same test. But it is also worthwhile to take into consi-
deration the best result among more repetitions of the
same test. 
REFERENCES
1. Bala, G. (1999). Motor behaviour evaluation of pre-school children
on the basis of different result registration procedures of motor test
performance. In V. Strojnik, & A. U{aj (Eds.), Proceedings I. – 6.
Sport Kinetics Conference ’99 »Theories of Human Motor Perfor-
mance and their Reflections in Practice« (pp. 62-65). Ljubljana: Uni-
versity of Ljubljana, Faculty of Sport.
2. Gredelj, M., Metiko{, D., Ho{ek, A., & Momirovi}, K. (1975). Mo-
del hijerarhijske strukture motorickih sposobnosti. 1. Rezultati do-
bijeni primjenom jednog neoklasi~nog postupka za procjenu latent-
nih dimenzija. [Model of a hierarchic structure of motor abilities. 1.
The results obtained using a neo-classical method for estimating la-
tent dimensions]. Kineziologija, 5(1-2), 7-81.
3. Kureli}, N., Momirovi}, K., Stojanovi}, M., [turm, J., Radojevi}, \.,
& Viski}-[talec, N. (1975). Struktura i razvoj morfolo{kih i motori~-
kih dimenzija omladine [Structure and development of morpholo-
gic and motor dimensions of youth]. Belgrade: Institute for Research
of the Faculty of Physical Education.
4. Madi}, D. (1995). Konstrukcija i metrijske karakteristike motori~kih
testova specifi~ne gipkosti gimnasti~arki [Construction and metric
characteristics of motor tests of specific flexibility of female gym-
nasts]. Master’s thesis, Novi Sad: Fakultet fizi~ke kulture.
5. Madi}, D. (2000). Povezanost antropolo{kih dimenzija studenata fi-
zi~ke kulture sa njihovom uspe{no{}u ve`banja na spravama [Rela-
tionship between anthropological dimensions of physical education
students and successful exercising on gymnastic apparatuses]. Doc-
toral thesis, Novi Sad: Fakultet fizi~ke kulture.
6. Metiko{, D., Prot, F., Hofman, E., Pinter, @., & Oreb, G. (1989).
Mjerenje bazi~nih motori~kih dimenzija sporta{a [Measurement of
basic motor abilities of sportsmen]. Zagreb: Fakultet za fizi~ku kul-
turu.
7. Momirovi}, K., Wolf, B., & Popovi}, D. (1999). Uvod u teoriju me-
renja i interne metrijske karakteristike kompozitnih mernih instru-
menata [Introduction in the theory of measurement and internal
characteristics of composite measuring instruments]. Pri{tina: Fa-
kultet za fizi~ku kulturu.
8. Moravec, R. (Ed.). (1996). Eurofit – Physique and motor fitness of the
Slovak school youth. Bratislava: Slovak Scientific Society for Physi-
cal Education and Sports.
12 Bala, G. (2001). Dependence of the motor dimension difinition on the mode of result registration procedure… KinSI 7(1–2), 5–12