Reliability and Validity of the Four Station Fundamental Motor Test (4-SFMT) for Assessing Motor Skills in Preschool Children—A Pilot Study

Abstract

The main aim of this study was to create a new test to assess the motor skills of children of preschool and early school age and to determine its reliability and validity. Thirty children (5–6 years old) were tested on two occasions 14 days apart. The testing procedures included measuring the participants’ performance on the Four Station Fundamental Motor Test (4-SFMT). The newly constructed 4-SFMT tested four fundamental skills/tasks: space covering (ROLL), overcoming resistance (PULL), object control (BALL), and overcoming obstacles (CLIMB). The performance was evaluated with a 22-point scale with different criteria for each skill, and the time taken to perform the test was measured. The concurrent validity was assessed by determining the correlation with a Test of Gross Motor Development (TGMD-2). The level of agreement across trials was statistically significant for all three raters, with two variables presenting excellent reliability (ICC > 0.9) and two variables having good reliability (ICC > 0.75 and <0.9). No significant differences were found between the test and retest scores, indicating the test’s high reliability. A factor analysis isolated only one motor factor (accounting for 43.99% of the variance with the eigenvalue of 1.768) from the four tasks. There was a large correlation (r = −0.576, p < 0.01) between the process- and product-oriented assessments of the 4-SFMT. Moreover, significant correlations were found between the 4-SFMT and TGMD-2 regarding the scores (r = 0.824, p < 0.001) and time taken to perform the tasks (r = −0.652, p < 0.001), which indicates the good concurrent validity of the newly constructed test. Small to moderate correlations between tasks (0.016 to 0.497) and no differences between boys and girls in the total score (p = 0.943) and time taken to complete the tasks (p = 0.49) were determined. The 4-SFMT appears to be a valid and reliable tool that can be used to evaluate motor skills performance in children between the ages of 5 and 6 and is reasonably simple to use.

1. Introduction

Human motor development is a process that includes progressions and regressions of motor competence throughout life [1]. The term “motor competence” (MC) is globally understood and refers to the level of proficiency with which children can execute fundamental motor skills [2]. Process-oriented MC assessments evaluate how a movement is performed, while product-oriented assessments evaluate the outcome of a movement. It is widely believed that motor development is the most important form of development throughout the lifespan, and that it takes place during the most sensitive phase between birth and six years of age. [3]. In that period, as many as three phases of motor development change (reflex phase, phase of elementary movements, and phase of basic motor patterns). It is believed that children can hardly reach their full motor potential if their motor development is not stimulated in this specified period [4]. This is extremely important knowledge because motor competence (MC) in children and adolescents is directly related to numerous health-related outcomes including physical activity [5], physical fitness [6], lower body mass index [7], cardiorespiratory fitness [8], well-being [9], and even cognitive health [10]. Although there is a relatively large number of tests that assess MC in children and adolescents, most of them were developed for clinical purposes and are used to identify children with motor impairments or medical deficits [11]. Moreover, the implementation of most tests takes a long time [12], and a good number of them require special equipment and props, which also reduces their practicality, i.e., applicability. On the other hand, it is known that testing children in early childhood is very demanding, and care should be taken to ensure that it does not take too long, that the environment is safe, and that, in general, the administration of the test is not a negative experience for these children [13].

The results of some studies are interesting as they show that in early childhood, there is a significantly different structure of motor dimensions when compared to the older population. Namely, at an early age, there is no clear differentiation between different motor abilities and motor skills, as is the case in adults [14,15]. It seems that in early childhood, only one motor factor dominates, which determines the motor competence of the child. In practice, this would mean that by determining only one motor segment of the child (regardless of whether motor performance or qualitative motor achievements were tested), the general state of the MC of that subject could be assessed. This assessment should be more accurate as the test becomes more complex. Therefore, is it justified to carry out long-term testing protocols with a whole series of subtests for the assessment of motor competence if it is possible to determine this with one simple and quick test? This is particularly relevant in situations where the testing time is limited, such as in preschool and school institutions during physical education lessons. The goal is not to determine motor impairment or medical deficits but simply to assess some of the fundamental motor skills as a part of the motor competence of preschool- or early school-age children, or for the possible identification sports talent [16]. However, caution is needed and further research on the young athletes’ population is still needed. Motor competence is often assessed within the general population to identify children who may benefit from additional training. Nevertheless, motor competence assessments for sports and talent identification are becoming increasingly prevalent [17].

For the purposes of this research, such a test was constructed. Its duration is relatively short and can be applied as a product- and process-oriented test or even both. Process-oriented MC assessments evaluate how a movement is performed, while product-oriented assessments evaluate the outcome of a movement. Additionally, the newly constructed test has a relatively high ecological validity because the tasks are carried out alternately in continuity [18] and it does not require expensive equipment. Therefore, the main goal of this research was to create a new test to assess the motor skills of children of preschool and early school age and to determine some of its psychometric properties. For the purposes of the psychometric properties assessment, the test–retest method was applied, and a comparison was made with the TGMD-2. Moreover, it was hypothesized that there would be sex differentiation and moderate correlations between the test items.

2. Materials and Methods

2.1. Study Design

The Four Station Fundamental Motor Test (4-SFMT) construction and evaluation process consisted of several phases: the development stage (identification of the 4-SFMT items/tasks); the reliability study stage (assessment of test consistency, where two aspects were addressed, i.e., test–retest reliability and inter-rater and intra-rater reliability); and the validation stage (comparison with the criterion test for concurrent validity assessment and correlation between tasks with sex differentiation for construct validity).

2.2. Participants

In order to establish the psychometric properties of the 4-SFMT, a convenience sample of 30 preschool children (12 girls and 18 boys) from 5 to 6 years old was recruited. There were no behavioral, executive function (e.g., attention deficit hyperactivity disorder), or musculoskeletal issues or learning impairments among the children who participated in this experiment. Before the children participated in this study, their parents and guardians received a letter with information about the study and signed the informed consent form. Parents were clearly informed that the child could withdraw from the experiment at any time without giving a reason.

2.3. 4-SFMT Development

The 4-SFMT was created by a group of four experts with extensive knowledge in motor skill development, early childhood physical education, measurement methodologies, and sports pedagogy. The panel group was requested to evaluate each skill for age appropriateness and feasibility. Additionally, experts were asked to give their opinions and suggestions about particular skills and the test. The 4-SFMT was designed to follow the curriculum for physical education, which defined minimal standards for preschools of the Republic of Croatia [19]. Children’s physical education classes were analyzed in order to evaluate the tasks and skills that kids typically use in these settings. Furthermore, test items (tasks) were categorized into the following groups according to their utility [20]: (1) space-covering skills (various types of rolling, looping, crawling, walking, and running that allow humans to cover distances on various types of surfaces and in various directions); (2) overcoming resistance skills (a variety of pushing, pulling, holding, and carrying techniques used to overcome the passive resistance of objects of different volumes and shapes); (3) object-control skills (simple and complex operations of managing objects that differ in quantity, shape, and volume in a specific time and place by using a variety of throwing and catching, targeting, and shooting skills); and (4) overcoming obstacles skills (different forms of crawling through a narrow space, climbing, landing, and jumping that assists us in overcoming various types of vertical, diagonal, and horizontal obstacles without the use of any technical or other types of devices).

After careful examination, the test was deemed to include only one task from all four groups of basic motor skills, which fulfilled the study’s main aim of constructing a simple and short test. Therefore, the expert panel proceeded to develop a framework for the 4-SFMT by selecting the four tasks that best represent a certain movement skills area:

(1)

Rolling on a mat in an upward arms position, which measures space-covering skills (ROLL). The task was performed on a soft mat with dimensions of 200 cm × 100 cm and a thickness of 5 cm. The child had to roll with their hands in an upward position from the beginning to the end of the mat.

(2)

Pulling one’s body on a bench, which measures overcoming resistance skills (PULL). The task was performed on a wooden bench with a 340 × 26 surface and a height of 36 cm. The child had to pull his body from the beginning of the bench to the end. Only cotton shirts were allowed to avoid unnecessary friction due to the nature of the material (e.g., plastic print).

(3)

Pushing a ball over a bench, which measures object-control skills (BALL). The task was performed on the same wooden bench that was used in the PULL task. The child had to take a 400 g plastic ball with a diameter of 20 cm (which is usually used in rhythmic gymnastics) and guide it on the surface of the bench from the beginning to the end. The task was performed with the child’s dominant hand, which was determined by having the kid do three unimanual tasks: drawing a line with a pencil, cutting paper with scissors, and inserting a peg in the instructor’s hand. The dominant hand is the hand used for most tasks [21].

(4)

Climbing on a wooden ramp, which measures overcoming obstacles skills (CLIMB). The task was performed on an 85 cm × 250 cm wooden ramp that was set on a Swedish ladder that was 110 cm tall. The ramp had twenty-nine 15 cm × 8 cm holes, which made climbing easier. The child had to climb the board and touch the mark set at 130 cm.

The final version of the test consisted of four tasks that the children were required to perform in one trial in a circular manner (Figure 1). Every task was photographed and shown to the children to ensure they understood the type of movement selected for the test. The exact order of execution was as follows: ROLL, PULL, BALL, and CLIMB. The tasks were arranged in this order to ensure the participating children’s safety and allow for several transitions between various skill sets. Additionally, it was observed that the children occasionally had difficulty transitioning easily from one skill to the next, so the less complex skills were administered first. Moreover, the CLIMB task was administered last to properly ensure a safe environment for the children. The performance assessment was conducted in two ways: quantitatively (by measuring time) and qualitatively (by scoring each task).

2.4. Development of a 22-Point Skill Scale

To evaluate each task, a 22-point score assessment scale was developed. The scale consisted of different criterions for each task: three for the ROLL, BALL, and CLIMB, and two for the PULL. For each criterion, ratings of 0, 1, and 2 were assigned (a detailed description of each criterion is presented in

Table 1. Description of a 22-point skill score.

Task	Criterion	Description	Score
ROLL	arms	extended elbow joint + folded arms (outstretched and folded arms)	2
		some of the elements (spread, flexed)	1
		flexed elbow joint, abduction of the shoulder joint (bent and extended arms)	0
	movement direction	in a straight line	2
		small deviations	1
		turns, significant loss of the direction of straight movement	0
	legs	extended knees + contracted legs (extended and contracted legs)	2
		some of the elements	1
		flexed knees, abduction of the hip joint (bent and extended legs)	0
PULL	arms	sync + full range	2
		incomplete or excessive extent of withdrawal or one brief period where dominant arm was leading	1
		one of the arms is dominant	0
	legs	knees extended and contracted (legs extended and contracted)	2
		succeeds, but with flexed knees or extended knees and abduction in the hip joint	1
		falls to the side, pushes off with his toes	0
BALL	palm control	pushes the ball in a controlled manner with the entire palm	2
		uncomplete control—fingers or palm or occasionally helps with the other hand	1
		bounces the ball with fingers or leads with both palms	0
	the direction of ball movement	direction is followed	2
		zigzag movement	1
		falls	0
	ball disposal	the ball is disposed in a controlled manner	2
		disposal started but failed	1
		does not complete the task, does not dispose the ball, has no intention	0
CLIMB	movement pattern	contralateral pattern	2
		combination of two	1
		ipsilateral pattern	0
	body length	maximum use of body length	2
		incomplete, something in between	1
		shortened, incomplete movements	0
	the use of the climbing board	holes are used correctly	2
		occasional mistakes	1
		constant mistakes, does not enter the foot	0

). As a result, each participant had the chance to achieve a maximum of 22 points. The panel of three experts (research scientists with a Ph.D.) with a background in motor learning and skill development was recruited to establish the instrument’s face and content validity.

2.5. Procedures

The experiment was conducted in a sports hall that resembled a typical preschool gymnasium during the day. Individual tests on the children were conducted by the principal researcher and an assistant (both kinesiology experts) trained in the testing procedures. First, the children performed the 4-SFMT. The children were given clear instructions on how to perform the test in the exact prescribed order, starting with ROLL and ending with CLIMB. Each item (station) was demonstrated and described to the children. Following the principal researcher’s example, the children completed two try-out trials, during which verbal assistance and feedback on how to properly execute the skills were given to the participants. This feedback enabled the children to perform the measurement trial independently. The principal researcher instructed the child to prepare and start the test by saying “Go!” while the assistant started measuring time with a stopwatch. The test was finished when the child performed the last task in the test. Each child performed the test only one time. After the 4-SFMT, the children were instructed to perform the Test of Gross Motor Development (TGMD-2) [22]. The test is divided into two subtests (the locomotor skills subtest and the object-control skills subtest) with six skills each. The locomotor subtest consists of the following six skills: run, gallop, hop, leap, jump, and slide. The object-control subtest consists of the following six skills: striking a stationary ball, stationary dribble, catching, kicking, overhand throw, and underhand roll. Depending on the test, each skill has a set of three to five criteria, and each one is scored with a 0 or 1. The child performs each skill twice; therefore, each skill’s maximum score ranges from 6 to 10. The test began when the principal researcher instructed the child to prepare and start the test at their free will.

2.6. Reliability Estimates

The reliability of a test can be described as the consistency of the test results, e.g., the extent to which that same test would produce the same results under the same conditions when repeated several times [23]. The consistency of the 4-SFMT was evaluated using test–retest and inter-rater/intra-rater reliability. Therefore, the children performed the 4-SFMT on two separate occasions with a time frame of two weeks between the test and retest. The amount of time between two test administrations can have an impact on the test–retest reliability. Carryover effects due to memory or practice are more likely if the length interval is very short, whereas a longer interval can increase the chances of changes in these parameters. Each performance was videotaped to assess the inter-rater and intra-rater reliability of a skill scale, and the participants were fully informed that they would be videotaped throughout the test. To improve the quality of the video assessment, two cameras were used to record the videos (GoPro Hero 7 Black; GoPro, San Mateo, CA, USA). After the video recordings were processed, three raters with extensive experience in physical education (all three were educated kinesiologists) conducted the performance evaluation. The recordings were given to the raters to evaluate (score) the performance using guidelines after they were briefed on the procedures and study aims. The following instructions were given to each rater: (a) strictly follow the assessment scoring scale; (b) when scoring is complete, do not rewind the clip; and (c) try to complete the evaluations at the same time of day [24]. The order of the videos was randomly chosen.

2.7. Validity

Newly constructed tests should be assessed for various types of validity when scores are used for their intended purpose. Two types of validity were tested for the 4-SFMT: concurrent and construct validity. Concurrent validity involves correlating a newly constructed instrument concurrently with some criterion (reference test). The TGMD-2 was the criterion used for comparison with the 4-SFMT. The TGDM-2 is widely considered to be a robust process-oriented tool and has been extensively used to assess the motor skill performance of children up to 10 years of age. Moreover, previous research has shown that this test is reliable and valid for this age group [25,26]. Construct validity refers to the degree to which a test measures the construct it was designed to measure, and several factors can be used to demonstrate this type of validity. For the purposes of this study, sex differentiation and the correlation between the four tasks was assessed. As this test consists of four tasks representing four groups of basic motor skills (dimensions) [20], each task should have a significant positive correlation with the overall test score and a moderate correlation with each other.

2.8. Statistical Analysis

All the data were analyzed with SPSS 28.0 statistical software (SPSS, Chicago, IL, USA) and GraphPad Prism 9 (GraphPad Software, Inc., San Diego, CA, USA) and are presented as the mean and standard deviation after the normality of the data was confirmed with a Shapiro–Wilk test. To evaluate the level of agreement across the trials to establish evidence of inter-rater and intra-rater objectivity and test–retest reliability, intraclass correlation coefficients (ICC) with 95% confidence intervals (95%CI) were calculated from a two-way mixed-effects model for absolute agreement. The ICC is a value between 0 and 1, where values below 0.5 indicate poor reliability, values between 0.5 and 0.75 indicate moderate reliability, values between 0.75 and 0.9 indicate good reliability, and any value above 0.9 indicates excellent reliability [27]. A paired t-test was used to determine systematic bias/differences between two testing occasions (test/retest). Additionally, the standard error of measurement (SEm) and standard error of measurement expressed as a coefficient of variation (CV%) were calculated to determine within-individual variation, and a 95% confidence interval was also presented. Bartlett’s Test of Sphericity [28] and The Kaiser–Meyer–Olkin Measure (KMO) of Sampling Adequacy [29] was used to determine the suitability of the data for factor analysis, which was performed to find meaningful underlying dimensions. A value of 0.60 was the minimum standard to determine matrix factorability in the KMO. An exploratory factor analysis (EFA) was performed using principal component extraction and eigenvalues > 1. The scree test was used to decide how many factors to retain [30]. However, although considered the best and easy to administer, the scree test involves searches for sharp distinctions between the eigenvalues, and sometimes there may be more than one demarcation point. Moreover, the reliability of scree plot interpretations is found to be low [31]. Therefore, Parallel Analysis (PA) and Velicer’s Minimum Average Partial (MAP) test [32] were conducted as supplementary analyses. Both the PA and MAP were conducted using the O’Connor SPSS syntax [33]. As suggested by Comrey and Lee [34], the factor loadings were interpreted as excellent (>0.71), very good (>0.63 and <0.71), good (>0.55 and <0.63), fair (>0.45 and <0.55), and poor (>0.32 and <0.45). Values of 0.32 should be the minimum threshold used to identify significant factor loadings [35]. The Pearson product-moment correlation coefficient (r) associated with 95%CI and the coefficient of determination (R²) were calculated to examine the relationship between each item and the total score and performance time and to establish the relationship between the 4-SFMT and TGDM-2. The magnitude of the correlations was also determined using the modified scale by Hopkins [36]: 0.1, trivial; 0.1–0.3, small; 0.3–0.5, moderate; 0.5–0.7, large; 0.7–0.9, very large; 0.0.9, nearly perfect. An unpaired t-test was conducted to determine the differences in the 4-SFMT raw scores and the performance time between genders. All the effect sizes for the appropriate analyses were calculated using Cohen’s d [37], with values of <0.2, >0.2 and <0.6, >0.6 and <1.2, >1.2 and <2.0, and ≥2.0 considered as trivial, small, medium, large, and very large effects, respectively. The level of statistical significance for the analyses was set at p < 0.05.

3. Results

3.1. Inter-Rater and Intra-Rater Reliability

Intra- and inter-rater reliability values for the 4-SFMT score are presented in

Table 2. Intra- and inter-rater reliability for 4-SFMT score.

Reliability		ICC	95%CI
Inter-rater
Raters	ROLL	0.927	0.819 to 0.968
	PULL	0.953	0.909 to 0.976
	BALL	0.823	0.515 to 0.926
	CLIMB	0.778	0.433 to 0.904
	Total	0.914	0.395 to 0.975
Intra-rater
Rater 1	ROLL	0.89	0.678 to 0.963
	PULL	0.826	0.474 to 0.942
	BALL	0.742	0.203 to 0.914
	CLIMB	0.846	0.558 to 0.948
	Total	0.961	0.883 to 0.987
Rater 2	ROLL	0.916	0.75 to 0.972
	PULL	0.751	0.272 to 0.916
	BALL	0.714	0.147 to 0.904
	CLIMB	0.703	0.088 to 0.901
	Total	0.963	0.893 to 0.988
Rater 3	ROLL	0.932	0.8 to 0.977
	PULL	0.911	0.736 to 0.97
	BALL	0.732	0.176 to 0.911
	CLIMB	0.816	0.467 to 0.938
	Total	0.967	0.903 to 0.989

ICC = intraclass correlation coefficient; 95%CI = 95% confidence interval.

. Evidence for inter-rater objectivity was excellent (ICC > 0.9) for ROLL, PULL, and the overall score and was good (ICC > 0.75 and <0.9) for BALL and CLIMB. Evidence for intra-rater reliability for the skill score ranged from excellent for the overall score to moderate for BALL for all three raters.

3.2. Test–Retest Reliability

Table 3. Means and standard deviations of test and retest scores and measures of reliability.

	Test		Retest
Variable	Mean	SD	Mean	SD	ES	ICC	95%CI	SEm	95%CI	CV%
ROLL	2.91	1.39	2.58	1.53	0.223	0.877	0.645 to 0.958	0.637	0.438 to 1.164	5.94
PULL	2.8	0.86	2.82	1.04	0.035	0.889	0.665 to 0.963	0.401	0.276 to 0.733	21.15
BALL	2.68	0.89	2.73	0.75	0.097	0.711	0.109 to 0.904	0.626	0.433 to 1.148	14.28
CLIMB	4.34	0.74	4.18	0.76	0.214	0.85	0.569 to 0.949	0.402	0.277 to 0.734	11.2
Total score (a.u.)	12.22	3.41	12.34	3.12	0.036	0.973	0.920 to 0.991	0.724	0.498 to 1.323	7.2
Time (s)	76.08	27.07	74.54	25.79	0.058	0.995	0.984 to 0.998	2.652	1.824 to 4.842	3.34

ES = effect size; ICC = intraclass correlation coefficient; SEm = standard error of measurement; CV% = standard error of measurement expressed as coefficient of variation; 95%CI = 95% confidence interval; a.u. = arbitrary units.

shows the means and standard deviations of the test and retest scores, CV%, and the 95% confidence intervals for the ICCs and SEm. The ICCs between the test and retest scores ranged from 0.711 to 0.995. No significant differences were found between the test and retest scores, although small effects were detected for ROLL and CLIMB.

3.3. Determination of the Factorial Structure with EFA

The factorial structure of the 4-SFMT is presented in

Table 4. Factor loadings from EFA.

Variable	Factor 1
ROLL	0.8
PULL	0.46
BALL	0.81
CLIMB	0.51

. The significant Bartlett’s Test of Sphericity (χ² = 11.471, p = 0.075) and the KMO value of 0.61 indicates that the correlation matrix had significant correlations among some of its components and that the sample was adequate, respectively. By looking at the scree plot (not presented), one factor had an eigenvalue greater than one, accounting for 43.99% of the variance with the eigenvalue of 1.768. The PA revealed that the raw data eigenvalue from the actual data was greater than the eigenvalues of the 95th percentile of the random data distribution for one factor. Additionally, Velicer’s MAP test confirmed a one-factor solution. A chi-square goodness-of-fit test determined that the one-factor model fit the data well, χ²(2) = 0.18, p = 0.916. As shown in Table 4, ROLL and PULL had excellent loadings for Factor 1, while PULL and CLIMB had fair loadings.

3.4. Construct Validity

Table 5. Correlation analysis between 4-SFMT total test score and time and interitem correlation.

Item	Correlation with Total Score	95%CI	Inter Item Correlation
			ROLL	PULL	BALL	CLIMB	Time
ROLL	0.724 **	0.492 to 0.86	1	0.224	0.497 *	0.199	−0.509 **
PULL	0.533 **	0.213 to 0.749		1	0.191	0.016	−0.233
BALL	0.757 **	0.545 to 0.878			1	0.269	−0.467 **
CLIMB	0.505 **	0.177 to 0.732				1	−0.348
Time (s)	−0.576 **	−0.775 to −0.272					1

ICC = interclass correlation coefficient; 95%CI = 95% confidence interval; * p > 0.05; ** p > 0.01.

shows the results of the correlation between the 4-SFMT total score and interitem correlation matrix. As expected, all the item correlations with the total score were positive and significant. The highest correlation with the total score was with BALL (R² = 0.572, t = 6.122, very large). A significant negative correlation between the total score and performance time was observed (R² = 0.332, t = −3.727, moderate). Correlations between scores on individual items were significant only between ROLL and BALL (R² = 0.247, t = 3.03, small). Performance time significantly correlated with ROLL (R² = 0.259, t = −3.127, small) and BALL (R² = 0.218, t = −2.796, small).

Differences between boys and girls regarding the 4-SFMT total score and performance time are presented in

Table 6. Differences between boys and girls in 4-SFMT total score and performance time.

	Boys (n = 18)		Girls (n = 12)
Variable	Mean	SD	Mean	SD	p
Total score (a.u.)	12.15	2.62	12.22	2.91	0.943
Time (s)	65.81	24.77	71.97	21.78	0.49

p—probability value; a.u. = arbitrary units.

. There were no significant differences in both variables (total score (t = 0.072; d = 0.014: trivial) and performance time (t = 0.669; d = 0.14: trivial)).

3.5. Concurrent Validity

Correlations between the 4-SFMT total score and performance time and the TGMD-2 score are presented in Figure 2 and Figure 3. Figure 2 shows significant positive correlations between the 4-SFMT total score and TGMD-2 score (r = 0.824, R² = 0.679, t = 7.696 p < 0.001). The correlation between the 4-SFMT performance time and TGMD-2 score was significant and negative (r = −0.652, R² = 0.425, t = −4.574, p < 0.001), as shown in Figure 3.

4. Discussion

The main aim of this study was to create a new test to assess the motor skills of children of preschool and early school age and to determine its reliability and validity. The main findings were as follows: (a) all the ICC values that represent the level of agreement between the raters were statistically significant for all three raters, with two variables presenting excellent (ICC > 0.9) reliability and two variables having good reliability (ICC > 0.75 and <0.9); (b) no significant differences were found between the test and retest scores, although small effects were detected for two tasks, indicating the test’s high reliability; (c) through factor analysis, one motor factor F1 was isolated from the four tasks representing space covering, overcoming resistance, object control, and overcoming obstacles skills; (d) there was a moderate correlation between process- and product-oriented assessments of the 4-SFMT; (e) statistically significant correlations (from large to very large) were found between the process- and product-oriented assessments of the test 4-SFMT and test TGMD-2, which indicates the good concurrent validity of the newly constructed test; (f) there was no statistically significant difference in the quantitative and qualitative results of the newly constructed test between boys and girls.

The results confirmed that the criteria for evaluating certain motor skills with the 4-SFMT were well defined because all indicators of agreement between the experts had acceptable values (ICC from 0.778 to 0.953). These results are in line with other product-oriented assessment tests such as the Movement Assessment Battery for Children (MABC), Bruininks–Oseretsky Test of Motor Proficiency (BOT), and Körpercoordinationstest für Kinder (KTK), where values of inter-rater reliability were in the range from good (ICC > 0.75–0.9) to excellent (ICC > 0.9) in several studies [38,39,40]. Moreover, the excellent test–retest ICC values of 0.973 and 0.995 for score and time for the 4-SFMT are comparable with the Athletic Skills Track (AST; ICC = 0.891) [41] and Canadian Agility and Movement Skill Assessment (CAMSA; ICC = 0.91) [42]. Somewhat lower reliability values (test–retest scores) of the 4-SFMT could be observed only for one task, while the values of the remaining tasks as well as the overall process and product of the obtained results were excellent. The task with lower reliability values was the task used to assess object-control skills (BALL), whereby children had to roll a ball on a bench. This task was probably the most complex in terms of its structure and was less familiar to children than the other locomotor tasks in the test. It is known that object-control skills are acquired later and are generally more difficult than locomotor skills [1]. From the results of the intra-rater reliability (Table 2), it is also noticeable that all three experts showed the lowest values of agreement precisely in that task.

One of this study’s interesting findings is the factor analysis results. Namely, although the 4-SFMT included four items that assessed the slightly different skill domains of gross MC, the factor analysis isolated only one construct—the factor of general motor competence. Such results are in accordance with some previous findings [15,20] and confirm the hypothesis stated in the introduction regarding the absence of strict differentiation in MC in children of that age. However, further research on a larger sample of subjects and with a larger number of tests is necessary for such conclusions to receive relevant scientific confirmation. In other tests used to assess FMS, it was found that there was a different number of factors obtained through factor analysis. Regarding the MBAC test, there was a disagreement in the obtained structures, so different studies demonstrated a two-factor, three-factor, or even five-factor structure [43], for example. Furthermore, four factors were confirmed with the Polygon test [20]; however, due to a small number of studies on structural validity, these results should be taken with caution. In all variants of the TGDM test, a stable structure has been demonstrated where all studies confirmed a two-factor model (locomotion and object control) despite cultural differences [43]. Similar to the findings of the 4-SFMT, the KTK test also found to have a one-factor structure, with the factor being interpreted as the coordination of the body [44].

The correlation coefficients between the scores on individual items were relatively low and were significant only between ROLL and BALL (R² = 0.247, t = 3.03). However, when determining the validity, it is recommended that the correlations between the tasks should not be too high because this would mean that they measure the same ability to the same degree and therefore are redundant [22]. On the other hand, there was a significant connection between performance time and total score, which points to the conclusion that the quality of the performance of given motor skills significantly affects the speed of performance of these skills. This is in line with previous empirical evidence that indicates a moderate to very large association between process- and product-oriented assessments [45,46]).

One of the goals of this research was to determine the concurrent validity of the newly constructed test. For this purpose, a Pearson’s correlation coefficient was calculated to determine the correlation between the performance time and total score of the 4-SFMT on the one hand and the results of the TGMD on the other (Figure 2 and Figure 3). Both correlations were significant, with a higher coefficient recorded in the total score compared to the performance time (r = 0.824 versus r = −0.652). Such a result was expected since the total score was obtained via a qualitative assessment of motor skills, which is very close to the method of evaluating the TGMD, which belongs to the category of criterion-referenced tests. Ultimately, no significant differences were found in the analyzed variables between boys and girls (total score (t = 0.072, p = 0.943); performance time (t = 0.669 p = 0.49)). This finding contradicts the hypothesis of sex differentiation in motor performance. However, there are numerous studies that have been conducted to determine the differences in MC between genders in preschool children, where a certain inconsistency has been observed in the results obtained [3]. Nevertheless, in most studies, boys performed better than girls in tasks assessing object-control skills, and girls performed better than boys in balancing and locomotor tasks. These differences in motor skills are attributed to environmental rather than biological factors [47].

This research also has certain limitations. For more relevant scientific conclusions, the sample of participants should be expanded in terms of number and age. Small samples can be a limitation in validity studies because they may not be representative of the larger population, which can lead to inaccurate or incomplete findings. Although concurrent validity was determined with large to very large correlations with the TGMD-2, the evidence for construct validity is weak. Therefore, future studies should include much larger samples with a wider age range as a variable to assess the construct validity. Additionally, it would be desirable for future research to include the body mass index (BMI) variable in children, which is directly related to performance in motor competence tests [48].

5. Conclusions

In conclusion, the 4-SFMT is a new tool that can be used to assess the performance of motor skills of 5–6-year-old children. It is relatively easy to administer, whether as a process- or product-oriented test. It can be carried out quickly and it does not contain expensive props, which makes it ecologically valid. Overall, this study provides valuable insights into the assessment of motor skills in children and may have implications for future research in this area.