A School-Based Screening Tool for Adolescents With Low Motor Coordination Abilities

Abstract

This study sought to select the most relevant test items from the Bruininks-Oseretsky Test of Motor Proficiency, Second Edition (BOTMP-2) and from a selection of health-related fitness tests for identifying school teenagers with poor motor coordination. The 241 participants in this study (144 boys, 97 girls aged 13–14 years old) were tested on the short form of the BOTMP-2 and on the following additional fitness tests: (a) seated medicine ball test, (b) broad jump, (c) handgrip strength, (d) alternate hand ball wall toss, (e) 10 × 5-meter agility shuttle run, and (f) Chester step test. We performed a factor analysis of participant scores on these various tasks and BOTMP-2 test items to reduce them to the least number of meaningful and useful items. Four factors explained 45% of the data variance: gross motor skills and power (including broad jump, hand ball toss, shuttle run, and sit-ups tests); fine motor skills (including copying star, following the maze and paper folding); core strength and balance (including push-ups, hopping, and balance beam); and general body strength (including medicine ball throw and handgrip). We conclude that an efficient school-based battery of test items to screen 13-14 year old adolescents for fitness and coordination should assess these four factors and might especially rely upon the broad jump, copying a star shape, hopping handgrip strength, aerobic fitness, and wall ball toss.

Keywords

gross motor skills fine motor skills strength balance factor analysis

Introduction

Approximately 5-6% of school-aged children and adolescents in Europe experience significant motor control and coordination difficulties (Gillberg & Kadesjo, 2003). These children and adolescents are less likely to be involved in general play and organized sports not only during childhood but also later in life (Barnett, van Beurden, Morgan, Brooks, & Beard, 2009). Decreased levels of physical activity (PA) associated with poor motor coordination in these children and adolescents could compromise their overall health and well-being, given their poor musculoskeletal fitness (Cantell, Crawford, & Tish Doyle-Baker, 2008) and their higher risk of developing cardiovascular disease (Rivilis et al., 2011) and anxiety and depression (Cairney, Rigoli, & Piek, 2013). This worrisome profile highlights the need to identify and address poor health-related fitness and PA participation in this population.

Several school-based exercise interventions have been developed to engage children and adolescents with poor motor proficiency (Smits-Engelsman et al., 2018), relying on their diagnosis through a variety of standardized motor competence test batteries, such as the Movement Assessment Battery for Children—Second Edition (Henderson, Rose, & Henderson, 1992), the Test for Gross Motor Development, Second Edition (Issartel et al., 2017), or the Bruininks-Oseretsky Test of Motor Proficiency, Second Edition (BOTMP-2; Bruininks, 2005). These tests assess a range of motor proficiencies including mainly manual dexterity and fine motor control, body coordination, gross motor skills, aiming and catching, locomotor skills, and balance and object control; these tests categorize participants according to precise criteria for clinical and research purposes (Geuze, Schoemaker, & Smits-Engelsman, 2015). Some tests are process-orientated, while others are focused on the product (Logan, Barnett, Goodway, & Stodden, 2017). However, in school or sport settings in which broad categorizations are the goal, there is less need for precision, and a deviation from this specific criterion-based approach may be of greater benefit (Geuze et al., 2015). In their current form, these diagnostic tools are lengthy and labor intensive, and they require clear and closely followed instructions, compromising their efficiency for mass screening (Bruininks, 2005). In addition, some test items are highly prone to a ceiling effect (Brahler, Donahoe-Fillmore, Mrowzinski, Aebker, & Kreill, 2012), highlighting the need for a closer examination of each item for its relevance to large-scale school-based testing that would reduce testing time and increase assessment efficiency.

Health-related fitness test batteries are also frequently involved in school-based exercise interventions targeting children and adolescents with or without poor motor proficiency to categorize them at the start of the intervention and assess any benefits linked to exercise (Ortega et al., 2008; Vanhelst, Beghin, Fardy, Ulmer, & Czaplicki, 2016). It is interesting to note that many of the tests involved in these large batteries, assessing strength, power, speed, agility, muscular endurance, or cardiorespiratory fitness, also heavily rely on coordination. For example, Hands (2008) highlighted that the fitness components of jumping involve very specific elements of motor control, such as precise timing and positioning of the limbs during the different phases of these skills. However, some tests include more coordination elements than others do; for example, agility performance relies more on coordination than straight-line speed, and assessing cardiorespiratory fitness with the Chester step test (CST) is more demanding in terms of coordination than the Cooper (12-minute run test around a stadium) test. In the context of youth testing, fitness tests involving more coordination might be more interesting as they replicate the context in which they play and interact with others.

Several studies have attempted to categorize health-related fitness or motor competency tests into separate factors, either to assess the validity of these test items in the context of performance or to adapt testing to time constraints, such as those associated with large-scale school screening (Brown, 2019; Hassan, 2001; Issartel et al., 2017; Kambas & Aggeloussis, 2006; Psotta & Brom, 2016). However, past studies have considered either motor or health-related fitness test batteries, while there is currently no study, to our knowledge, that has examined both types collectively. In addition, many of the previously mentioned studies were performed on young children, while few have addressed adolescents in this context. It is crucial to examine this population as there is a well-established decrease in PA levels at this age, widening the gap in health-related fitness between adolescents with and without coordination issues (Schott, Alof, Hultsch, & Meermann, 2007). Therefore, the present study aimed to select the most relevant tests from among a selection of motor coordination and health-related fitness tests with the goal of assisting mass screening in schools to identify adolescents with poor motor coordination.

Method

Participants

Two hundred and forty-one adolescents (144 boys, 97 girls) aged 13–14 years (Year 9 in school) participated in this study. They were recruited from a mainstream school in the Oxfordshire area, classified in the second quintile of economic deprivation with a score of 13.03, based on the Office of National Statistic Indices of Multiple Deprivation 2010. Participants were mainly Caucasian, and 15.4% were overweight according to the World Health Organization cutoffs for body mass index (kg·m⁻²) for this age-group (de Onis et al., 2007). We obtained permission to collect data from each school’s head teacher, and parents or legal guardians returned a signed consent form to exclude their child from the study after details about the study procedures were sent to them. This opt-out recruitment method was approved by the University Research Ethics Committee at the time of the study.

Design Overview

Data collection took place in the sports hall of each school during physical education classes, in the form of a circuit of various stations overseen by physical education teachers. Tests within this fitness test battery included the BOTMP-2 short form, as well as a selection of health-related fitness tests. Participants were randomly divided into groups of six, each of which rotated between stations.

Tests

The short version of the BOTMP-2 is a popular motor assessment battery used for clinically identifying movement difficulties in children and young people aged 4-21 years (Bruininks, 2005). The BOTMP-2 is characterized by excellent interrater reliability (range: 0.88–0.92), good test–retest reliability (range: 0.62–0.73; Lucas et al., 2013), and moderate to good levels of agreement (validity) compared with similar tests (Fransen et al., 2014). The BOTMP-2 consists of eight subtests including a total of 14 items for the assessment of fine motor precision (drawing lines through paths-crooked, folding paper); fine motor integration (copying a star shape, copying a square shape); manual dexterity (transferring pennies); bilateral coordination (jumping in place same side synchronized, tapping feet and fingers same side synchronized); balance (walking forward in a line, standing on one leg on a balance beam with eyes open); running speed and agility (one-legged stationary hop); upper limb coordination (ball dropping, ball dribbling); and strength (push-ups, sit-ups). The setup, instructions, and scoring system of each of the items are described in detail in the BOTMP-2 manual (Bruininks, 2005).

Past literature documents the use of a variety of health-related fitness tests as part of large assessment batteries, covering mainly strength, power, speed, agility, balance, flexibility, muscular endurance, and cardiorespiratory fitness (Kambas & Venetsanou, 2014; Ortega et al., 2008; Vanhelst et al., 2016). We chose tests for this research based on their requirement for coordination as well as fitness. On these bases, as well as time constraints, we selected no measures of flexibility or speed, as we believed that those tests relied less on coordination than did other tests. We also excluded tests of muscular endurance and balance to avoid redundancy with measures within the BOTMP-2 short form.

The seated medicine ball throw assesses upper limb power and was selected for its ease of implementation and its coordination requirements (compared with the bench press for example). The medicine ball throw required participants to sit on the floor with legs fully extended, feet (∼60 cm) apart, and their backs against a wall. A 4-kg medicine ball was held with the back of the hands facing the center of the chest and the forearms parallel to the ground. Participants were instructed to throw the medicine ball (by pushing the hands away from the chest) vigorously as far straight forward as they could while keeping their back against the wall. We then used a measuring tape to acquire the distance thrown (from the wall to where the ball landed), and we recorded the best distance achieved out of three trials recorded. This task has shown very good test–retest reliability in children and adolescents in its past usage (intraclass correlation coefficient of .93 (Vanhelst et al., 2016).

The broad jump is among the most commonly used measures of lower limb power in children and adolescents; we chose it, rather than a vertical jump test, for its ease of implementation and cost-effectiveness. The broad jump is characterized by very good reliability (test–retest intraclass correlation coefficient of .91, standard error of measurement of 12.23, and coefficient of variation of 6.89%; Gillen, Miramonti, McKay, Leutzinger, & Cramer, 2018). Participants started in a standing position with feet together behind the start line on a jumping mat. They then jumped horizontally as far as possible. We measured the distance between the starting line and the heel of the foot that was most backward, and we kept the longest jump of two trials.

We assessed handgrip strength of the dominant hand (the one used for writing) using a handgrip dynamometer (Takei 5001, Tokyo, Japan). We chose this test because it is quick to administer and is a good predictor of total body strength (Wind, Takken, Helders, & Engelbert, 2010). In a standing position, we instructed participants to squeeze as hard as they could while simultaneously swaying their arm down in front of them, and we recorded the best of two trials. Past researchers have reported excellent test–retest reliability for this test (mean intertrial difference of 0.3 [SD = 2.5] and 0.0 [SD = 1.8]) for boys and girls, respectively; Ortega et al., 2008).

The alternate hand wall toss is a relatively new test of upper limb coordination (Du Toit, Krüger, Fowler, Govender, & Clark, 2010) which consisted of standing one meter away from a wall and tossing a tennis ball with one hand against the wall in an underarm manoeuver and then catching it with the opposite hand. The ball was then thrown back against the wall with the hand that caught it, and, then, it was caught again with the initial throwing hand. The test continued for 30 seconds, and we recorded the number of successful catches.

We assessed agility with the 10 × 5-meter shuttle run (Baquet, Berthoin, Gerbeaux, & Van Praagh, 2001). The number of 180° turns in the 10 × 5-meter shuttle requires more coordination than other agility tests. Participants started with one foot directly behind a line traced on the floor, and we instructed them to run and step on an opposite line placed 5 m away, turn and then run back to the starting line. This was repeated five times, and we recorded the duration (in seconds) required to run these 50 m. Prior researchers have reported good test–retest reliability for this test (r = .69; Baquet et al., 2001).

The CST is a submaximal multistaged fitness test that consisted of stepping on a 30-cm high step (The Step, USA) at a gradually increasing frequency (15–35 cycles per minute) set by a metronome for five stages of 2 minutes each (Buckley, Sim, Eston, Hession, & Fox, 2004). We chose this test because of its high reliance on coordination and for its minimal space requirements. One cycle is defined as stepping on and off the step with both feet. The test started with a brief introduction that familiarized the participants with the task, followed by a demonstration of the initial stepping rate. Throughout the test, we encouraged participants to step at the appropriate rate. We measured heart rate (HR, beats·min⁻¹) during the last few seconds of each stage, using a pulse oximeter and expressed results relative to each participant’s theoretical maximal HR (HR_max = 220 − age). We then used the five HR readings for each participant to predict maximal oxygen consumption (VO_2max), based on the extrapolation of a line of best fit, which passes through HR readings for each stage, up to a level equal to the participants’ estimated HR_max (Buckley et al., 2004).

Statistical Analyses

We conducted all statistical analyses with SPSS 23 for Windows (SPSS Inc., Chicago, IL, USA). We calculated participant means and standard deviations on all test scores. Subsequently, we ran a factor analysis to reduce the test items to the least number of meaningful and useful items. The extraction method chosen was principal axis factoring, and the rotation method was oblique oblimin with Kaiser normalization. An oblique rotation allows the selected factors to be correlated with one another. We used the Kaiser–Meyer–Olkin (KMO) statistic to test for sampling adequacy, with values less than 0.5 indicating that sampling was not adequate for factor analysis (Rosenblad, 2009). To determine the number of underlying factors in the data set, we ran an initial analysis to obtain eigenvalues for each factor. Eigenvalues represent the amount of variance explained by a factor. We included only factors with eigenvalues above Kaiser’s criterion of one (i.e., a substantial amount of variation; Rosenblad, 2009). We also used scree plots for that purpose; the point of strong inflection in a scree plot is regarded as a cutoff for the number of factors extracted (Rosenblad, 2009). We excluded missing cases list-wise. We conducted multiple analyses before the final analysis, with the aim of obtaining a simple structure, in which each variable loaded highly onto one factor only. Finally, we performed a reliability analysis using Cronbach’s alpha on each factor identified by the main analysis.

Results

Table 1 shows the descriptive data of scores participants obtained on all performed tests. The average total score on BOTMP-2 short form was 75.6 (SD = 4.4). We performed the factor analysis only on the raw test scores, and we excluded the following items because very little or no variability (ceiling effect) might have hindered the function of the correlation matrix (Rosenblad, 2009): jumping synch, tapping hand and foot synchronized and walking forward in a line We then ran a preliminary analysis, eliminating the VO_2max score as it did not correlate with any other variables.

Table 1.

Descriptive Data of Scores Obtained on all Performed Tests.

Test item	M	SD	Range
Height (cm)	161.7	9.2	101.0–187.0
Weight (kg)	50.7	9.6	32.9–84.9
BMI (kg·m⁻²) boys	19.0	3.1	14.2–31.6
BMI (z score) boys	0.18	1.36	−1.98–5.77
BMI (kg·m⁻²) girls	19.9	4.0	14.4–44.9
BMI (z score) girls	0.26	1.52	−1.83–9.86
Ball dropping and catching with two hands (number of successful attempts out of 5)	4.8	0.8	0–5
Ball dribbling with alternate hands (number of successful attempts out of 10)	9.4	1.7	1–10
Push-ups (number in 30 s)	22	7.6	4–50
Sit-ups (number in 30 s)	21	5.1	10–36
Balance beam (s, maximum 10 s)	8.2	2.9	1–10
Walking forward in a line (number of steps)	6.0	0.1	5–6
Penny transfer (number in 15 s)	16	2.2	7–20
Tapping coordinated (number of successful attempts out of 10)	9.9	0.6	5–10
Jumping synch (number of successful attempts out of 5)	5.0	0.0	5–5
Hopping (number in 15 s)	40	7	7–58
Paper folding (s)	6.1	1.5	0–7
Copying star shape (s)	4.8	0.7	0–5
Copying square shape (s)	5.0	0.2	4–5
Maze (s)	6.8	1.0	0–7
Broad jump (m)	1.6	0.3	0–3.1
Medicine ball throw (m)	3.5	0.8	1.3–6.4
Handgrip (kg)	25.6	6.3	9–48
Tennis wall toss (catches in 15 s)	21	8	0–40
10 × 5-m shuttle run (s)	19.9	2.2	14.7–29.4
VO_2max (ml·kg⁻¹·min⁻¹)	37.1	4.8	22.3–48.7

Note. Balance beam: standing on one leg on a balance beam, eyes open; tapping coordinated: tapping feet and fingers same side synchronized; jumping synch: jumping in place same side synchronized; hopping: one-legged stationary hopping, maze: drawing line through paths-crooked (s). BMI = body mass index; VO_2max = maximal oxygen consumption.

We conducted a first principal factor analysis on the remaining test scores (16 items) with oblique rotation (direct oblimin, Table 2). The KMO measure verified the sampling adequacy for the analysis (KMO = 0.67), which corresponded to Mediocre; and all KMO values of individual items were larger than the acceptable level of 0.5 (except for ball dribbling: 0.416). The resulting scree plot as well as the eigenvalues suggested the extraction of five factors. The results of this analysis led to the exclusion of two items: ball dropping and catching and ball dribbling for several reasons. First, both items failed to correlate highly with any of the other variables in the correlation matrix. Second, the item ball dropping and catching did not load on any of the factors after rotation, and the KMO value of the ball dribbling was below the acceptable level of 0.5. Finally, both items were at extreme risk for a ceiling effect.

Table 2.

Pattern Matrix of the Rotated Factors (Oblique Rotation) for the Initial and Second Analyses.

Test item	Factors initial analysis
Test item	1	2	3	4	5
10 × 5-m shuttle run	−.747
Broad jump	.563			.316
Sit-ups	.562
Ball dropping and catching
Maze		.714	.355
Paper folding	.403	−.608
Copying star		−.528
Copying square		−.425
Push-ups			.600
Hopping			.592
Balance beam			.534
Penny transfer			.356
Medicine ball				.832
Handgrip				.806
Ball dribbling					.655
Tennis wall toss	.369				.398
	Factors second analysis
		1	2	3	4
10 × 5-m shuttle run		−.778
Broad jump		.628
Sit-ups	.573
Tennis wall toss	.474
Maze		.798	.307
Paper folding		−.690
Copying star		−.445
Copying square		−.361
Push-ups			.579
Hopping			.558
Balance beam			.549
Penny transfer			.394
Medicine ball				.799
Handgrip				.789

Next, we conducted a second principal factor analysis on the 14 remaining items with oblique rotation (direct oblimin, Table 2). The KMO measure verified the sampling adequacy for the analysis (KMO = 0.7), which corresponded to Meddling and all KMO values of individual items were larger than 0.66 which is well above the acceptable level of 0.5. We ran an initial analysis to obtain eigenvalues for each factor in the data. Four factors had eigenvalues over Kaiser’s criterion of one and in combination explained 45% of the variation. The scree plot suggested the extraction of four factors or latent variables (Figure 1). The first factor comprised the shuttle run, broad jump, ball wall toss, and sit-ups tests and could therefore be labeled gross motor skills and power The second factor included the items copying star, following the maze and paper folding; hence, we named it fine motor skills The third latent variable was comprised of push-ups, hopping, and balance beam and was labeled core strength and balance and, finally, the fourth factor included the medicine ball throw and handgrip and so was named general body strength. Results from the reliability analysis are presented in Table 3.

Figure 1.

Scree plot of the second factor analysis. The inflection point suggested the extraction of four factors.

Table 3.

Results of the Reliability Analysis on the Factors Identified by the Second Analysis.

	Corrected item-total correlation	Cronbach’s alpha if item deleted
Factor 1 Cronbach’s alpha = .225
Shuttle run	−.653	.355
Broad jump	.336	.058
Sit-ups	.429	.090
Tennis wall toss	.348	.050
Factor 2 Cronbach’s alpha = .560
Maze	.521	.331
Paper folding	.422	.515
Copying star	.433	.444
Copying square	.266	.602
Factor 3 Cronbach’s alpha = .483
Maze	.101	.508
Push-ups	.387	.334
Hopping	.411	.307
Balance beam	.395	.400
Penny transfer	.226	.471
Factor 4 Cronbach’s alpha = .263
Medicine ball	.617	–
Handgrip	.617	–

Discussion

The main finding of this study shows that our selection of motor proficiency and fitness tests, when used with 13-14 year-old adolescents, can be grouped into a four-factor structure of (a) gross motor skills and power, (b) fine motor skills, (c) core strength and balance, and (d) general body strength. These categories could be particularly useful when trying to reduce motor proficiency test items for large-scale screening in schools. The following paragraph will discuss which test items would be most relevant in each of these four categories.

Before discussing the latent variables identified by our analysis, it is important to consider the excluded items. Three items of the BOTMP-2 short form were excluded before running the analysis for having very little to no variability. This finding is similar to reports by Brahler et al. (2012) who characterized multiple BOTMP-2 short form items as showing no variability and as being highly susceptible to a ceiling effect. The VO_2max scores obtained from the CST were also excluded for having no correlation with any other test items for two possible reasons. First, cardiovascular fitness might be a stand-alone ability that shares no common features with other motor skills. However, this is very unlikely in light of vast literature describing a clear relationship between coordination, motor proficiency, and fitness (Barnett et al., 2009; Haga, 2009; Rivilis et al., 2011). A second and more plausible explanation is that the CST did not yield reliableVO_2max measures, possibly because the short duration for obtaining these measurements (10-second rest between stages) together with the slow recording of HR data by the pulse oximeters meant that the student’s HR would have already changed by the time it was measured. Also, this technology is not suitable in winter months, as pulse measurement from the fingers could not be obtained on occasions when participants’ fingers were too cold to register it. With these challenges in mind, we suggest that CST is not a suitable test of cardiovascular fitness for mass screening. However, because of a well-established link between CST and life-threatening adult conditions like cardiovascular disease (Rivilis et al., 2011) and depression (Cairney et al., 2013), CST should be tested in school for these purposes. It should be replaced with a test that does not rely on HR measurement, such as the 20-m shuttle run.

We labeled the first latent factor revealed by our analysis gross motor skills and power, and it included the shuttle run test, broad jump, sit-ups, and ball wall toss. This latent factor is similar to the group of factors identified by Kambas and Venetsanou (2014) in their examination of a motor screening tool (Democritos Movement Screening Tool for preschool children) in preschool children that these authors called gross motor control, as it included a variety of jumping and running tests, as well as one upper body element (running and carrying and placing a ball in a box). Furthermore, an analysis of the BOTMP-2 in preschool and primary school children also resulted in the classification of various tests into a gross motor skill ability, with tasks that included the broad jump and a speed and agility test, similar to tasks in our study (Kambas & Aggeloussis, 2006). The similarity between our findings and those of previous studies highlights that no matter the age-group considered, gross motor skills are essential in the screening of children and adolescents for fitness or coordination purposes. Indeed, these skills form the basis of the games undertaken in playgrounds, and poor competence in these skills could be a main cause of children’s low PA levels (Barnett et al., 2009; Cairney et al., 2005). Within this first gross motor skill factor, the items that loaded the highest in the present study were the agility shuttle run and broad jump, with correlation coefficients of −.0778 and .628, respectively. In a test battery for fitness or coordination screenings in schools, these tests should be included. The broad jump test is part of multiple motor proficiency batteries like the ALPHA health-related fitness battery for children and the European test of Physical Fitness (EUROFIT; Baquet et al., 2001; Ortega et al., 2008). When the question of using both or only one of these tests arises, different arguments could be put forward. In favor of using both tests, Salaj and Markovic (2011) showed that jumps, including the broad jump, and quick changes in direction, such as the agility shuttle run, are two distinct abilities and should be tested separately. However, while both tests had acceptable reliability in our study, the Cronbach’s alpha when the shuttle run (.355) was deleted was greater than the overall factor reliability (.225). Thus, removing this item would improve the overall reliability of the factor. The shuttle run might not be the most meaningful for this factor. Consequently, we suggest using the broad jump as a more meaningful test for this factor. The sit-up test was also among the variables selected in this first factor. A very recent study (Brown, 2019) on the structural validity of the BOTMP-2 short form identified the sit-up test as one of the five crucial elements to keep in a revised version of this test, further highlighting its importance in testing fitness/coordination.

We labeled the second latent factor identified by our analysis fine motor skills, and it was comprised of the maze, paper folding, copying star, and copying square items. Fine motor skills are an essential component of coordination, and all batteries of tests for motor proficiency include this category (Bruininks, 2005; Henderson et al., 1992). Furthermore, similar to our findings, a recent study based on the factorial structure of the Movement Assessment Battery for Children—Second Edition identified a group of similar items called manual dexterity (Psotta & Brom, 2016). The items included within this category corresponded to two subcategories of the BOTMP-2, namely fine motor precision (maze, folding paper) and fine motor integration (copying a star shape, copying a square shape; Bruininks, 2005), with the items related to fine motor precision loading highly on factor two in our analysis. This suggests that motor integration might be less important in this context. However, in practical terms, administering these two items was tricky as participants (especially boys) tended to lost patience with them and made an insufficient effort to perform them correctly. In contrast, the copying items were quicker to perform. We recommend the copying items for school screening as their sensitivity is greater in children and adolescents. Indeed, adolescents’ shorter attention spans are exacerbated by indoor versus outdoor activities (Rogerson, Gladwell, Gallagher, & Barton, 2016). Further studies are needed for determining how this variable affects the sensitivity of tests performed in the context of indoor school screening. Our reliability analysis indicated that the copying star test would be a better choice, compared with the copying square test, because of its greater reliability (.433 vs. .266) and its Cronbach’s alpha if item deleted (.444) below the overall reliability of the factor (.530), unlike the “copying square” test (.602).

The third group of items revealed by our analysis included push-ups, hopping, balance beam, penny transfer, and maze items. We labeled it core strength and balance. Several studies relying on reduction analysis included balance as a latent factor, whether they used the BOTMP-2 or other tests, such as the M-ABC (Hassan, 2001; Psotta & Brom, 2016). Balance was also identified as an important discriminating factor between children with poor coordination and children with good coordination (Hands, 2008). It is interesting to note that this group of test items is the most heterogeneous of the four groups identified in the present study, with items such as penny transfers or maze included, even though they did not reach the threshold of 0.5 that one would instinctively characterize as necessary for their inclusion among tests of fine motor skills. This observation, together with the fact the mazes item loaded on two factors, questions the separation between fine and gross motor skills, highlighted by Hassan (2001). Indeed, Hassan suggested that there is some moderate degree of factorial balance overlap between components of motor proficiency assessments and that gross and fine motor skills should be considered on a continuum rather than as separate skills (Hassan, 2001). While push-ups loaded highly on this factor, when reflecting on the practical aspects of this test, we noticed that the technique required for performing it was difficult, particularly for participants who had not done push-ups before the screening. As this observation affected the suitability of this test in the context of school screenings, we suggest that efficiency-minded examiners use another item, such as hopping, that also loaded highly on this factor. In favor of this choice, hopping was one of the five test items of the BOTMP-2 short form retained after a structural validity analysis (Brown, 2019). Our reliability analysis showed acceptable values for this test (reliability of .411 and Cronbach’s alpha if item deleted (.307) lower than the overall reliability of the factor (.483)).

Finally, the last latent factor from our analysis included handgrip strength and medicine ball throw. Although these tests are performed with the upper limb, they are also associated with general coordination and strength (Luz et al., 2018; Wind et al., 2010). Hence, we named it general strength As both tests obtained a high loading score, the choice of only one test for this category might rely on the practical aspects associated with their use in mass screening. While the medicine ball test might be the cheapest option, the correct technique for it was quite difficult to judge. This test requires participants to throw the medicine ball vigorously as far straight forward as they could while maintaining their back against the wall, but there was no objective way to determine if the trajectory of the ball was horizontal and if participants’ back remained against the wall. For this reason, we recommend using the handgrip strength test in school screening, even though it requires more equipment. This is a reliable test (.617 in our study), and it is objective and easy to administer. It has also shown good predictive ability for total body strength (Wind et al., 2010) and bone mineral density (Chan et al., 2008).

Although the tennis ball wall toss test (Du Toit et al., 2010) loaded on the first factor, suggesting that this factor may represent a gross motor skill, we believe it is important to include this test as a separate item. It is the only test that assessed eye–hand coordination, highlighted as a fundamental aspect of motor proficiency in the BOTMP-2 (Kambas & Aggeloussis, 2006). In addition, Psotta and Brom (2016) noted that aiming and catching was one of the three categories for the factorial structure of a test battery for motor proficiency. As this task has an obvious relationship to a great many sporting activities involving catching and throwing, it may be particularly predictive of later engagement in sports and PA generally.

The main limitations of this study were that our participants came from only one school and may not be representative of the overall British adolescent population. Indeed, factors such as academic performance, geographical location, socioeconomic status, and existing sports opportunities and success were restricted in this ample. Further studies with more diverse populations are needed. In addition, we did not measure the maturational status of our participants, and it is likely that 13-14 year-old boys and girls may be at different stages of maturation. Finally, we could not get information about our participants’ PA levels or sport participation, and data regarding these variables could help interpret our results.

Conclusion

In conclusion, our findings suggest that a relevant battery of tests to screen for fitness and coordination in school settings should assess four main factors: (a) gross motor skills and power, (b) fine motor skills, (c) core strength and balance, and (d) general body strength. Tests of choice within these areas should be the broad jump, copying a star shape, hopping, and handgrip strength. Future studies should assess the feasibility of large-scale school screening using these tests and should evaluate their association to engagement in sports and PA at various subsequent ages during development.

Footnotes

Article Notes

References

Baquet

Berthoin

Gerbeaux

Van Praagh

(2001) High-intensity aerobic training during a 10 week one-hour physical education cycle: Effects on physical fitness of adolescents aged 11 to 16. International Journal of Sports Medicine 22(4): 295–300. doi:10.1055/s-2001-14343.

Barnett

L. M.

van Beurden

Morgan

P. J.

Brooks

L. O.

Beard

J. R.

(2009) Childhood motor skill proficiency as a predictor of adolescent physical activity. The Journal of Adolescent Health: Official Publication of the Society for Adolescent Medicine 44(3): 252–259. doi:10.1016/j.jadohealth.2008.07.004.

Brahler

C. J.

Donahoe-Fillmore

Mrowzinski

Aebker

Kreill

(2012) Numerous test items in the complete and short forms of the BOT-2 do not contribute substantially to motor performance assessments in typically developing children six to ten years old. Journal of Occupational Therapy, Schools, & Early Intervention 5(1): 73–84. doi:10.1080/19411243.2012.674746.

Brown

(2019) Structural validity of the Bruininks-Oseretsky test of motor proficiency – Second edition brief form (BOT-2-BF). Research in Developmental Disabilities 85: 92–103. doi:10.1016/j.ridd.2018.11.010.

Bruininks, R. H. (2005). Bruininks-Oseretsky test of motor proficiency, 2nd ed. Minneapolis, Minnesota: Pearson Assessments.

Buckley

J. P.

Sim

Eston

R. G.

Hession

Fox

(2004) Reliability and validity of measures taken during the Chester step test to predict aerobic power and to prescribe aerobic exercise. British Journal of Sports Medicine 38(2): 197–205.

Cairney

Hay

J. A.

Faught

B. E.

Wade

T. J.

Corna

Flouris

(2005) Developmental coordination disorder, generalized self-efficacy toward physical activity, and participation in organized and free play activities. The Journal of Pediatrics 147(4): 515–520. doi:10.1016/j.jpeds.2005.05.013.

Cairney

Rigoli

Piek

(2013) Developmental coordination disorder and internalizing problems in children: The environmental stress hypothesis elaborated. Developmental Review 33(3): 224–238. doi:10.1016/j.dr.2013.07.002.

Cantell

Crawford

S. G.

Tish Doyle-Baker

P. K.

(2008) Physical fitness and health indices in children, adolescents and adults with high or low motor competence. Human Movement Science 27(2): 344–362. doi:10.1016/j.humov.2008.02.007.

10.

Chan

D. C. C.

Lee

W. T. K.

D. H. S.

Leung

J. C. S.

Kwok

A. W. L.

Leung

P. C.

(2008) Relationship between grip strength and bone mineral density in healthy Hong Kong adolescents. Osteoporosis International: A Journal Established as Result of Cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 19(10): 1485–1495. doi:10.1007/s00198-008-0595-1.

11.

de Onis

Onyango

A. W.

Borghi

Siyam

Nishida

Siekmann

(2007) Development of a WHO growth reference for school-aged children and adolescents. Bulletin of the World Health Organization 85(9): 660–667.

12.

Du Toit

Krüger

Fowler

Govender

Clark

(2010) Influence of sports vision techniques on adult male rugby players. African Journal for Physical, Health Education, Recreation and Dance 16(3): 487–494. doi:10.4314/ajpherd.v16i3.60931.

13.

Fransen

D’Hondt

Bourgois

Vaeyens

Philippaerts

R. M.

Lenoir

(2014) Motor competence assessment in children: Convergent and discriminant validity between the BOT-2 short form and KTK testing batteries. Research in Developmental Disabilities 35(6): 1375–1383. doi:10.1016/j.ridd.2014.03.011.

14.

Geuze

R. H.

Schoemaker

M. M.

Smits-Engelsman

B. C. M.

(2015) Clinical and research criteria for developmental coordination disorder-should they be one and the same? Current Developmental Disorders Reports 2(2): 127–130. doi:10.1007/s40474-015-0043-9.

15.

Gillberg

Kadesjo

(2003) Why bother about clumsiness? The implications of having developmental coordination disorder (DCD). Neural Plasticity 10(1–2): 59–68. doi:10.1155/NP.2003.59.

16.

Gillen

Z. M.

Miramonti

A. A.

McKay

B. D.

Leutzinger

T. J.

Cramer

J. T.

(2018) Test-retest reliability and concurrent validity of athletic performance combine tests in 6-15-year-old male athletes. Journal of Strength and Conditioning Research 32(10): 2783–2794. doi:10.1519/JSC.0000000000002498.

17.

Haga

(2009) Physical fitness in children with high motor competence is different from that in children with low motor competence. Physical Therapy 89(10): 1089–1097. doi:10.2522/ptj.20090052.

18.

Hands

(2008) Changes in motor skill and fitness measures among children with high and low motor competence: A five-year longitudinal study. Journal of Science and Medicine in Sport 11(2): 155–162. doi:10.1016/j.jsams.2007.02.012.

19.

Hassan

M. M.

(2001) Validity and reliability for the Bruininks-Oseretsky Test of Motor Proficiency-Short Form as applied in the United Arab Emirates culture. Perceptual and Motor Skills 92(1): 157–166. doi:10.2466/pms.2001.92.1.157.

20.

Henderson

Rose

Henderson

(1992) Reaction time and movement time in children with a developmental coordination disorder. Journal of Child Psychology and Psychiatry, and Allied Disciplines 33(5): 895–905.

21.

Issartel

McGrane

Fletcher

O’Brien

Powell

Belton

(2017) A cross-validation study of the TGMD-2: The case of an adolescent population. Journal of Science and Medicine in Sport 20(5): 475–479. doi:10.1016/j.jsams.2016.09.013.

22.

Kambas

Aggeloussis

(2006) Construct validity of the Bruininks-Oseretsky test of motor proficiency-short form for a sample of Greek preschool and primary school children. Perceptual and Motor Skills 102: 65–72. doi:10.2466/pms.102.1.65-75.

23.

Kambas

Venetsanou

(2014) The Democritos Movement Screening Tool for preschool children (DEMOST-PRE©): Development and factorial validity. Research in Developmental Disabilities 35(7): 1528–1533. doi:10.1016/j.ridd.2014.03.046.

24.

Logan

S. W.

Barnett

L. M.

Goodway

J. D.

Stodden

D. F.

(2017) Comparison of performance on process- and product-oriented assessments of fundamental motor skills across childhood. Journal of Sports Sciences 35(7): 634–641. doi:10.1080/02640414.2016.1183803.

25.

Lucas

B. R.

Latimer

Doney

Ferreira

M. L.

Adams

Hawkes

Elliott

E. J.

(2013) The Bruininks-Oseretsky Test of Motor Proficiency-Short Form is reliable in children living in remote Australian Aboriginal communities. BMC Pediatrics 13: 135 . doi:10.1186/1471-2431-13-135.

26.

Luz

L. G. O.

Coelho-E-Silva

M. J.

Duarte

J. P.

Valente-Dos-Santos

Machado-Rodrigues

Seabra

Malina

R. M.

(2018) Multivariate relationships among morphology, fitness and motor coordination in prepubertal girls. Journal of Sports Science & Medicine 17(2): 197–204.

27.

Ortega

F. B.

Artero

E. G.

Ruiz

J. R.

Vicente-Rodriguez

Bergman

Hagstromer

Castillo

M. J.

(2008) Reliability of health-related physical fitness tests in European adolescents. The HELENA Study. International Journal of Obesity (2005) 32(Suppl 5): S49–S57. doi:10.1038/ijo.2008.183.

28.

Psotta

Brom

(2016) Factorial structure of the Movement Assessment Battery for Children Test-Second Edition in preschool children. Perceptual and Motor Skills 123(3): 702–716. doi:10.1177/0031512516666072.

29.

Rivilis

Hay

Cairney

Klentrou

Liu

Faught

B. E.

(2011) Physical activity and fitness in children with developmental coordination disorder: A systematic review. Research in Developmental Disabilities 32(3): 894–910. doi:10.1016/j.ridd.2011.01.017.

30.

Rogerson

Gladwell

V. F.

Gallagher

D. J.

Barton

J. L.

(2016) Influences of green outdoors versus indoors environmental settings on psychological and social outcomes of controlled exercise. International Journal of Environmental Research and Public Health 13(4): 363 . doi:10.3390/ijerph13040363.

31.

Rosenblad

(2009) Applied multivariate statistics for the social sciences, fifth edition by James P. Stevens. International Statistical Review 77(3): 476 doi:10.1111/j.1751-5823.2009.00095_13.x.

32.

Salaj

Markovic

(2011) Specificity of jumping, sprinting, and quick change-of-direction motor abilities. Journal of Strength and Conditioning Research 25(5): 1249–1255. doi:10.1519/JSC.0b013e3181da77df.

33.

Schott

Alof

Hultsch

Meermann

(2007) Physical fitness in children with developmental coordination disorder. Research Quarterly for Exercise and Sport 78(5): 438–450. doi:10.1080/02701367.2007.10599444.

34.

Smits-Engelsman

Vinçon

Blank

Quadrado

V. H.

Polatajko

Wilson

P. H.

(2018) Evaluating the evidence for motor-based interventions in developmental coordination disorder: A systematic review and meta-analysis. Research in Developmental Disabilities 74: 72–102. doi:10.1016/j.ridd.2018.01.002.

35.

Vanhelst

Beghin

Fardy

P. S.

Ulmer

Czaplicki

(2016) Reliability of health-related physical fitness tests in adolescents: The MOVE program. Clinical Physiology and Functional Imaging 36(2): 106–111. doi:10.1111/cpf.12202.

36.

Wind

A. E.

Takken

Helders

P. J. M.

Engelbert

R. H. H.

(2010) Is grip strength a predictor for total muscle strength in healthy children, adolescents, and young adults? European Journal of Pediatrics 169(3): 281–287. doi:10.1007/s00431-009-1010-4.