VALIDITY AND RELIABILITY OF 2 UPPER-BODY STRENGTH TESTS FOR PRESCHOOL CHILDREN CARLOS AYA´N PE´REZ,1 JOSE´ M. CANCELA,1 IRI´A SENRA,2
AND
EVA QUIREZA2
1
Department of Special Didactics; 2Healthy-Fit Research Group, Faculty of Education and Sport Science, University of Vigo, Pontevedra, Spain ABSTRACT
INTRODUCTION
Aya´n Pe´rez, C, Cancela, JM, Senra, I, and Quireza, E. Validity and reliability of 2 upper-body strength tests for preschool children. J Strength Cond Res 28(11): 3224–3233, 2014— The “bent knee push-up” and the “timed dipping” are 2 upperbody strength assessment tests whose simplicity and ease of use make them suitable to be applied to preschool children. Nevertheless, their psychometric properties have not been properly studied for this age group. This study aimed to fill this gap. To accomplish this, both tests were administered 3 times (familiarization, test, and retest) to a group of 120 children (mean age, 48.60 6 9.94 months; body mass index, 17.05 6 2.04 kg$m22; 50% girls), with a week interval between each assessment. The obtained results in each test were compared with those obtained in the “handheld dynamometry” (the criterion measure) with the aim of identifying their construct validity. The reliability of the tests was analyzed by comparing the obtained results during the second (test) and the third (retest) time that each 1 was carried out. Intraclass correlation coefficients (ICCs) were used to examine reliability, and Pearson’s correlation coefficients were used to examine validity. The bent knee push-up showed a moderate reliability for 3-year-old children (ICC, 0.690; 95% confidence interval [95% CI], 0.379–0.845) and high for those aged 4 years (ICC, 0.848; 95% CI, 0.715–0.919) and 5 (ICC, 0.702; 95% CI, 0.441–0.841). Similar results were obtained for the timed dipping, in children aged 3 years (ICC, 0.522; 95% CI, 0.422– 0.761), 4 years (ICC, 0.766; 95% CI, 0.560–0.875), and 5 years (ICC, 0.828; 95% CI, 0.677–0.908). Both tests showed a good construct validity, with the exception of the timed dipping in 3-year-old children (r = 0.200; p = 0.256). The results in this study suggest that the bent knee push-up and the timed dipping show psychometric properties that make them suitable to be used in research carried out with preschoolers.
KEY WORDS fitness, muscular, kindergarten
Address correspondence to Carlos Aya´n Pe´rez, [email protected]. 28(11)/3224–3233 Journal of Strength and Conditioning Research Ó 2014 National Strength and Conditioning Association
3224
the
H
ealth-related physical fitness (HRPF) assessment can help to identify children at the risk for major public health diseases and to establish dose-response relationships between activity and health outcomes, and intervention effectiveness (26). Given the emerging physical inactivity detected among preschool-aged children and the future health problems that may be developed as a consequence, HRPF assessment seems to be especially important in early childhood (23). When measuring HRPF in preschool children, the identification of the muscular strength level seems to be a matter of great interest, although little research has dealt with this issue (13). One of the possible reasons for this could be the lack of clinically viable and practical-to-use performance-based tests for the strength-level assessment in children younger than 6 years (24). These kind of tests are defined by a number of characteristics, among which we must highlight the lack of complicated equipment requirements, short administration time, easiness, simple employment and interpretation scoring system, and the presence of excellent psychometric properties, determined basically by the reliability the test shows when it is used with the target population (28). A test is reliable when it demonstrates that test scores are stable over time (test-retest reliability). Validity can be demonstrated by showing that the measure correlates significantly with another test that measures the same constructs (criterion validity). Although in the scientific literature, we can find muscular strength assessment tests that fulfill these characteristics, those are basically focused on the lower-body performance (3). Thus, we can claim that there is an obvious lack of assessment tests for the upper-body muscular strength (9). Under these circumstances, it seems necessary to carry out some research to analyze the reliability and validity of tests that, being easy to carry out and without complicated equipment requirements, allow us to know the upper-body strength level in preschool children. A test that seems to fulfill these characteristics is the “bent knee push-up” included in the “Bruininks-Oseretsky Test of Motor Proficiency” battery under the name of “modified push-up test” (6). Nevertheless, and although the battery is widely used with preschool children, the analysis of its validity and reliability has not been carried out with samples
TM
Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the
TM
Journal of Strength and Conditioning Research
| www.nsca.com
TABLE 1. Descriptive statistics for the sample.* 3 year old, Mean 6 SD 4 year old, Mean 6 SD 5 year old, Mean 6 SD (median) (median) (median) Height (m) Weight (kg) BMI (kg$m22) Bent knee push-up_test (repetition) Bent knee push-up_retest (repetition) Timed dipping_test (s) Timed dipping_retest (s) Handheld dynamometry_test (kg)† Handheld dynamometry_retest (kg)z Handheld dynamometry_ test (kg)§ Handheld dynamometry_ retest (kg)k
1.01 17.76 17.48 6.68
6 6 6 6
0.43 2.11 2.12 4.50
(1.01) (18.40) (17.26) (4.99)
5.86 6 5.08 (4.12)
1.08 19.48 16.78 11.24
6 6 6 6
0.52 2.95 1.79 8.49
(1.08) 1.15 6 0.52 (19.00) 22.31 6 3.78 (16.45) 16.89 6 2.22 (8.96) 20.33 6 18.15
(115.50) (22.00) (16.70) (14.85)
ANOVA F
Sig
77.410 22.469 1.285 12.949
0.001 0.001 0.281 0.001
12.18 6 10.45 (10.22) 25.50 6 22.83 (15.83) 16.706 0.001
2.37 6 2.21 (2.00) 3.62 6 2.09 (4.00) 4.45 6 1.70 (4.00)
4.76 6 3.09 (5.00) 5.22 6 3.74 (5.00) 5.91 6 2.07 (6.00)
9.81 6 5.03 (9.00) 7.71 6 3.64 (7.50) 7.83 6 2.07 (8.00)
41.052 0.001 14.964 0.001 29.652 0.001
4.44 6 1.57 (4.50)
6.29 6 1.65 (6.00)
7.69 6 1.86 (7.50)
26.696 0.001
4.46 6 1.07 (5.00)
5.70 6 1.52 (5.50)
7.15 6 2.04 (7.00)
34.604 0.001
4.71 6 1.52 (4.75)
6.18 6 1.58 (6.00)
7.71 6 2.22 (8.00)
25.747 0.001
*ANOVA = analysis of variance; BMI = body mass index; Sig = p , 0.05. †Performed before the bent knee push-up (test). zPerformed before the bent knee push-up (retest). §Performed before the timed dipping (test). kPerformed before the timed dipping (retest).
specifically composed of preschool children (10). Thus, it seems necessary to establish the psychometric characteristics of the bent knee push-up test when it is performed by children younger than 6 years. Similarly, another test that initially seems to fulfill the previously mentioned characteristics is the “timed dipping” (20). Nevertheless, although its psychometric properties have been contrasted when the test was administered to a group of Japanese preschoolers (16), they should be further analyzed for 2 reasons. First, there are no data on the timed dipping criterion validity, which offers greater objectivity because it involves correlating test scores with another established test that also measures the same fitness component. Second, there is a need to test the psychometric properties of the timed dipping when it is administered to a different ethnic group, as it is a variable that may have some influence over them (17). Given this situation, the aim of this investigation was to study the reliability and construct validity of the bent knee push-up and timed dipping tests when applied to a group of European preschoolers.
METHODS Experimental Approach to the Problem
The purpose of this study was to establish reliability and construct validity evidence for the bent knee push-up and
timed dipping tests in a sample composed of children older than 3 years and younger than 6 years. There are no specific independent and dependent variables for the designed study. Therefore, to analyze the reliability of the strength tests targeted in this study, we followed the protocol that was established by other authors who have analyzed the psychometric properties of different performance-based fitness measures with preschool children (25). Thus, the tests were carried out in 3 stages, with 1-week interval. The first stage was used for the introduction and explanation of the tests with the aim of getting the right performance of the children and avoiding a possible learning effect. The obtained results in each test during this first attempt were not taken into account for further analysis. The second stage was used as a test phase and the third stage as a retest phase. The comparison of the obtained scores in the second and in the third stage in each performance allowed us to analyze the reliability of the upper-body muscular strength tests. To be able to compare the construct validity of both the bent knee push-up and the timed dipping tests, we decided to compare the obtained results in each of them with those obtained after the “handheld dynamometry” test performance. This test was chosen as criterion measure because it can be considered as a gold standard when it comes to assess the upper-body strength (29), and it has proved to be VOLUME 28 | NUMBER 11 | NOVEMBER 2014 |
3225
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
Upper-Body Strength Preschool Children valid and reliable when it is employed with preschool children (15). Having into account that this investigation design included 2 valid times of assessment and data collection for each test (test and retest phases) and that the handheld dynamometry was always administered before the performance of the bent knee push-up and the timed dipping, the construct validity of both tests could be analyzed twice throughout the study with the aim of showing more solid results. Thus, the bent knee push-up (test) score was compared with the score obtained in the handheld dynamometry, (a) which was carried out just before it. In the same way, the score obtained in the third performance of the bent knee push-up (retest) was compared with the results obtained in the handheld dynamometry (b) carried out during the same assessment session. Also, the obtained score after the second performance of the timed dipping (test) was compared with the one obtained in the handheld dynamometry (c) carried out before the first. Finally, the obtained scores after the third performance of the timed dipping (retest) and the previous performance to this attempt of the handheld dynamometry (d) were also compared. A priori power calculation for the validity study indicated that a sample size of 120 would be necessary, given the effect size of 0.4, the power of 0.95, and an expected proportion loss 0.5. Power calculation is not applicable for a reliability study. Subjects Figure 1. Analysis of the responders and nonresponders to the “bent knee push-up” and “timed dipping” tests.
3226
the
The research was conducted on a sample of 120 Caucasian healthy Spanish urban children recruited from 2 kindergartens in the north of Spain. Inclusion
TM
Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the
TM
Journal of Strength and Conditioning Research
| www.nsca.com
TABLE 2. Upper-body strength tests reliability.* Reliability statistics 3 year olds
Bent knee push-up_test-retest (repetition) Timed dipping_test-retest (s) Handheld dynamometry_test-retest (kg)†z Handheld dynamometry_test-retest (kg)§k
4 year olds
5 year olds
ICC
95% CI
ICC
95% CI
ICC
95% CI
0.690 0.522 0.891 0.821
0.379–0.845 0.422–0.761 0.825–0.938 0.698–0.903
0.848 0.766 0.924 0.843
0.715–0.919 0.560–0.875 0.883–0.955 0.747–0.909
0.702 0.828 0.876 0.871
0.441–0.841 0.677–0.908 0.809–0.926 0.758–0.931
*ICC = intraclass correlation coefficient; 95% CI = 95% confidence interval. †Performed before the bent knee push-up (test). zPerformed before the bent knee push-up (retest). §Performed before the timed dipping (test). kPerformed before the timed dipping (retest).
criteria for participation were being enrolled in the first and second levels of the first stage of the Spanish Education System and having no pathologies that prevented the performance of the muscular strength tests proposed for the study. At the end of the research, the data were collected from 60 boys (mean age, 49.20 6 9.80 months; mean body mass index (BMI), 17.57 6 2.06 kg$m22) and 60 girls (mean age, 48.00 6 10.12 months; mean BMI, 16.50 6 1.93 kg$m22). Written informed consent was obtained from school principals and parents or guardians of all the children who took part in the study, before participating in the research protocol. The study was conducted in accordance with the Declaration of Helsinki, and all the study procedures were approved by our institutional ethics committee. Procedure
The research took place during the second term of the academic year 2012-2013, being all the tests carried out during each of the 2 weekly sessions of psychomotricity in which children were taking part and that are part of the preschool academic curriculum in Spain. Such sessions took place 3 hours after the children started their school day and were carried out with a 3-day interval between them. The children not allowed to drink or eat any kind of food during the 2 hours before the assessment sessions, which were carried out in groups of 20 children in 5-week blocks. One kindergarten teacher and 1 preschool teacher controlled and supervised the performance of the assessment sessions. Those responsible of administering the tests were 2 students on the last year of the degree in Early Childhood Education who were specializing in physical education and who were familiar with the performance protocol of those tests. During the first assessment session of the first week, the BMI of half of the group of students who were going to be tested was established. Body weight and height were determined by means of a Tefal digital scale (type PP1200VO) (Groupe Seb
Iberica, Barcelona, Spain) and a Seca mobile rod (type 225) (Seca, Birmingan, UK). Measurements were taken while the children were wearing light clothing without shoes. The BMI was calculated from the body weight in kilograms divided by height in meters squared. Afterward, the protocol of the handheld dynamometry test was carefully explained, and the children carried it out. After a 10-minute break, the protocol for the bent knee push-up test was explained, and then, it was performed. During the second assessment session, the BMI of the other half of the group was tested, and the children carried out the handheld dynamometry test again. After a 10-minute break, they received an accurate explanation of the timed dipping test protocol, and they carried it out. These first week sessions aimed at the knowledge and familiarization of the children with the correct execution of the 3 strength tests, with the purpose of avoiding a learning effect in the next assessment sessions. During the following week, the group carried out normal psychomotricity sessions, in which played tasks were included and the strength tests execution was reviewed. The third week of the research was used as a test phase. During its first session, the children carried out the handheld dynamometry and the bent knee push-up with a 10-minute interval between both the tests. During the second session, the handheld dynamometry and the timed dipping were administered to each child, also with a 10-minute interval. After the fourth week performing normal psychomotricity sessions, the tests were carried out for the third and last time during the fifth week of the research (retest) following the procedure used during the third week (test). When the bent knee push-up test was administered, 1 researcher demonstrated the skill. A mat was used to avoid the possibility of any graze on their knees when touching the floor. The children were face down on hands and knees position, with the hands under the shoulders, elbows slightly spread, knees close together touching the mat, and feet raised VOLUME 28 | NUMBER 11 | NOVEMBER 2014 |
3227
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
Upper-Body Strength Preschool Children
Figure 2. Dispersal graphs and confidence interval in the reliability analysis of the bent knee push-up, handheld dynamometry, and timed dipping tests at 2 different points (test-retest).
in the air. With the body straight, the children extended the arms completely so that their weight was resting on hands and knees. This was the up position. Then, they bent their arms, lowering the body to the ground so that the chest touched the ground. This was the down position. The children were told to perform as many repetitions as possible. The score was the number of push-ups executed correctly. For the timed dipping, 1 researcher also demonstrated the skill before this test was administered. Two desks of 60 3 60 cm height each were used to carry out the test. The children stood between the 2 desks, which were shoulder-width apart, keeping their elbows bent and their feet touching the floor. In those cases in which the desks were too high for this position, the assessor got up the child and held his or her waist. The test started when the child stretched out his or her elbows, raised his or her feet and kept suspended, although his or her weight was supported by the palm of his or her hands, placed on the respective desk. The score
3228
the
was the amount of time (in seconds) that the position was kept. The test ended when one or both elbows bent or when 1 or 2 hands came off the desk. Handheld Dynamometry. For the purpose of this study, a Takei (Takei Scientific Instruments, Ltd., Yashiroda, Japan) dynamometer (adjustable rectified and complacent handle shape, electromechanical system) was used. The test was conducted in standing position with the wrist in neutral position and the elbow extended, in accordance with the protocol used when applying this dynamometer to children (7). The arm was allowed to move from 1808 of flexion to near 08. The children were given verbal encouragement to apply maximal effort between approximately 908 and 08 of flexion and “squeeze as hard as possible” for at least 2 seconds. Statistical Analyses
The statistical analysis began with a descriptive analysis by means of central tendency measures (mean 6 SD and
TM
Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the
TM
Journal of Strength and Conditioning Research
| www.nsca.com
TABLE 3. Upper-body strength tests validity. Validity statistics Pearson’s correlation coefficient Handheld dynamometry_test (kg)* 3 year olds Bent knee pushr = 0.260; up_test (repetition) Sig = 0.046 Bent knee pushup_retest (repetition) Timed dipping_test (s) Timed dipping_retest (s) 4 year olds Bent knee pushr = 0.521; up_test (repetition) Sig = 0.001 Bent knee pushup_retest (repetition) Timed dipping_test (s) Timed dipping_retest (s) 5 year olds Bent knee pushr = 0.675; up_test (repetition) Sig = 0.001 Bent knee pushup_retest (repetition) Timed dipping_test (s)
Handheld dynamometry_retest (kg)†
Handheld dynamometry_test (kg)z
Handheld dynamometry_retest (kg)§
r = 0.358; Sig = 0.037 r = 0.200; Sig = 0.256 r = 0.330; Sig = 0.046
r = 0.636; Sig = 0.032 r = 0.662; Sig = 0.001 r = 0.673; Sig = 0.001
r = 0.818; Sig = 0.001 r = 0.699; Sig = 0.001
Timed dipping_retest (s)
r = 0.705; Sig = 0.001
Sig = p , 0.05.
*Performed before the bent knee push-up (test). †Performed before the bent knee push-up (retest). zPerformed before the timed dipping (test). §Performed before the timed dipping (retest).
median) stratifying the analysis by age. With the aim of considering whether there are significant differences of the analyzed parameters according to the age, an analysis of variance (ANOVA) was made. Afterward, there was an analysis of the reliability of the tests: bent knee push-up, timed dipping, and handheld dynamometry. The reliability analysis was carried out taking into account the age of the children and the moments of assessment of each test, identifying the coefficient of interclass correlation and the confidence interval. To analyze the construct validity of the bent knee push-up and timed dipping tests, a Pearson’s correlation analysis was carried out. This analysis was developed stratifying it by age and using the handheld dynamometry test as a criterion measure. Statistical analyses were conducted with Statistical Package for Social Sciences 19 (SPSS, Inc., Chicago, IL, USA),
and a linear statistical model was considered for the analysis. The level of significance was set at p # 0.05.
RESULTS Descriptive statistics for the sample, including means, SD, and medians are presented in Table 1. This table shows how the average obtained scores in each of the 3 tests increased according to the performers’ age. Significant differences were observed among the 3 groups in this regard. Responders and nonresponders to the bent knee push-up and timed dipping tests are shown in Figure 1. The statistical analysis carried out from the obtained results in the test and retest of the bent knee push-up proved that this test showed a moderate reliability for children younger than 3 years (intraclass correlation coefficient [ICC], 0.690; 95% confidence interval [95% CI], 0.379–0.845) and high for those VOLUME 28 | NUMBER 11 | NOVEMBER 2014 |
3229
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
Upper-Body Strength Preschool Children
Figure 3. Dispersal graphs and confidence interval in the validity analysis of the bent knee push-up and timed dipping tests at 2 different points (test-retest).
older than 4 years (ICC, 0.848; 95% CI, 0.715–0.919) and 5 years (ICC, 0.702; 95% CI, 0.441–0.841). The contrast between the results of the test and retest of the timed dipping showed that its reliability was also moderate for the children younger than 3 years (ICC, 0.522; 95% CI, 0.422–0.761) and high for those younger than 4 years (ICC, 0.766; 95% CI, 0.560–0.875) and 5 years (ICC, 0.828; 95% CI, 0.677–0.908). The data obtained during the research made it possible to verify the reliability of the handheld dynamometry test in 2 different moments (“a” vs. “b” and “c” vs. “d”), which turned out to be high for the 3 age groups (Table 2). Data regarding the reliability of the 3 tests (test-retest) is shown in Figure 2. The information concerning the construct validity analyzed twice for each test can be observed in Table 3. The mean scores of the handheld dynamometry test (a) showed a significant correlation with the mean scores obtained in the bent knee push-up test (test) regardless of the children’s age. This tendency was also observed when the mean scores of the handheld dynamometry test (b) were compared with those of the bent knee push-up (retest). Regarding the
3230
the
construct validity observed for the timed dipping test, significant correlations were found between its average scores (test) and those obtained in the handheld dynamometry (c) in the 4-year-old and 5-year-old children but not in the 3-year-old children (r = 0.200; p = 0.256). Nevertheless, the comparison of the obtained scores in the timed dipping (retest) and the handheld dynamometry (d) were significant for all the age groups. Data regarding the construct validity of the bent knee push-up and timed dipping tests are shown in Figure 3.
DISCUSSION This research tried to analyze the reliability and criterion validity of the 2 upper-body strength measurement tests when performed by preschoolers. The obtained scores indicate that the bent knee push-up and the timed dipping show a suitable reliability and criterion validity, specially for 4 and 5 year old children. The obtained values for the reliability of the bent knee push-up are similar to those previously informed by other authors (Pearson product
TM
Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the
TM
Journal of Strength and Conditioning Research moment correlation coefficient $ 0.80). However, it should be pointed out that this data was obtained with a smaller sample (n = 43) in which children who were not preschoolers were included (10). Similarly, the reliability of the timed dipping found in this research is similar to the previously observed in a wide sample of preschoolers (ICC, 0.829). Nevertheless, such research was carried out with a non-Caucasian population, and age intervals were not taken into account (16). These aspects, together with the fact that it has not been found any research in which criterion validity for both tests have been analyzed by means of its comparison with a criterion measure in preschool children, give special relevance to the results of this investigation. Still, it is necessary to make some remarks on them. First, although the contrast of the average scores obtained in the test and the retest for each performance, carried out with a week interval, showed the existence of a good reliability, their execution according to the administering protocol was not exempt from certain difficulties. For instance, a considerable number of children showed some difficulty when performing the bent knee push-up correctly, given that they were not able to keep their back straight during the first and even the second trial. This could be the result of the lack of strength in the shoulder muscular groups, which play an important role during the performance of the test (19). However, this test is not considered as a very muscularly demanding task (32). Thus, it can be thought that the aforementioned difficulty that has been observed in these little children could be related to the fact that their body-map (perception of the way their body is located in the space) was not totally developed. Similarly, the execution of the timed dipping was in some cases influenced by the score obtained by other children. Indeed, some of the participants in the study decided to put an end to the test during the first trial when they thought they had already beaten other partners, regardless of the level of fatigue that they were experiencing. These 2 aspects may have had an influence on the test reliability, especially with the younger children, among which we could find the lowest values of the test. Second, although both the bent knee push-up and the timed dipping showed good criterion validity, it is important to point out a methodological issue regarding the psychometric property assessment carried out in the present study. To identify the criterion validity, we decided to use the handheld dynamometry test as a criterion measure, which was executed by means of a Takei dynamometer. Although the validity and reliability of this device has been tested with young people (1) and teenagers (11), according to the authors’ knowledge, its psychometric properties when used by preschoolers have not been established. This fact could be pointed out as an important methodological limitation. Nevertheless, the design of this study made it possible to solve this problem. Thus, the reliability of the handheld dynamometry could be doubly analyzed and for 3 age groups separately, obtaining positive results. This finding supports
| www.nsca.com
the use of the Takei in the preschool setting and, therefore, its use in this investigation. Finally, we must highlight the existent difference between the obtained average score in the test stage and the retest stage in the bent knee push-up by the 5-year-old children, much higher than the one obtained by the rest of the analyzed age groups for this and for the other 2 strength tests carried out. In this regard, although the design of the investigation included a first stage of previous familiarization to avoid a learning effect that could influence the test performance during the test and retest stages, it could be possible that this strategy was not totally effective for the 5-year-old children who carried out the bent knee push-up. Nevertheless, the reliability and construct validity of this test for this age group did not seem to be specially influenced by this fact. According to the authors’ knowledge, there are hardly any test for the upper-body strength assessment in preschoolers whose psychometric properties have been properly studied. Thus, there are some easy execution and administering tests, such as the “throwing distance tennis ball” (27) or the “chair push-up test” (2), but no investigations informing about their validity and reliability have been found. In this line, there are also more frequently used tests, such as the “bent arm hang” whose psychometric properties in the preschool context have been analyzed. Nevertheless, the studies for this purpose show some methodological weaknesses like using a small simple size, not including children younger than 5 years or not having assessed by means of a criterion measure its criterion validity (4,12). Finally, other tests that are widely accepted because they are included in the “Fitnessgram” battery (21), such as the “modified pull-up” or the “908 push-up test,” have also seen their psychometric properties criticized. In the case of the first test, its criterion validity to assess the upper-body strength in preschoolers has been questioned (9). In the second test case, some authors have suggested that its psychometric properties should be assessed again because they were established with small samples, which were conveniently selected (22). Furthermore, other authors have realized that the protocol of the 908 push-up test should be modified to improve its standards (5). Thus, just the “medicine ball throw” test seemed to be a valid and reliable measure of upper-body strength for kindergarten children (9), although its criterion validity through the use of a valid criterion measure needs to be further analyzed. Given this situation, the results of this study are a relevant contribution to the scientific literature referent to the validity and reliability of the tests that can be used to assess upperbody strength in preschool children. Still, the aforementioned results must be interpreted taking into account some limitations in the design of this study and which limit their generalization. First, the maturity level and the initial psychomotricity level of the sample object of study were not assessed. Second, neither the kind of physical VOLUME 28 | NUMBER 11 | NOVEMBER 2014 |
3231
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
Upper-Body Strength Preschool Children activities carried out outside the school timetable nor their own performance before the test or retest stages of the tests in this study were controlled. Finally, although the result of the tests were steady in time (test-retest reliability), it was not assessed the influence that the examiner may have in their performance and assessment (interrater reliability), aspect that may have given more consistency to the outcome in this study.
PRACTICAL APPLICATIONS Physical inactivity contributes to the growing prevalence of health-related problems, which have been observed among preschool-aged children (18). Given that preschool age is a key time for the development of health behaviors (14), effective preventive interventions targeting this population are needed. Consequently, health and physical education professionals must know and be able to administer performance-based measurements that make it possible to identify the HRPF level of preschool-aged children and assess the effects of this kind of interventions. Nevertheless, it is difficult to find tests with these characteristics as the performance-based measures traditionally designed to be used with preschool children aim at assessing the motor competence and the gross motor coordination (8,31). Thus, it is considered that there is scarcity of studies giving information about health-related fitness test that can be used with such young children. The results of this study contribute to solve this situation by providing information about the psychometric characteristics of the bent knee push-up and the timed dipping, tests that are easy to carry out and administer, which make it possible to evaluate upper-body strength level among preschool children. These findings are of importance because it has been observed that upper-body muscular strength can be considered a valid health indicator in growing children (30). The implication for practitioners in using the bent knee push-up and the timed dipping for children older than 3 years and younger than 6 years is that both tests seem to be a valid and reliable field test of upper-body strength for that age group, although some aspects must be taken into account for their administration. Thus, regarding the bent knee push-up, we cannot rule out the influence of some kind learning effect, especially with 5-year-old children. Consequently, it is very important to make sure that the children have understood the test protocol and to give them enough time to become familiar with its performance. In this line, it was observed that the performance of the timed dipping may be shortened according to the perception of the performer about the result he get, having into account the scores obtained by his peers. For this reason, children must try to understand that the test must be performed till the highest level of fatigue that they can endure and the examiner must motivate them during the performance to accomplish it, as well as to compete with themselves. The usefulness of the bent knee push-up and the timed dipping must be confirmed by further research aimed at
3232
the
assessing the effects on the preschool children HRPF level of interventions based on the promotion of physical activity among them. These tests should be used to identify possible expected changes in their muscular strength level induced by such interventions.
REFERENCES 1. Amaral, JF, Mancini, M, and Novo, JM. Comparison of three hand dynamometers in relation to the accuracy and precision of the measurements. Rev Bras Fisioter 16: 216–224, 2012. 2. Arnheim, DD and Sinclair, WA. The Clumsy Child: A Program of Motor Therapy. St. Louis, MO: CV Mosby, 1975. 3. Aya´n, C. Assessing health related fitness in the pre-school setting by means of physical performance batteries: A narrative review. J Physic Educ Sport 13: 287–297, 2013. 4. Bala, G. Some problems and suggestions in measuring motor behaviour of pre-school children. Kinesiologia Slovenica 5: 5–10, 1999. 5. Baumgartner, TA, Oh, S, Chung, H, and Hales, D. Objectivity, reliability, and validity for a revised push-up test protocol. Meas Phys Educ Exerc Sci 6: 225–242, 2002. 6. Bruininks, RH. Bruininks-oseretsky Test of Motor Proficiency: Owner’s Manual. Circle Pines, MN: American Guidance Service, 1978. 7. Cohen, DD, Voss, C, Taylor, MJ, Stasinopoulos, DM, Delextrat, A, and Sandercock, GR. Handgrip strength in English schoolchildren. Acta Paediatr 99: 1065–1072, 2010. 8. Cools, W, De Martelaer, K, Samaey, C, and Andries, C. Movement skill assessment of typically developing preschool children: A review of seven movement skill assessment tools. J Sports Sci Med 8: 154– 168, 2009. 9. Davis, KL, Kang, M, Boswell, BB, DuBose, KD, Altman, SR, and Binkley, HM. Validity and reliability of the medicine ball throw for kindergarten children. J Strength Cond Res 22: 1958–1963, 2008. 10. Deitz, JC, Kartin, D, and Kopp, K. Review of the BruininksOseretsky test of motor proficiency, (BOT-2). Phys Occup Ther Pediatr 27: 87–102, 2007. 11. Espan˜a-Romero, V, Ortega, FB, Vicente-Rodrı´guez, G, Artero, EG, Rey, JP, and Ruiz, JR. Elbow position affects handgrip strength in adolescents: Validity and reliability of Jamar, DynEx, and TKK dynamometers. J Strength Cond Res 24: 272–277, 2010. 12. Fjortoft, I. Motor fitness in pre-primary school children: The EUROFIT motor fitness test explored on 5-7-year-old children. Pediatr Exerc Sci 12: 424–436, 2000. 13. Gabel, L, Obeid, J, Nguyen, T, Proudfoot, N, and Timmons, B. Short-term muscle power and speed in preschoolers exhibit stronger tracking than physical activity. Appl Physiol Nutr Metab 36: 939–945, 2011. 14. Hardy, L, King, L, Espinel, P, Okely, A, and Bauman, A. Methods of the NSW Schools Physical Activity and Nutrition Survey 2010 (SPANS 2010). J Sci Med Sport 14: 390–396, 2011. 15. He´bert, LJ, Maltais, DB, Lepage, C, Saulnier, J, Creˆte, M, and Perron, M. Isometric muscle strength in youth assessed by handheld dynamometry: A feasibility, reliability, and validity study. Pediatr Phys Ther 23: 289–299, 2011. 16. Ikeda, T and Aoyagi, O. Relationships between test characteristics and movement patterns, physical fitness, and measurement characteristics: Suggestions for developing new test items for 2-to 6-year-old children. Human Perform Meas 5: 9–22, 2008. 17. Kambas, A, Venetsanou, F, Giannakidou, D, Fatouros, I, Avloniti, A, Chatzinikolaou, A, Dragandis, D, and Zimmer, R. The MotorProficiency-Test for children between 4 and 6 years of age (MOT 4–6): An investigation of its suitability in Greece. Res Dev Disabil 33: 1626–1632, 2012.
TM
Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the
TM
Journal of Strength and Conditioning Research 18. Klein, D, De Toia, D, Weber, S, Wessely, N, Koch, B, Dordel, S, Sreeram, N, Tokarski, W, Stru¨der, H, and Graf, C. Effects of a low threshold health promotion intervention on the BMI in pre-school children under consideration of parental participation. ESPEN J 5: e125–e131, 2010. 19. Maenhout, A, Van Praet, K, Pizzi, L, Van Herzeele, M, and Cools, A. Electromyographic analysis of knee push up plus variations: What is the influence of the kinetic chain on scapular muscle activity? Br J Sports Med 44: 1010–1015, 2010. 20. Matsuda, I and Kondo, M. A Study on the Motor Abilities of Preschool Children: A Composition of a Standard of the Development of the Motor Abilities of Preschool Children. Vol. 7. Tokyo: Bulletin of the Faculty of Physical Education, Tokyo University of Education, 1968. pp. 33–45. 21. Meredith, MD and Welk, GJ. Fitnessgram_/Activitygram_ Test Administration Manual (4th ed.). Dallas, TX: The Cooper Institute, 2007. 22. Morrow, J, Martin, SB, and Jackson, AW. Reliability and validity of the Fitnessgram: Quality of teacher-collected health-related fitness surveillance data. Res Q Exerc Sport 81: 24S–30S, 2010. 23. Obeid, J, Nguyen, T, Gabel, L, and Timmons, B. Physical activity in Ontario preschoolers: Prevalence and measurement issues. Appl Physiol Nutr Metab 36: 291–297, 2011. 24. Pe´rez, C. Fitness evaluation in the context of early childhood education: Practical applications. Apunts: Educacion Fisica e Sport 112: 52–62, 2013. 25. Rikli, R. The reliability of distance run tests for children in grades K-4. Res Q Exerc Sport 63: 270–276, 1992.
| www.nsca.com
26. Ruiz, JR, Ortega, FB, Gutierrez, A, Meusel, D, Sjo¨stro¨m, M, and Castillo, MJ. Health-related fitness assessment in childhood and adolescence: A European approach based on the AVENA, EYHS and HELENA studies. J Public Health 14: 269–277, 2006. 27. Science and Education Department of China State Sport General Administration. Research of Physical Fitness Surveillance System of China. Beijing: Beijing University of Physical Education Publishing House, 2000. 28. Slater, LM, Hillier, SL, and Civetta, LR. The clinimetric properties of performance-based gross motor tests used for children with developmental coordination disorder: A systematic review. Pediatr Phys Ther 22: 170–179, 2010. 29. Stark, T, Walker, B, Phillips, JK, Fejer, R, and Beck, R. Hand-held dynamometry correlation with the gold standard isokinetic dynamometry: A systematic review. PM R 3: 472–479, 2011. 30. Valtuen˜a, J, Gracia-Marco, L, Huybrechts, I, Breidenassel, C, Ferrari, M, Gottrand, F, Dallongeville, J, Sioen, I, Gutierrez, A, Kersting, M, Kafatos, A, Manios, Y, Widhalm, K, Moreno, LA, and Gonza´lez-Gross, M; On Behalf of the Helena Study Group. Cardiorespiratory fitness in males, and upper limbs muscular strength in females, are positively related with 25-hydroxyvitamin d plasma concentrations in european adolescents: The HELENA study. QJM 106: 809–821, 2013. 31. Wiart, L and Darrah, J. Review of four tests of gross motor development. Dev Med Child Neurol 43: 279–285, 2001. 32. Wood, H and Baumgartner, T. Objectivity, reliability, and validity of the bent-knee push-up for college-age women. Meas Phys Educ Exerc Sci 8: 203–212, 2004.
VOLUME 28 | NUMBER 11 | NOVEMBER 2014 |
3233
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
AND
EVA QUIREZA2
1
Department of Special Didactics; 2Healthy-Fit Research Group, Faculty of Education and Sport Science, University of Vigo, Pontevedra, Spain ABSTRACT
INTRODUCTION
Aya´n Pe´rez, C, Cancela, JM, Senra, I, and Quireza, E. Validity and reliability of 2 upper-body strength tests for preschool children. J Strength Cond Res 28(11): 3224–3233, 2014— The “bent knee push-up” and the “timed dipping” are 2 upperbody strength assessment tests whose simplicity and ease of use make them suitable to be applied to preschool children. Nevertheless, their psychometric properties have not been properly studied for this age group. This study aimed to fill this gap. To accomplish this, both tests were administered 3 times (familiarization, test, and retest) to a group of 120 children (mean age, 48.60 6 9.94 months; body mass index, 17.05 6 2.04 kg$m22; 50% girls), with a week interval between each assessment. The obtained results in each test were compared with those obtained in the “handheld dynamometry” (the criterion measure) with the aim of identifying their construct validity. The reliability of the tests was analyzed by comparing the obtained results during the second (test) and the third (retest) time that each 1 was carried out. Intraclass correlation coefficients (ICCs) were used to examine reliability, and Pearson’s correlation coefficients were used to examine validity. The bent knee push-up showed a moderate reliability for 3-year-old children (ICC, 0.690; 95% confidence interval [95% CI], 0.379–0.845) and high for those aged 4 years (ICC, 0.848; 95% CI, 0.715–0.919) and 5 (ICC, 0.702; 95% CI, 0.441–0.841). Similar results were obtained for the timed dipping, in children aged 3 years (ICC, 0.522; 95% CI, 0.422– 0.761), 4 years (ICC, 0.766; 95% CI, 0.560–0.875), and 5 years (ICC, 0.828; 95% CI, 0.677–0.908). Both tests showed a good construct validity, with the exception of the timed dipping in 3-year-old children (r = 0.200; p = 0.256). The results in this study suggest that the bent knee push-up and the timed dipping show psychometric properties that make them suitable to be used in research carried out with preschoolers.
KEY WORDS fitness, muscular, kindergarten
Address correspondence to Carlos Aya´n Pe´rez, [email protected]. 28(11)/3224–3233 Journal of Strength and Conditioning Research Ó 2014 National Strength and Conditioning Association
3224
the
H
ealth-related physical fitness (HRPF) assessment can help to identify children at the risk for major public health diseases and to establish dose-response relationships between activity and health outcomes, and intervention effectiveness (26). Given the emerging physical inactivity detected among preschool-aged children and the future health problems that may be developed as a consequence, HRPF assessment seems to be especially important in early childhood (23). When measuring HRPF in preschool children, the identification of the muscular strength level seems to be a matter of great interest, although little research has dealt with this issue (13). One of the possible reasons for this could be the lack of clinically viable and practical-to-use performance-based tests for the strength-level assessment in children younger than 6 years (24). These kind of tests are defined by a number of characteristics, among which we must highlight the lack of complicated equipment requirements, short administration time, easiness, simple employment and interpretation scoring system, and the presence of excellent psychometric properties, determined basically by the reliability the test shows when it is used with the target population (28). A test is reliable when it demonstrates that test scores are stable over time (test-retest reliability). Validity can be demonstrated by showing that the measure correlates significantly with another test that measures the same constructs (criterion validity). Although in the scientific literature, we can find muscular strength assessment tests that fulfill these characteristics, those are basically focused on the lower-body performance (3). Thus, we can claim that there is an obvious lack of assessment tests for the upper-body muscular strength (9). Under these circumstances, it seems necessary to carry out some research to analyze the reliability and validity of tests that, being easy to carry out and without complicated equipment requirements, allow us to know the upper-body strength level in preschool children. A test that seems to fulfill these characteristics is the “bent knee push-up” included in the “Bruininks-Oseretsky Test of Motor Proficiency” battery under the name of “modified push-up test” (6). Nevertheless, and although the battery is widely used with preschool children, the analysis of its validity and reliability has not been carried out with samples
TM
Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the
TM
Journal of Strength and Conditioning Research
| www.nsca.com
TABLE 1. Descriptive statistics for the sample.* 3 year old, Mean 6 SD 4 year old, Mean 6 SD 5 year old, Mean 6 SD (median) (median) (median) Height (m) Weight (kg) BMI (kg$m22) Bent knee push-up_test (repetition) Bent knee push-up_retest (repetition) Timed dipping_test (s) Timed dipping_retest (s) Handheld dynamometry_test (kg)† Handheld dynamometry_retest (kg)z Handheld dynamometry_ test (kg)§ Handheld dynamometry_ retest (kg)k
1.01 17.76 17.48 6.68
6 6 6 6
0.43 2.11 2.12 4.50
(1.01) (18.40) (17.26) (4.99)
5.86 6 5.08 (4.12)
1.08 19.48 16.78 11.24
6 6 6 6
0.52 2.95 1.79 8.49
(1.08) 1.15 6 0.52 (19.00) 22.31 6 3.78 (16.45) 16.89 6 2.22 (8.96) 20.33 6 18.15
(115.50) (22.00) (16.70) (14.85)
ANOVA F
Sig
77.410 22.469 1.285 12.949
0.001 0.001 0.281 0.001
12.18 6 10.45 (10.22) 25.50 6 22.83 (15.83) 16.706 0.001
2.37 6 2.21 (2.00) 3.62 6 2.09 (4.00) 4.45 6 1.70 (4.00)
4.76 6 3.09 (5.00) 5.22 6 3.74 (5.00) 5.91 6 2.07 (6.00)
9.81 6 5.03 (9.00) 7.71 6 3.64 (7.50) 7.83 6 2.07 (8.00)
41.052 0.001 14.964 0.001 29.652 0.001
4.44 6 1.57 (4.50)
6.29 6 1.65 (6.00)
7.69 6 1.86 (7.50)
26.696 0.001
4.46 6 1.07 (5.00)
5.70 6 1.52 (5.50)
7.15 6 2.04 (7.00)
34.604 0.001
4.71 6 1.52 (4.75)
6.18 6 1.58 (6.00)
7.71 6 2.22 (8.00)
25.747 0.001
*ANOVA = analysis of variance; BMI = body mass index; Sig = p , 0.05. †Performed before the bent knee push-up (test). zPerformed before the bent knee push-up (retest). §Performed before the timed dipping (test). kPerformed before the timed dipping (retest).
specifically composed of preschool children (10). Thus, it seems necessary to establish the psychometric characteristics of the bent knee push-up test when it is performed by children younger than 6 years. Similarly, another test that initially seems to fulfill the previously mentioned characteristics is the “timed dipping” (20). Nevertheless, although its psychometric properties have been contrasted when the test was administered to a group of Japanese preschoolers (16), they should be further analyzed for 2 reasons. First, there are no data on the timed dipping criterion validity, which offers greater objectivity because it involves correlating test scores with another established test that also measures the same fitness component. Second, there is a need to test the psychometric properties of the timed dipping when it is administered to a different ethnic group, as it is a variable that may have some influence over them (17). Given this situation, the aim of this investigation was to study the reliability and construct validity of the bent knee push-up and timed dipping tests when applied to a group of European preschoolers.
METHODS Experimental Approach to the Problem
The purpose of this study was to establish reliability and construct validity evidence for the bent knee push-up and
timed dipping tests in a sample composed of children older than 3 years and younger than 6 years. There are no specific independent and dependent variables for the designed study. Therefore, to analyze the reliability of the strength tests targeted in this study, we followed the protocol that was established by other authors who have analyzed the psychometric properties of different performance-based fitness measures with preschool children (25). Thus, the tests were carried out in 3 stages, with 1-week interval. The first stage was used for the introduction and explanation of the tests with the aim of getting the right performance of the children and avoiding a possible learning effect. The obtained results in each test during this first attempt were not taken into account for further analysis. The second stage was used as a test phase and the third stage as a retest phase. The comparison of the obtained scores in the second and in the third stage in each performance allowed us to analyze the reliability of the upper-body muscular strength tests. To be able to compare the construct validity of both the bent knee push-up and the timed dipping tests, we decided to compare the obtained results in each of them with those obtained after the “handheld dynamometry” test performance. This test was chosen as criterion measure because it can be considered as a gold standard when it comes to assess the upper-body strength (29), and it has proved to be VOLUME 28 | NUMBER 11 | NOVEMBER 2014 |
3225
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
Upper-Body Strength Preschool Children valid and reliable when it is employed with preschool children (15). Having into account that this investigation design included 2 valid times of assessment and data collection for each test (test and retest phases) and that the handheld dynamometry was always administered before the performance of the bent knee push-up and the timed dipping, the construct validity of both tests could be analyzed twice throughout the study with the aim of showing more solid results. Thus, the bent knee push-up (test) score was compared with the score obtained in the handheld dynamometry, (a) which was carried out just before it. In the same way, the score obtained in the third performance of the bent knee push-up (retest) was compared with the results obtained in the handheld dynamometry (b) carried out during the same assessment session. Also, the obtained score after the second performance of the timed dipping (test) was compared with the one obtained in the handheld dynamometry (c) carried out before the first. Finally, the obtained scores after the third performance of the timed dipping (retest) and the previous performance to this attempt of the handheld dynamometry (d) were also compared. A priori power calculation for the validity study indicated that a sample size of 120 would be necessary, given the effect size of 0.4, the power of 0.95, and an expected proportion loss 0.5. Power calculation is not applicable for a reliability study. Subjects Figure 1. Analysis of the responders and nonresponders to the “bent knee push-up” and “timed dipping” tests.
3226
the
The research was conducted on a sample of 120 Caucasian healthy Spanish urban children recruited from 2 kindergartens in the north of Spain. Inclusion
TM
Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the
TM
Journal of Strength and Conditioning Research
| www.nsca.com
TABLE 2. Upper-body strength tests reliability.* Reliability statistics 3 year olds
Bent knee push-up_test-retest (repetition) Timed dipping_test-retest (s) Handheld dynamometry_test-retest (kg)†z Handheld dynamometry_test-retest (kg)§k
4 year olds
5 year olds
ICC
95% CI
ICC
95% CI
ICC
95% CI
0.690 0.522 0.891 0.821
0.379–0.845 0.422–0.761 0.825–0.938 0.698–0.903
0.848 0.766 0.924 0.843
0.715–0.919 0.560–0.875 0.883–0.955 0.747–0.909
0.702 0.828 0.876 0.871
0.441–0.841 0.677–0.908 0.809–0.926 0.758–0.931
*ICC = intraclass correlation coefficient; 95% CI = 95% confidence interval. †Performed before the bent knee push-up (test). zPerformed before the bent knee push-up (retest). §Performed before the timed dipping (test). kPerformed before the timed dipping (retest).
criteria for participation were being enrolled in the first and second levels of the first stage of the Spanish Education System and having no pathologies that prevented the performance of the muscular strength tests proposed for the study. At the end of the research, the data were collected from 60 boys (mean age, 49.20 6 9.80 months; mean body mass index (BMI), 17.57 6 2.06 kg$m22) and 60 girls (mean age, 48.00 6 10.12 months; mean BMI, 16.50 6 1.93 kg$m22). Written informed consent was obtained from school principals and parents or guardians of all the children who took part in the study, before participating in the research protocol. The study was conducted in accordance with the Declaration of Helsinki, and all the study procedures were approved by our institutional ethics committee. Procedure
The research took place during the second term of the academic year 2012-2013, being all the tests carried out during each of the 2 weekly sessions of psychomotricity in which children were taking part and that are part of the preschool academic curriculum in Spain. Such sessions took place 3 hours after the children started their school day and were carried out with a 3-day interval between them. The children not allowed to drink or eat any kind of food during the 2 hours before the assessment sessions, which were carried out in groups of 20 children in 5-week blocks. One kindergarten teacher and 1 preschool teacher controlled and supervised the performance of the assessment sessions. Those responsible of administering the tests were 2 students on the last year of the degree in Early Childhood Education who were specializing in physical education and who were familiar with the performance protocol of those tests. During the first assessment session of the first week, the BMI of half of the group of students who were going to be tested was established. Body weight and height were determined by means of a Tefal digital scale (type PP1200VO) (Groupe Seb
Iberica, Barcelona, Spain) and a Seca mobile rod (type 225) (Seca, Birmingan, UK). Measurements were taken while the children were wearing light clothing without shoes. The BMI was calculated from the body weight in kilograms divided by height in meters squared. Afterward, the protocol of the handheld dynamometry test was carefully explained, and the children carried it out. After a 10-minute break, the protocol for the bent knee push-up test was explained, and then, it was performed. During the second assessment session, the BMI of the other half of the group was tested, and the children carried out the handheld dynamometry test again. After a 10-minute break, they received an accurate explanation of the timed dipping test protocol, and they carried it out. These first week sessions aimed at the knowledge and familiarization of the children with the correct execution of the 3 strength tests, with the purpose of avoiding a learning effect in the next assessment sessions. During the following week, the group carried out normal psychomotricity sessions, in which played tasks were included and the strength tests execution was reviewed. The third week of the research was used as a test phase. During its first session, the children carried out the handheld dynamometry and the bent knee push-up with a 10-minute interval between both the tests. During the second session, the handheld dynamometry and the timed dipping were administered to each child, also with a 10-minute interval. After the fourth week performing normal psychomotricity sessions, the tests were carried out for the third and last time during the fifth week of the research (retest) following the procedure used during the third week (test). When the bent knee push-up test was administered, 1 researcher demonstrated the skill. A mat was used to avoid the possibility of any graze on their knees when touching the floor. The children were face down on hands and knees position, with the hands under the shoulders, elbows slightly spread, knees close together touching the mat, and feet raised VOLUME 28 | NUMBER 11 | NOVEMBER 2014 |
3227
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
Upper-Body Strength Preschool Children
Figure 2. Dispersal graphs and confidence interval in the reliability analysis of the bent knee push-up, handheld dynamometry, and timed dipping tests at 2 different points (test-retest).
in the air. With the body straight, the children extended the arms completely so that their weight was resting on hands and knees. This was the up position. Then, they bent their arms, lowering the body to the ground so that the chest touched the ground. This was the down position. The children were told to perform as many repetitions as possible. The score was the number of push-ups executed correctly. For the timed dipping, 1 researcher also demonstrated the skill before this test was administered. Two desks of 60 3 60 cm height each were used to carry out the test. The children stood between the 2 desks, which were shoulder-width apart, keeping their elbows bent and their feet touching the floor. In those cases in which the desks were too high for this position, the assessor got up the child and held his or her waist. The test started when the child stretched out his or her elbows, raised his or her feet and kept suspended, although his or her weight was supported by the palm of his or her hands, placed on the respective desk. The score
3228
the
was the amount of time (in seconds) that the position was kept. The test ended when one or both elbows bent or when 1 or 2 hands came off the desk. Handheld Dynamometry. For the purpose of this study, a Takei (Takei Scientific Instruments, Ltd., Yashiroda, Japan) dynamometer (adjustable rectified and complacent handle shape, electromechanical system) was used. The test was conducted in standing position with the wrist in neutral position and the elbow extended, in accordance with the protocol used when applying this dynamometer to children (7). The arm was allowed to move from 1808 of flexion to near 08. The children were given verbal encouragement to apply maximal effort between approximately 908 and 08 of flexion and “squeeze as hard as possible” for at least 2 seconds. Statistical Analyses
The statistical analysis began with a descriptive analysis by means of central tendency measures (mean 6 SD and
TM
Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the
TM
Journal of Strength and Conditioning Research
| www.nsca.com
TABLE 3. Upper-body strength tests validity. Validity statistics Pearson’s correlation coefficient Handheld dynamometry_test (kg)* 3 year olds Bent knee pushr = 0.260; up_test (repetition) Sig = 0.046 Bent knee pushup_retest (repetition) Timed dipping_test (s) Timed dipping_retest (s) 4 year olds Bent knee pushr = 0.521; up_test (repetition) Sig = 0.001 Bent knee pushup_retest (repetition) Timed dipping_test (s) Timed dipping_retest (s) 5 year olds Bent knee pushr = 0.675; up_test (repetition) Sig = 0.001 Bent knee pushup_retest (repetition) Timed dipping_test (s)
Handheld dynamometry_retest (kg)†
Handheld dynamometry_test (kg)z
Handheld dynamometry_retest (kg)§
r = 0.358; Sig = 0.037 r = 0.200; Sig = 0.256 r = 0.330; Sig = 0.046
r = 0.636; Sig = 0.032 r = 0.662; Sig = 0.001 r = 0.673; Sig = 0.001
r = 0.818; Sig = 0.001 r = 0.699; Sig = 0.001
Timed dipping_retest (s)
r = 0.705; Sig = 0.001
Sig = p , 0.05.
*Performed before the bent knee push-up (test). †Performed before the bent knee push-up (retest). zPerformed before the timed dipping (test). §Performed before the timed dipping (retest).
median) stratifying the analysis by age. With the aim of considering whether there are significant differences of the analyzed parameters according to the age, an analysis of variance (ANOVA) was made. Afterward, there was an analysis of the reliability of the tests: bent knee push-up, timed dipping, and handheld dynamometry. The reliability analysis was carried out taking into account the age of the children and the moments of assessment of each test, identifying the coefficient of interclass correlation and the confidence interval. To analyze the construct validity of the bent knee push-up and timed dipping tests, a Pearson’s correlation analysis was carried out. This analysis was developed stratifying it by age and using the handheld dynamometry test as a criterion measure. Statistical analyses were conducted with Statistical Package for Social Sciences 19 (SPSS, Inc., Chicago, IL, USA),
and a linear statistical model was considered for the analysis. The level of significance was set at p # 0.05.
RESULTS Descriptive statistics for the sample, including means, SD, and medians are presented in Table 1. This table shows how the average obtained scores in each of the 3 tests increased according to the performers’ age. Significant differences were observed among the 3 groups in this regard. Responders and nonresponders to the bent knee push-up and timed dipping tests are shown in Figure 1. The statistical analysis carried out from the obtained results in the test and retest of the bent knee push-up proved that this test showed a moderate reliability for children younger than 3 years (intraclass correlation coefficient [ICC], 0.690; 95% confidence interval [95% CI], 0.379–0.845) and high for those VOLUME 28 | NUMBER 11 | NOVEMBER 2014 |
3229
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
Upper-Body Strength Preschool Children
Figure 3. Dispersal graphs and confidence interval in the validity analysis of the bent knee push-up and timed dipping tests at 2 different points (test-retest).
older than 4 years (ICC, 0.848; 95% CI, 0.715–0.919) and 5 years (ICC, 0.702; 95% CI, 0.441–0.841). The contrast between the results of the test and retest of the timed dipping showed that its reliability was also moderate for the children younger than 3 years (ICC, 0.522; 95% CI, 0.422–0.761) and high for those younger than 4 years (ICC, 0.766; 95% CI, 0.560–0.875) and 5 years (ICC, 0.828; 95% CI, 0.677–0.908). The data obtained during the research made it possible to verify the reliability of the handheld dynamometry test in 2 different moments (“a” vs. “b” and “c” vs. “d”), which turned out to be high for the 3 age groups (Table 2). Data regarding the reliability of the 3 tests (test-retest) is shown in Figure 2. The information concerning the construct validity analyzed twice for each test can be observed in Table 3. The mean scores of the handheld dynamometry test (a) showed a significant correlation with the mean scores obtained in the bent knee push-up test (test) regardless of the children’s age. This tendency was also observed when the mean scores of the handheld dynamometry test (b) were compared with those of the bent knee push-up (retest). Regarding the
3230
the
construct validity observed for the timed dipping test, significant correlations were found between its average scores (test) and those obtained in the handheld dynamometry (c) in the 4-year-old and 5-year-old children but not in the 3-year-old children (r = 0.200; p = 0.256). Nevertheless, the comparison of the obtained scores in the timed dipping (retest) and the handheld dynamometry (d) were significant for all the age groups. Data regarding the construct validity of the bent knee push-up and timed dipping tests are shown in Figure 3.
DISCUSSION This research tried to analyze the reliability and criterion validity of the 2 upper-body strength measurement tests when performed by preschoolers. The obtained scores indicate that the bent knee push-up and the timed dipping show a suitable reliability and criterion validity, specially for 4 and 5 year old children. The obtained values for the reliability of the bent knee push-up are similar to those previously informed by other authors (Pearson product
TM
Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the
TM
Journal of Strength and Conditioning Research moment correlation coefficient $ 0.80). However, it should be pointed out that this data was obtained with a smaller sample (n = 43) in which children who were not preschoolers were included (10). Similarly, the reliability of the timed dipping found in this research is similar to the previously observed in a wide sample of preschoolers (ICC, 0.829). Nevertheless, such research was carried out with a non-Caucasian population, and age intervals were not taken into account (16). These aspects, together with the fact that it has not been found any research in which criterion validity for both tests have been analyzed by means of its comparison with a criterion measure in preschool children, give special relevance to the results of this investigation. Still, it is necessary to make some remarks on them. First, although the contrast of the average scores obtained in the test and the retest for each performance, carried out with a week interval, showed the existence of a good reliability, their execution according to the administering protocol was not exempt from certain difficulties. For instance, a considerable number of children showed some difficulty when performing the bent knee push-up correctly, given that they were not able to keep their back straight during the first and even the second trial. This could be the result of the lack of strength in the shoulder muscular groups, which play an important role during the performance of the test (19). However, this test is not considered as a very muscularly demanding task (32). Thus, it can be thought that the aforementioned difficulty that has been observed in these little children could be related to the fact that their body-map (perception of the way their body is located in the space) was not totally developed. Similarly, the execution of the timed dipping was in some cases influenced by the score obtained by other children. Indeed, some of the participants in the study decided to put an end to the test during the first trial when they thought they had already beaten other partners, regardless of the level of fatigue that they were experiencing. These 2 aspects may have had an influence on the test reliability, especially with the younger children, among which we could find the lowest values of the test. Second, although both the bent knee push-up and the timed dipping showed good criterion validity, it is important to point out a methodological issue regarding the psychometric property assessment carried out in the present study. To identify the criterion validity, we decided to use the handheld dynamometry test as a criterion measure, which was executed by means of a Takei dynamometer. Although the validity and reliability of this device has been tested with young people (1) and teenagers (11), according to the authors’ knowledge, its psychometric properties when used by preschoolers have not been established. This fact could be pointed out as an important methodological limitation. Nevertheless, the design of this study made it possible to solve this problem. Thus, the reliability of the handheld dynamometry could be doubly analyzed and for 3 age groups separately, obtaining positive results. This finding supports
| www.nsca.com
the use of the Takei in the preschool setting and, therefore, its use in this investigation. Finally, we must highlight the existent difference between the obtained average score in the test stage and the retest stage in the bent knee push-up by the 5-year-old children, much higher than the one obtained by the rest of the analyzed age groups for this and for the other 2 strength tests carried out. In this regard, although the design of the investigation included a first stage of previous familiarization to avoid a learning effect that could influence the test performance during the test and retest stages, it could be possible that this strategy was not totally effective for the 5-year-old children who carried out the bent knee push-up. Nevertheless, the reliability and construct validity of this test for this age group did not seem to be specially influenced by this fact. According to the authors’ knowledge, there are hardly any test for the upper-body strength assessment in preschoolers whose psychometric properties have been properly studied. Thus, there are some easy execution and administering tests, such as the “throwing distance tennis ball” (27) or the “chair push-up test” (2), but no investigations informing about their validity and reliability have been found. In this line, there are also more frequently used tests, such as the “bent arm hang” whose psychometric properties in the preschool context have been analyzed. Nevertheless, the studies for this purpose show some methodological weaknesses like using a small simple size, not including children younger than 5 years or not having assessed by means of a criterion measure its criterion validity (4,12). Finally, other tests that are widely accepted because they are included in the “Fitnessgram” battery (21), such as the “modified pull-up” or the “908 push-up test,” have also seen their psychometric properties criticized. In the case of the first test, its criterion validity to assess the upper-body strength in preschoolers has been questioned (9). In the second test case, some authors have suggested that its psychometric properties should be assessed again because they were established with small samples, which were conveniently selected (22). Furthermore, other authors have realized that the protocol of the 908 push-up test should be modified to improve its standards (5). Thus, just the “medicine ball throw” test seemed to be a valid and reliable measure of upper-body strength for kindergarten children (9), although its criterion validity through the use of a valid criterion measure needs to be further analyzed. Given this situation, the results of this study are a relevant contribution to the scientific literature referent to the validity and reliability of the tests that can be used to assess upperbody strength in preschool children. Still, the aforementioned results must be interpreted taking into account some limitations in the design of this study and which limit their generalization. First, the maturity level and the initial psychomotricity level of the sample object of study were not assessed. Second, neither the kind of physical VOLUME 28 | NUMBER 11 | NOVEMBER 2014 |
3231
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
Upper-Body Strength Preschool Children activities carried out outside the school timetable nor their own performance before the test or retest stages of the tests in this study were controlled. Finally, although the result of the tests were steady in time (test-retest reliability), it was not assessed the influence that the examiner may have in their performance and assessment (interrater reliability), aspect that may have given more consistency to the outcome in this study.
PRACTICAL APPLICATIONS Physical inactivity contributes to the growing prevalence of health-related problems, which have been observed among preschool-aged children (18). Given that preschool age is a key time for the development of health behaviors (14), effective preventive interventions targeting this population are needed. Consequently, health and physical education professionals must know and be able to administer performance-based measurements that make it possible to identify the HRPF level of preschool-aged children and assess the effects of this kind of interventions. Nevertheless, it is difficult to find tests with these characteristics as the performance-based measures traditionally designed to be used with preschool children aim at assessing the motor competence and the gross motor coordination (8,31). Thus, it is considered that there is scarcity of studies giving information about health-related fitness test that can be used with such young children. The results of this study contribute to solve this situation by providing information about the psychometric characteristics of the bent knee push-up and the timed dipping, tests that are easy to carry out and administer, which make it possible to evaluate upper-body strength level among preschool children. These findings are of importance because it has been observed that upper-body muscular strength can be considered a valid health indicator in growing children (30). The implication for practitioners in using the bent knee push-up and the timed dipping for children older than 3 years and younger than 6 years is that both tests seem to be a valid and reliable field test of upper-body strength for that age group, although some aspects must be taken into account for their administration. Thus, regarding the bent knee push-up, we cannot rule out the influence of some kind learning effect, especially with 5-year-old children. Consequently, it is very important to make sure that the children have understood the test protocol and to give them enough time to become familiar with its performance. In this line, it was observed that the performance of the timed dipping may be shortened according to the perception of the performer about the result he get, having into account the scores obtained by his peers. For this reason, children must try to understand that the test must be performed till the highest level of fatigue that they can endure and the examiner must motivate them during the performance to accomplish it, as well as to compete with themselves. The usefulness of the bent knee push-up and the timed dipping must be confirmed by further research aimed at
3232
the
assessing the effects on the preschool children HRPF level of interventions based on the promotion of physical activity among them. These tests should be used to identify possible expected changes in their muscular strength level induced by such interventions.
REFERENCES 1. Amaral, JF, Mancini, M, and Novo, JM. Comparison of three hand dynamometers in relation to the accuracy and precision of the measurements. Rev Bras Fisioter 16: 216–224, 2012. 2. Arnheim, DD and Sinclair, WA. The Clumsy Child: A Program of Motor Therapy. St. Louis, MO: CV Mosby, 1975. 3. Aya´n, C. Assessing health related fitness in the pre-school setting by means of physical performance batteries: A narrative review. J Physic Educ Sport 13: 287–297, 2013. 4. Bala, G. Some problems and suggestions in measuring motor behaviour of pre-school children. Kinesiologia Slovenica 5: 5–10, 1999. 5. Baumgartner, TA, Oh, S, Chung, H, and Hales, D. Objectivity, reliability, and validity for a revised push-up test protocol. Meas Phys Educ Exerc Sci 6: 225–242, 2002. 6. Bruininks, RH. Bruininks-oseretsky Test of Motor Proficiency: Owner’s Manual. Circle Pines, MN: American Guidance Service, 1978. 7. Cohen, DD, Voss, C, Taylor, MJ, Stasinopoulos, DM, Delextrat, A, and Sandercock, GR. Handgrip strength in English schoolchildren. Acta Paediatr 99: 1065–1072, 2010. 8. Cools, W, De Martelaer, K, Samaey, C, and Andries, C. Movement skill assessment of typically developing preschool children: A review of seven movement skill assessment tools. J Sports Sci Med 8: 154– 168, 2009. 9. Davis, KL, Kang, M, Boswell, BB, DuBose, KD, Altman, SR, and Binkley, HM. Validity and reliability of the medicine ball throw for kindergarten children. J Strength Cond Res 22: 1958–1963, 2008. 10. Deitz, JC, Kartin, D, and Kopp, K. Review of the BruininksOseretsky test of motor proficiency, (BOT-2). Phys Occup Ther Pediatr 27: 87–102, 2007. 11. Espan˜a-Romero, V, Ortega, FB, Vicente-Rodrı´guez, G, Artero, EG, Rey, JP, and Ruiz, JR. Elbow position affects handgrip strength in adolescents: Validity and reliability of Jamar, DynEx, and TKK dynamometers. J Strength Cond Res 24: 272–277, 2010. 12. Fjortoft, I. Motor fitness in pre-primary school children: The EUROFIT motor fitness test explored on 5-7-year-old children. Pediatr Exerc Sci 12: 424–436, 2000. 13. Gabel, L, Obeid, J, Nguyen, T, Proudfoot, N, and Timmons, B. Short-term muscle power and speed in preschoolers exhibit stronger tracking than physical activity. Appl Physiol Nutr Metab 36: 939–945, 2011. 14. Hardy, L, King, L, Espinel, P, Okely, A, and Bauman, A. Methods of the NSW Schools Physical Activity and Nutrition Survey 2010 (SPANS 2010). J Sci Med Sport 14: 390–396, 2011. 15. He´bert, LJ, Maltais, DB, Lepage, C, Saulnier, J, Creˆte, M, and Perron, M. Isometric muscle strength in youth assessed by handheld dynamometry: A feasibility, reliability, and validity study. Pediatr Phys Ther 23: 289–299, 2011. 16. Ikeda, T and Aoyagi, O. Relationships between test characteristics and movement patterns, physical fitness, and measurement characteristics: Suggestions for developing new test items for 2-to 6-year-old children. Human Perform Meas 5: 9–22, 2008. 17. Kambas, A, Venetsanou, F, Giannakidou, D, Fatouros, I, Avloniti, A, Chatzinikolaou, A, Dragandis, D, and Zimmer, R. The MotorProficiency-Test for children between 4 and 6 years of age (MOT 4–6): An investigation of its suitability in Greece. Res Dev Disabil 33: 1626–1632, 2012.
TM
Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the
TM
Journal of Strength and Conditioning Research 18. Klein, D, De Toia, D, Weber, S, Wessely, N, Koch, B, Dordel, S, Sreeram, N, Tokarski, W, Stru¨der, H, and Graf, C. Effects of a low threshold health promotion intervention on the BMI in pre-school children under consideration of parental participation. ESPEN J 5: e125–e131, 2010. 19. Maenhout, A, Van Praet, K, Pizzi, L, Van Herzeele, M, and Cools, A. Electromyographic analysis of knee push up plus variations: What is the influence of the kinetic chain on scapular muscle activity? Br J Sports Med 44: 1010–1015, 2010. 20. Matsuda, I and Kondo, M. A Study on the Motor Abilities of Preschool Children: A Composition of a Standard of the Development of the Motor Abilities of Preschool Children. Vol. 7. Tokyo: Bulletin of the Faculty of Physical Education, Tokyo University of Education, 1968. pp. 33–45. 21. Meredith, MD and Welk, GJ. Fitnessgram_/Activitygram_ Test Administration Manual (4th ed.). Dallas, TX: The Cooper Institute, 2007. 22. Morrow, J, Martin, SB, and Jackson, AW. Reliability and validity of the Fitnessgram: Quality of teacher-collected health-related fitness surveillance data. Res Q Exerc Sport 81: 24S–30S, 2010. 23. Obeid, J, Nguyen, T, Gabel, L, and Timmons, B. Physical activity in Ontario preschoolers: Prevalence and measurement issues. Appl Physiol Nutr Metab 36: 291–297, 2011. 24. Pe´rez, C. Fitness evaluation in the context of early childhood education: Practical applications. Apunts: Educacion Fisica e Sport 112: 52–62, 2013. 25. Rikli, R. The reliability of distance run tests for children in grades K-4. Res Q Exerc Sport 63: 270–276, 1992.
| www.nsca.com
26. Ruiz, JR, Ortega, FB, Gutierrez, A, Meusel, D, Sjo¨stro¨m, M, and Castillo, MJ. Health-related fitness assessment in childhood and adolescence: A European approach based on the AVENA, EYHS and HELENA studies. J Public Health 14: 269–277, 2006. 27. Science and Education Department of China State Sport General Administration. Research of Physical Fitness Surveillance System of China. Beijing: Beijing University of Physical Education Publishing House, 2000. 28. Slater, LM, Hillier, SL, and Civetta, LR. The clinimetric properties of performance-based gross motor tests used for children with developmental coordination disorder: A systematic review. Pediatr Phys Ther 22: 170–179, 2010. 29. Stark, T, Walker, B, Phillips, JK, Fejer, R, and Beck, R. Hand-held dynamometry correlation with the gold standard isokinetic dynamometry: A systematic review. PM R 3: 472–479, 2011. 30. Valtuen˜a, J, Gracia-Marco, L, Huybrechts, I, Breidenassel, C, Ferrari, M, Gottrand, F, Dallongeville, J, Sioen, I, Gutierrez, A, Kersting, M, Kafatos, A, Manios, Y, Widhalm, K, Moreno, LA, and Gonza´lez-Gross, M; On Behalf of the Helena Study Group. Cardiorespiratory fitness in males, and upper limbs muscular strength in females, are positively related with 25-hydroxyvitamin d plasma concentrations in european adolescents: The HELENA study. QJM 106: 809–821, 2013. 31. Wiart, L and Darrah, J. Review of four tests of gross motor development. Dev Med Child Neurol 43: 279–285, 2001. 32. Wood, H and Baumgartner, T. Objectivity, reliability, and validity of the bent-knee push-up for college-age women. Meas Phys Educ Exerc Sci 8: 203–212, 2004.
VOLUME 28 | NUMBER 11 | NOVEMBER 2014 |
3233
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
