Investigation of the motor skills assessments of typically developing preschool children in China

Eleven of 17 studies described the current situation of motor skill assessments in China by using international tools [35‐45]. Three uniquely recognized MSAT were identified across the 11 studies. The Test of Gross Motor Development, second edition, (TGMD-2) was used in nine studies. The Test of Gross Motor Development, third edition, (TGMD-3), and the Movement Assessment Battery for Children, second edition, (MABC-2) were used in one study respectively. Of the included studies, data from 5,529 children from 53 childcare centers were incorporated. Thus, studies included between 47 and 1,823 participants from 1 to 24 childcare centers. Motor skills tests were most commonly conducted at the pre-school site (nine studies); only two were carried out in primary schools and kindergartens. Below are the major features demonstrated (See Table 2).

In terms of the translation of the tools, only Hua showed information about the translation of the MABC-2, and an independent translator back-translated the information [35]. About half of the studies described the recruitment and training process of testers. Li only mentioned that specially trained testers rated the participants’ motor performance according to scoring criteria [36]. The other studies showed training details, including testing process, instrument items, scoring standards, safety management, and practice assessment [35, 41, 43, 45].

Of nine articles that used the TGMD-2 in China, two-thirds used all original items [36‐39, 43], one-third of them used partial items from the instrument [40‐42]. Dai selected running, hopping, kicking, and stationary dribbling as the easiest and most difficult items from the TGMD-2. Horizontal jumping and catching were used to estimate the coordination skills of limbs and joints [40]. To explore the effect of ball game skills on object control skills in children, Chen used all items in the object control skills subscale [41]. In another study, Yu and Chen included horizontal jumping, running, hopping in locomotor skills subscale and catching, stationary dribbling, and underhand rolling in object control skills subscale, without explaining why and how the six items were determined [42]. Moreover, no study adjusted items according to the condition of local children.

Three studies verified both the reliability and validity of international tools [35, 39, 45]. Composite reliability score of locomotor skills and object control skills subscale of the TGMD-2 was 0.75 and 0.67 respectively. To determine the amount of convergence, Pearson's r was calculated and demonstrated the correlation between subscales was 0.56. Besides, the average variance extracted values were higher than the Pearson correlation coefficient, which suggested that subscales of the TGMD-2 had discriminant validity [39]. ICCs of “walking heels raised” and “drawing trail” in the MABC-2 were good but below 0.9, and if the items “walking heels raised” and “drawing trail” were removed, Cronbach’s alpha coefficient would increase. Moreover, the correlation between components of the MABC-2 and the PDMS-2 was high to weak. These findings indicated that the reliability and validity of age band 1 (3–6 years old) of the MABC-2 were fair [35]. Internal consistency reliability (0.81, Cronbach’s Alpha), test-retest reliability (r = 0.974), and inter-rater reliability (r = 0.983) were good; the construct validity was acceptable, indicating that the TGMD-3 was an appropriate tool [45].

Among the three international tools, only the TGMD-3 has norms in Shanghai, and the establishment process is complete [45]. The stratified sampling method was adopted to select 12 kindergartens and 12 primary schools with 1,118 children from both urban and rural environments. The difficulty (0.63) and discrimination (0.4) of the TGMD-3 were appropriate. Validity and reliability were examined. Significant gender and age differences were found in the ball skills subscale and significant age differences in the locomotor skills subscale and total scale of the TGMD-3. Thus, the norm of the ball skills subscale was established by gender and age, and norms of the total scale and locomotor skills subscale were established by age. The original score was converted into a Z score linearly. According to the rule of the Z score, the raw scores of the two subscales and the total scale were converted into the 1–5 level norm.

Eight studies showed that the competence of fundamental movement skills of pre-school children increased with age, which means that older children grasped motor skills better [36‐43]. Mo and Dai found no gender difference in total and subscale scores [37‐40]. In contrast, Ning and Liu observed that boys performed better than girls in gross motor skills [39, 43]; Jia and Liu found that boys were better at controlling balls and girls at locomotor skills [38, 43]. In China, only Liu observed that urban children got higher scores in object control skills than rural children, especially in kicking and overhand throwing [43].

Only three articles examined the reliability and validity of tools, and one mentioned the translation of the tool. The TGMD-2 was the most frequently used international MSAT in China. Two-thirds of studies applying TGMD-2 chose to use all items of the original tool, and one-third selected some of the items for specific test purposes.

A total of six studies among the included articles described four MSAT developed by Chinese researchers [46‐51]. These articles were published between 1994 and 2018. Children were recruited from the community, health care centers, and pre-schools. The sample size of included studies varied from 60 to 2,368 children (See Table 3).

The CDCC measures the development of Chinese children aged 3–6 years based on their psychological features. This test can be used to provide indicators of early childhood development for workers in pre-school education, health care, and pediatrics to facilitate education, assessment, and intervention. The CDCC was initially developed with 22 items in four subtests by combining the psychological development tests with the practical experience of early childhood educators. First, a preliminary test was carried out on 500 children in six Chinese cities. After statistical processing and summary, 16 items were extracted to form the scale. Next, a pilot test was conducted on 2,368 children in 18 cities in China, and the sample was stratified by age, gender, parental education level, and residence [46]. By analyzing the pilot test results, the test sequence and content of test items were adjusted, and the scale standardization was completed. Then the CDCC was field-tested nationwide to assess its feasibility and further modified according to the problems that originated from the pilot test results.

The CDCC consists of 11 items in the intellectual development subscale and five items in the motor development subscale. The motor development subscale includes four items to assess gross motor skills and skills to quickly pick up beans to measure fine motor skills. Each item in the motor development subscale is rated on a six-point rating scale from 0 (weakest performance) to 5 (best performance). Total score, percentile score, and subscale scores are used to express the development level. The CDCC has good reliability and validity. The internal consistency reliability ranges from 0.708 to 0.953, ICC of test-retest reliability was 0.893, and validity was high. Chen and Li found that children’s CDCC score (41.88 ± 28.76) was significantly related to corner activity [47]. Chen et al. used the motor development subscale of CDCC and investigated that the relationship between children’s motor development and outdoor activity quality was weak [48].

The Gross Motor Development Assessment Scale is a quick and convenient test to measure gross motor skills, which creates the opportunity for evaluation and guidance of the current situation of pre-school children’s motor development. Originally, 17 motor skills assessments were extracted to generate 34 items from three subtests. Then a questionnaire with 34 items compiled according to the Likert Scale five-point scoring method was sent to experts for their opinions. Ten items were selected, and scoring criteria were determined to form the first draft. A pilot test was conducted on 56 children to test the acceptability and universality of the first draft. According to the statistical properties results, ten items (below) were included in the final draft.

The scale is divided into locomotor skills, object control skills, and posture control skills subscales. The locomotor subtest consists of four items: running, hopping, horizontal jumping, and leaping. The object control part consists of four items: kicking, catching, one-hand throwing, and stationary dribbling. The posture control subtest consists of two items: one-leg balance and balance beam. A five-point scale from 0 to 4 is used to score the motor performance on each item. It takes about 15 minutes to administer the test. The inter-rater reliability (ranging from 0.954 to 0.988), test-retest reliability (ranging from 0.926 to 0.997), and homogeneity reliability (ranging from 0.837 to 0.861) of the Gross Motor Development Assessment Scale were good. In addition, the construct validity (ranging from 0.607 to 0.890), convergent validity, and discriminant validity were acceptable. Guo found no significant gender difference in subscale scores, while there were significant differences in subscales in different age groups. Therefore, in the establishment of norms, different criteria are required for each age group. However, only the reference version of evaluation standards established on 280 children is available [49].

The Athletic Ability Test Scale of Urban Community Children is designed to assess motor skills in 3–6 years old children in the urban community [50]. The scale is established through the discussion by experts in the field of pre-school education and physical education. It contains three subscales: balance coordination and agility (five items), strength (five items), and manual dexterity (two items). The highest score for each item is 5, and the lowest score is 1. Raw scores are converted into the total score and average score. Boys and girls in the same age group use the same scoring criteria. The test requires equipment such as sandbags and balls. Before the formal test, 12 children were primarily tested to verify the reliability and validity of the scale. Regrettably, no specific information about reliability, validity, and item selection was reported. Sixty children in Beijing participated in the test. Results suggested that the development of balance and coordination, agility, manual dexterity was good, the development of strength and endurance, especially for upper body strength, was not satisfactory. There was no significant gender difference in the total score, but boys performed better in balance, bounce, and endurance than girls. Girls performed better in manual dexterity.

The Evaluation System of Sports Ability is suitable for assessing fundamental movement skills for children aged three to six years. The test items and measurement were generated through two rounds of questionnaires, and items deemed important were retained. The test includes 11 items: 20 meters, standing long jump, walking on hands and feet, five meters back and forth, balance beam, sandbag throw, a ball bouncing, fast run, standing on one foot, skipping rope, and arms support. The scoring criteria are established by the percentile method and T standard score conversion method. Scale reliability of the test ranges from 0.732 to 0.824. Guo and Zhou found no significant gender difference, whereas there were significant differences in terms of age [51].

Among these four self-constructed MSAT, two tools are used to assess gross motor skills, and two measures fine and gross motor skills. Although the CDCC was the most utilized test tool in China, its intellectual development assessment was applied more than its motor development assessment. Therefore, current studies involving the motor development subscale were aimed to explore the relationship between motor development and physical activity. Except for the Athletic Ability Test Scale of Urban Community Children, all other self-constructed tools in China verify their reliability and validity.

The adoption of the international instrument includes back-translation of the test manual to ensure accurate capture of assessment items, preliminary test to explore the suitability, necessary examination of the psychometric property, item selection, and standardization on a fully representative sample [53, 54]. These steps make it possible to make full of the international tools and obtain the objective test results.

Any assessment tool would be useless if it were not valid and reliable [55]. Even if all the assessments are carried out in China, environmental and participant characteristics can affect the reliability and validity of the same tool. However, the composite reliability, convergent validity, and discriminant validity of the most widely used tool, TGMD-2, was verified just in one city [39]. The other studies using this tool in different regions did not examine reliability and validity before the formal trial. Besides, none of the included studies conducting item selection reported the reliability and validity data since the psychometric characteristics of the trimmed tool may change. Lacking information on psychometric properties may lead to the quality and applicability of this research assessment tool.

The item selection lacks a more scientific process and reason [56]. Among three studies using partial items in the TGMD-2 [40‐42], two of them interpreted that the specific purposes for item selection were to estimate coordination skills of joints and the effect of ball games on object control skills, respectively [40, 41]. They offered information about the aims of selection and which items were determined. However, these studies did not explain why these items were selected and whether the selection was validated. The arbitrary selection of items is problematic as it will threaten the comparability of test results across studies and make it difficult to replicate findings [57].

Moreover, the sequence of the preliminary investigation and item selection is missing. Only Dai followed the usual procedure to conduct the pilot test, then to determine the final items by considering the test results and experts’ suggestions together [40]. The other studies identified the test items without a pilot test. Besides, no study considered the sport habits of children while selecting test items, which may affect the test results. For example, when the TGMD-3 was introduced to Germany, two-hand striking was removed because children did not know how to do it [58].

When the international MAST was used in China, the researchers paid less attention to the translation. Specifically, only one study in this review mentioned the translation process, and its translation was administered by an independent translator [35]. Potentially biased translation occurs when the instrument is translated and administered by only one translator [59]. The shortage of rigorous translation processes and professional translators leads to misunderstanding of the instrument.

Furthermore, the testers with appropriate backgrounds play an essential role during the assessment. In this review, five studies using international tools described the source of testers, but most of them had no information about how testers were trained and whether the practice was carried out before formal tests. The testers’ lack of training and practice may lead to implementation problems during the assessment process and affect the data reliability. Besides, only two studies reported inter-rater reliability [35, 45], whereas inter-rater reliability can validate evaluation results.

In this review, the TGMD-2 appears in a large proportion of studies because of its good applicability. It is the most widely used tool globally and has good reliability and validity under different cultural backgrounds [59‐63]. Besides, the TGMD-2 has been translated into Chinese already [59], making it the first choice of MSAT in China. Other tools, such as the BOT-2, the Movement Assessment Battery for Children (MABC), and the Motor-Proficiency-Test for children between four and six years of age (MOT 4–6), are not used regularly or never used, which might result from their late investigation, including the lack of Chinese version. For example, compared with other countries that assessed the MABC in 1996, China used it until 2001 [64, 65].

Moreover, the MABC-2 is often used to identify children with motor impairment, such as developmental coordination disorder (DCD) [14]. Severe neurological and sensory injuries such as cerebral palsy occur in 10% to 15% of preterm/low birth weight infants, but almost 50% of preterm infants develop motor impairment such as DCD [66]. Therefore, several studies in China have used the MBC-2 to investigate the effects of preterm birth on motor performance in pre-school-age children [67, 68], and few studies used the MBC-2 to observe motor performance in normal preschoolers.

In this study, we found that a few international tools were used. The investigation of these tools was incomplete in terms of translation, verification of reliability and validity, item selection, and administration.

Most self-constructed tools are not widely used in China. Specifically, the majority are applied in one city or district, and the number of subjects was less than 1,000 except for the CDCC [46]. Furthermore, many self-constructed tools are developed by individual or research teams with little supports from the government or institutions, leading to limited generalizability and application in one district. Therefore, the sample size is too small and regional to guarantee the representativeness of subjects in terms of China’s geographic diversity. Combined with the application of the international tools described before, simply using the various international tools with good reliability and validity is easier and more convenient for researchers instead of committing substantial resources to develop a national tool. However, there is a strong need to construct a Chinese localized MSAT because of the cultural differences in motor development between Chinese and Western populations [69].

The test content of Chinese self-constructed tools is unbalanced since it focuses on assessing gross motor skills. Although the CDCC has one item, and the Athletic Ability Test Scale of Urban Community has two items to assess fine motor skills, the number of gross motor skills items is more than fine motor skills items. The Evaluation System of Sports Ability and the Gross Motor Development Assessment Scale does not measure fine motor skills. While several fine motor skill tools are referenced in the international context, cultural differences exist in the development progress. Unlike the gross motor skills assessment’s concern about muscles and joints, fine motor skills assessment involves other factors such as visual memory, making it difficult to consider the unidimensionality and equal interval of scoring [69].

The examination of reliability and validity is not comprehensive in self-constructed tools. Generally, the assessment evaluation includes inter-rater reliability, test-retest reliability, and construct validity [25]. However, the Athletic Ability Test Scale of Urban Community Children, the CDCC, and the Evaluation System of Sports Ability did not test the inter-rater reliability. In addition, the Athletic Ability Test Scale of Urban Community Children and the Evaluation System of Sports Ability did not examine the construct validity. The inter-rater reliability reflects the extent to which the data collected are correct representations of the variables measured. Constructing validity represents the extent to which the measurement tool matches what we want to measure [70, 71].

In some included studies, the presentation of instruction was missing. For example, the Athletic Ability Test Scale of Urban Community Children and the Evaluation System of Sports Ability did not provide instructions for readers, which may confuse future researchers who intend to use the same tool.

In this review, we found that the usage range of self-constructed tools is limited in China. In addition, the development process is generally complete but lacks detailed information such as instruction.

Even though some international tools have been examined to be reliable and valid for Chinese preschoolers, cross-cultural differences appeared on some items as the development of motor skills is affected by environmental background and societal attitudes. Therefore, unless the tool has been vigorously evaluated on a fully representative sample of a national subject, it is crucial to examine the reliability and validity when the tool is used in different areas.

The cross-cultural translation process is necessary since it clarifies scientific test instruction to the source instrument [75]. Therefore, if the instrument has no Chinese version yet, the translation procedure should be carried out and presented in the article. The double-back reverse independent translation is always adopted, which involves four bilingual professional translators to complete the independent translations. The translators have no access to the original version of the assessment. After the translation process, all translators will compare the translated version with the original version and revise it [59]. No matter which method is adopted, it is necessary to guarantee the original version’s linguistic, conceptual, operational, and metric equivalence through professional translators and proof panels [76, 77].

Generally, international tools such as adding and deleting items are revised after the preliminary investigation and examination of the psychometric property. Future research should show which items are tested and interpret why these items are determined and the scientific support behind the reason, including the preliminary test results, the advice from experts, and the psychometric quality of the tool. For example, the reliability and validity of items “walking heels raised” and “drawing trail” of the MABC-2 were lower than other items, which suggested that the two items need refinement when the MABC-2 is used on Chinese children [35]. In addition, the scoring criteria of items need to make relative adjustments according to the sport habits in children.

The standardization establishes consistent procedures, including observation, administration, equipment, and scoring rules. However, future research should show more details about the equipment and observation of assessment as they are not found in the included studies. Besides, the recruited testers should be able to master the basic knowledge of the theory and principles of assessments, professional training, and related experience of children [25].

Last but not least, future studies should expand the sample size into developing standardized versions and establishing national norms after the preliminary investigation, which will help in interpreting and comparing the test results.

In the stage for the development of the assessment tool, an initial examination of reliability and validity must be integrated into the construction plan. Firstly, the development details, including why the tool is developed, how items are determined, and how to rank a child’s performance, should be described clearly to strengthen its reliability and validity. Apart from the primary examination, including inter-rater reliability, test-retest reliability, construct validity, and content validity, other tests such as concurrent validity should be used to verify the quantitative and qualitative aspects of motor skills. This step is necessary for constructing the new instrument. Besides, the item selection should be guided by a conceptual and strong framework and examined by construct validity [78].

When a new tool is constructed, the item selection process should be clear and organized. Based on the test purpose, the review of related assessment should be conducted to find the eligible and similar items. For example, many existing tools regard items in the National Standard of Physical Fitness as reference. Then questionnaires should be used to seek expert advice for initial items. After determining original items, the pilot test is necessary to test the difficulty, discrimination, and correlation coefficient between the item and its dimension. During the pilot test, videotapes are found effective in clarifying procedural uncertainties and resolve problems of interpretation in training. The pilot test results will be combined with expert judgment to delete or add items to the tool. The test items can combine assessment with the game to enhance enjoyment and avoid the abnormal behaviors of children in front of strangers and unfamiliar test equipment [50].

When the final draft of the instrument is settled, it can be carried out in the field. Therefore, it is necessary to commit immense resources to collect data on a fully representative sample of a national subject and put the test to a thorough analysis, helping us establish national norms and promote generalization. Noticeably, the subjects from different areas exercise and play differently, have different school experiences, and have quite different gender stereotypes. The published norms and even the nature of the test items require radical reconsideration. Experts should independently score all the items and suggest a specific change of item. Then the modified instrument will be field-tested with a preliminary sample. The adaptation suggested by the experts and field tests should be incorporated into a comprehensive tool.

Given that current self-constructed tools mainly measure gross motor skills, a comprehensive MSAT aiming to assess fine and gross motor skills is needed in China. Fine motor skills are the set of capacities that form the motor skills system with gross motor skills to improve the physical activity level [79]. The pre-school years are when fine motor skills, including bimanual skills, manual dexterity, object manipulation, and eye-hand coordination, develop rapidly [80, 81]. Therefore, the assessment of fine motor skills can complete the understanding of children’s current motor development level and promote children’s later cognitive development, language, and writing skills [82‐84].