The Role of Knowledge and Understanding in Shaping Motivation, Self-Confidence, and Physical Competence in 8–12-Year-Old Students in Physical Education

Abstract

Physical Literacy (PL) assumes mutual reinforcement between its cognitive, affective, and physical domains. However, the role of the cognitive domain within this framework remains insufficiently explored. The aim of this study was to determine the extent to which the level of Knowledge and Understanding is related to Physical Competence, Motivation and Confidence, and Daily Behaviour in children aged 8–12 years. The study also examined whether grade group moderates these relationships. The Canadian Assessment of PL (CAPL-2) tool was administered to a sample of 2030 students from Poland. No consistent or practically meaningful correlations were observed between Knowledge and Understanding and Daily Behaviour across grade groups. Significant, although generally weak, associations were found between Knowledge and Understanding and both Motivation and Confidence and Physical Competence, with the Understanding subdomain showing stronger relationships. Moderation analysis indicated that the association between ‘Knowledge’ and Physical Competence decreased in older grade groups, while no consistent interaction effects were observed for ‘Understanding’. These findings suggest that, within the CAPL-2 operationalisation, the ‘Understanding’ subdomain showed relatively stronger (although still weak) associations; however, these differences were observed within generally weak relationships and should be interpreted cautiously.

1. Introduction

Physical literacy is commonly defined as the motivation and confidence, physical competence, daily behaviours, and knowledge and understanding to value and take responsibility for engagement in physical activities for life (Whitehead, 2001). In this framework, PL is not limited to motor proficiency but encompasses affective, cognitive, and behavioural dimensions of human functioning. According to Whitehead (2010a, 2010b) PL domains are closely related and mutually reinforcing, meaning that development in one area may facilitate growth in others.

With this integrated framework, knowledge and understanding are considered a central component of PL and support the development of other PL domains (Edwards et al., 2017; Ennis, 2015; Lundvall, 2015; Whitehead, 2010a). Physical competence alone may not be sufficient for lifelong engagement in physical activity, and motivation without cognitive awareness may not produce sustained commitment.

To date, numerous studies have been conducted on the relationship between PL domains and various aspects of child and adolescent functioning, including: (i) well-being (Caldwell et al., 2020; Melby et al., 2022); (ii) physical activity behaviours (Bélanger et al., 2018); (iii) sedentary lifestyles (Saunders et al., 2018); (iv) cardiorespiratory fitness (Lang et al., 2018); (v) mental resilience (Jefferies et al., 2019); and (vi) the body weight of the children (Nyström et al., 2018). Relatively few studies have focused specifically on the cognitive domain of PL and its relationships with other domains (Cale & Harris, 2018; Gunnell et al., 2018b). The cognitive domain, operationalised in many studies through the Knowledge and Understanding component of the Canadian Assessment of Physical Literacy (CAPL-2), refers to children’s knowledge about physical activity and health and aspects related to interpreting and applying this knowledge.

Furthermore, existing research suggests that levels of specific domains of physical literacy may vary by gender. Studies conducted in various cultural contexts show that boys more frequently achieve higher scores in terms of everyday behaviours, physical competencies, and overall levels of physical literacy (Li et al., 2020; Mendoza Muñoz et al., 2024). At the same time, findings regarding the cognitive domain are less clear-cut—some studies indicate no significant differences between genders in terms of knowledge and understanding (Gilic et al., 2022; Hadier et al., 2024), whilst others suggest slight or context-dependent differences. This indicates that gender may play a significant role in shaping the physical literacy profile, particularly with regard to the behavioural and physical domains, and should be taken into account in analyses of the relationship between the cognitive domain and the other components of PL.

This issue is particularly relevant because the theoretical construct of PL proposed by Whitehead (Whitehead, 2001) still requires empirical verification (Carl et al., 2022a, 2022b; Durden-Myers et al., 2018; Harvey & Pill, 2019; Lounsbery & McKenzie, 2015). All the more so because the validation studies of the CAPL-2 tool in different countries to date do not fully confirm the mutual reinforcement of the four distinguished domains, and thus the coherence of the theoretical construct of PL. Although CAPL-2 is widely used as an operationalisation of PL in children aged 8–12, it represents a measurement model that translates the philosophical concept of PL into assessable domains. Therefore, examining the relationships among CAPL-2 domains should be understood as an analysis within an operational measurement model of physical literacy, rather than a direct empirical test of its theoretical structure.

The results of previous studies confirm that the level of knowledge significantly correlates with the overall PL score (Dania et al., 2020; Gilic et al., 2022; Gunnell et al., 2018b; Li et al., 2020; Mendoza Muñoz et al., 2024). However, while Gunnell et al. (2018b) observed significant intercorrelations between all PL domains, in other studies, knowledge was most often correlated with the Physical Competence domain (Elsborg et al., 2021; Li et al., 2020), and less often with other PL domains. There are also studies that do not show a correlation of knowledge with any other domain (Dania et al., 2020; Mendoza Muñoz et al., 2024). These equivocal findings suggest that the operational definition of the knowledge construct within PL may influence the strength and pattern of observed relationships.

Despite the growing interest in linking knowledge to the general construct of PL, there is still a lack of empirical research that would analyse in detail the relationships between the separate sub-domain of Knowledge and Understanding and other PL domains. The present study adopts a perspective consistent with the CAPL-2 framework, in which knowledge is understood as a cognitive component of PL that may support engagement in physical activity (Anderson, 1982). This approach does not directly address ecological-dynamical interpretation emphasising performer–environment mutuality (Rudd et al., 2021), but focuses instead on individual cognitive aspects related to physical literacy.

Knowledge and Understanding in the Context of Physical Literacy

Cognitive psychology assumes that knowledge is a prerequisite for action (Strube & Wender, 1993), which suggests that knowledge is related to children’s daily physical activity behaviours. However, as Grimm (2006) notes, “knowledge is not equal”—that is not all forms of knowledge function in the same way or have the same practical implications. This author distinguishes between knowledge and understanding, emphasising that understanding is a special kind of knowledge—knowledge about causes. It is possible to understand something without knowing all the causes, and conversely, to know the causes without understanding their practical significance. According to Grimm (2006), understanding, as a deeper and more integrated form of knowledge, is crucial in the process of transforming information into action. Similarly, Kelp (2016) notes that increasing knowledge can deepen understanding, assuming that other factors remain constant. These perspectives provide a theoretical context; however, they are not directly tested in the present study.

Elgin (2009) emphasises that comprehension is not limited to remembering facts—it enables complex thinking: drawing conclusions, argumentation, and decision-making based on available information. From the perspective of PL, understanding may allow children to interpret bodily experience and recognise the health-related consequences of physical activity.

Within the CAPL-2 framework, the cognitive domain is operationalised as Knowledge and Understanding. In the present study, these components are treated as analytically distinct subdomains of the instrument. This distinction reflects the structure of the questionnaire rather than independently validated cognitive processes. Any interpretation regarding differences between these components should therefore be understood as theoretical and interpretative, not as empirical evidence of distinct cognitive constructs.

From a developmental perspective, the distinction between Knowledge and Understanding may be relevant in middle childhood. According to Piaget’s (1936/1966, 1964/1966) theory of cognitive development, children aged 7–11 years (concrete operations stage) begin to understand cause-and-effect relationships, but their conclusions are based mainly on direct experience. From about the age of 12 (formal operations stage), it becomes possible to think more abstractly. This developmental progression provides a theoretical basis for examining whether grade group moderates the relationships between cognitive components of physical literacy and other domains.

Building on this perspective, the present study examines whether differences in ‘Knowledge’ and ‘Understanding’ are reflected in their relationship with Physical Competence, Motivation and Confidence, and Daily Behaviour. In addition, grade group is tested as a potential moderator of these relationships.

By focusing on the Knowledge and Understanding components within the CAPL-2 framework, the study aims to provide a descriptive analysis of the relationship between the Knowledge and Understanding domain and other CAPL-2 domains within the operational framework of the instrument.

2. Materials and Methods

2.1. Participants

2030 children aged 8 to 12 from all over Poland took part in the survey, comprising 1060 boys (52.2%) and 970 girls (47.8%). A purposive sampling strategy was applied to ensure geographical and socio-demographic diversity of the participants. The research was carried out simultaneously by five independent research teams operating in different regions of Poland, which made it possible to include students from different school and social backgrounds. This approach was intended to maximise diversity rather than be representative of a full population. Although data were collected by five research teams, all participants followed the same national core curriculum for physical education, which ensured a high level of standardisation in educational content across settings. The mean age of the participants was 10.02 years (SD = 1.42). The criteria for inclusion in the study were the voluntary consent of parents or guardians and a health condition enabling participation in physical education classes. No additional exclusion criteria were applied prior to the analysis; however, due to the strict data collection schedule, participants who were absent were unable to complete all measures.

Table 1. Sample size (n) and missing data for total CAPL-2 score and individual domains.

CAPL-2 Measures	Total (n)	Boys	Girls	Missing Data
				n	%
Physical Competence	1943	1010	933	87	4.3
Daily Behaviour	1941	1009	932	89	4.4
Motivation and Confidence	2022	1056	966	8	0.0
Knowledge and Understanding	2030	1060	970	0	0.0
Overall CAPL-2 score	1903	984	919	127	6.3

summarises available sample sizes and percentages of missing data for total PL and domains. All available data were included in the analysis; cases with missing values were handled using available-case analysis, depending on the variables included in a given model.

Almost half of the pupils (n = 874; 43.1%) attended grades 2 or 3, following the core curriculum for general education at stage I (ISCED 1) (UNESCO Institute for Statistics, 2024). At this educational level, physical education classes are conducted by an early school education teacher. The remaining participants attended primary school at Stage II of their education, following the core curriculum for physical education in grade 4 (n = 384; 18.9%) or grades 5–6 (n = 772; 38.0%), respectively, all of which fall within ISCED level 2. At this stage, classes are conducted by a physical education teacher. Participants were grouped according to grade level (grades 2–3, grade 4, and grades 5–6), reflecting different stages of the physical education curriculum. The study was approved by the Senate Research Ethics Committee of the Józef Piłsudski University of Physical Education in Warsaw (SKE 01-33/2023).

2.2. Measures and Procedures

To assess the level of physical literacy, the standardised tool Canadian Assessment of Physical Literacy—CAPL-2 (Longmuir et al., 2018) was used. CAPL-2 measures PL across four domains: Daily Behaviour (DB); Physical Competence (PC); Knowledge and Understanding (KU); and Motivation and Confidence (MC). Physical Competence is assessed using the Progressive Aerobic Cardiovascular Endurance Run (PACER) (Scott et al., 2013); the plank isometric hold (Boyer et al., 2013); and the Canadian Agility and Movement Skill Assessment (CAMSA) (Longmuir et al., 2017).

The number of steps was recorded using wrist-worn activity trackers (Xiaomi Band 8), worn on the wrist for seven consecutive days. Processing of step count data was carried out in accordance with the standard procedures used in the CAPL-2 protocol (HALO, 2017). A day was considered valid if the number of steps fell within the range of 1000 to 30,000 and the device was worn for at least 10 h. Inclusion of data in the analysis required at least four days meeting the validity criteria, including at least one weekend day. The wristbands were removed during activities involving contact with water. In the case of missing or unreliable data (e.g., device failure, illness, or non-compliance with the procedure), such data were excluded from the analysis. The average daily step count was calculated based on all days meeting the validity criteria. Given the large sample size and the field-based nature of the study, the use of commercially available devices enabled data collection in a feasible and standardised manner across all participants.

The Polish adaptation of the full CAPL-2 protocol, including all four domains, has previously been translated, culturally adapted, and validated (Firek et al., 2025). In that study, the Motivation and Confidence domain demonstrated high internal consistency (Cronbach’s α = 0.876) and test–retest reliability (ICC = 0.905), while the Knowledge and Understanding domain showed acceptable internal consistency (Cronbach’s α = 0.70) and substantial reliability (ICC = 0.714). The construct validity of the overall CAPL-2 model was confirmed using CFA, indicating a good model fit (χ²(37) = 72.7, p < 0.001, χ²/df = 1.95, CFI = 0.982, TLI = 0.973, RMSEA = 0.031, SRMR = 0.024). These findings support the validity and reliability of the Polish version of CAPL-2 in children aged 8–12. The instrument was administered in accordance with the standardised CAPL-2 protocol. Motivation and Confidence were assessed using the CAPL-2 questionnaire, which measures intrinsic motivation, physical activity competence, predilection, and adequacy. The Knowledge and Understanding questionnaire consists of two subdomains: Knowledge (4 items) and Understanding (6 items), which were also considered separately in this analysis. Younger children in grades 2 and 3 (ISCED 1) and pupils with special educational needs were able to benefit from the support of an assistant or support teacher. Each child completed the questionnaire at their own pace, with no time limit. Participants could obtain a maximum of 10 points for correct answers (Knowledge and Understanding domain). The domains of Motivation and Confidence, Physical Competence, and Daily Behavior are each scored on a 30-point scale.

2.3. Data Analysis

Statistical analyses were conducted using IBM SPSS Statistics 29 (IBM Corp., Armonk, NY, USA, 2024). In the present study, age was operationalised as a categorical variable reflecting grade group, corresponding to different stages of the physical education curriculum (grades 2–3, grade 4, grade 5–6), rather than as a continuous variable. To compare the results of the Knowledge and Understanding Scale (KU) between groups of students, distinguished on the basis of the core curriculum for physical education, a non-parametric Kruskal–Wallis test was used due to the failure to meet the assumption of normality of the distribution and the lack of equivalence of the groups. The relationships between CAPL-2 domains were analysed using Spearman’s rho correlation coefficient. To analyse the moderation effect, a General Linear Model (OLS) with standardised coefficients and robust standard errors was used, accounting for the heteroscedasticity of the residuals. Moderation analysis was performed using JAMOVI 2.6.44 (The Jamovi project, 2024). Grade group (three categories: grades 2–3, grade 4, and grades 5–6) was included as a categorical moderator variable in the regression models. For all analyses, the level of statistical significance α = 0.05 was assumed.

3. Results

The descriptive statistics for the four CAPL-2 domains and the total scores are presented in

Table 2. Descriptive statistics for the CAPL-2 domains and overall score (N = 2030).

CAPL-2 Domains	M	SD	Min	Max	Skewness	Kurtosis
Physical Competence (max. 30 points)	18.2	7.73	3.0	30.0	−0.167	−0.607
Boys	18.6	6.03	3.1	30.0	−0.173	−0.762
Girls	17.7	5.36	3.0	30.0	−0.226	−0.423
Daily Behaviour (max. 30 points)	14.5	7.01	0.0	30.0	0.219	−0.667
Boys	15.2	7.42	0.0	30.0	0.093	−0.836
Girls	13.9	6.65	0.0	30.0	0.337	−0.394
Knowledge and Understanding (max. 10 points)	5.8	2.37	0.0	10.0	−0.373	−0.676
Boys	5.6	2.44	0.0	10.0	−0.310	−0.846
Girls	6.0	2.26	0.0	10.0	−0.422	−0.466
Motivation and Confidence (max. 30 points)	25.12	4.12	7.8	30.0	−0.895	0.329
Boys	25.0	4.2	7.8	30.0	−0.899	0.462
Girls	25.2	4.09	9.6	30.0	−0.891	0.179
Overall CAPL-2 Score (max. 100 points)	63.88	12.56	24.9	99.6	−0.017	−0.346
Boys	64.7	13.18	26.8	99.6	−0.021	−0.451
Girls	63.0	11.80	24.9	93.7	−0.066	−0.268

. The mean overall CAPL-2 score was 63.88 points (SD = 12.56) in the range of 24.9 to 99.6, indicating an overall mid-range level of physical literacy within the sample. Among the individual domains, the highest average was recorded in Motivation and Confidence, suggesting that the children perceived themselves as motivated and confident in the context of physical activity. At the same time, this domain had the greatest negative skewness (Sk = −0.895), indicating a tendency for the results to be clustered at the upper end of the scale. In the case of the Physical Competence domain, the average score was 18.2 points (SD = 7.73), with a relatively wide spread (3–30 points), which may indicate a large variation in the level of physical fitness in the studied sample. The average score in the Daily Behavior domain was 14.5 points, suggesting relatively low daily physical activity within the scoring range. In the knowledge test, the children received an average of 5.83 points. The results of this domain will be analysed in detail later in the paper.

Independent-sample t-tests with Welch correction were conducted to examine gender differences across the analysed physical literacy domains. Statistically significant differences were observed in all variables. Boys demonstrated higher scores in Daily Behaviour (M = 15.19 vs. 13.85; t(1937.29) = 4.19; p < 0.001; d = 0.19), Physical Competence (M = 18.56 vs. 17.71; t(1938.14) = 3.32; p = 0.001; d = 0.15), and Total CAPL score (M = 64.70 vs. 63.00; t(1897.61) = 2.96; p = 0.003; d = 0.14). Girls scored higher in Knowledge and Understanding (M = 6.04 vs. 5.64; t(2027.81) = −3.83; p < 0.001; d = 0.17). Effect sizes were small across all comparisons (Cohen’s d = 0.14–0.19).

3.1. Differences in Knowledge and Understanding Across Grade Group

Students were divided into three age groups depending on the core curriculum for physical education. Descriptive statistics (mean, standard deviation, minimum and maximum values, skewness and kurtosis) are presented in

Table 3. Basic descriptive statistics of the results of the CAPL-2 Knowledge and Understanding Test (N = 2030).

Grade	n	M	SD	Min	Max	Skewness	Kurtosis
Grades 2 to 3 (ISCED 1) (max. 10 points)	874	5.02	2.32	0	10	−0.035	−0.831
Boys	448	4.78	2.34	0	10	0.042	−0.913
Girls	426	5.28	2.26	0	10	−0.102	−0.708
Grade 4 (ISCED 2) (max. 10 points)	384	5.93	2.28	0	10	−0.569	−0.270
Boys	190	5.67	2.36	0	10	−0.480	−0.573
Girls	194	6.18	1.19	0	10	−0.648	0.130
Grades 5–6 (ISCED 2) (max 10 points)	772	6.69	2.14	0	10	−0.730	0.032
Boys	422	6.53	2.27	0	10	−0.710	−0.181
Girls	350	6.89	1.96	0	10	−0.680	0.167

, along with results stratified by sex.

A comparative analysis by gender across grade levels revealed statistically significant differences in the Knowledge and Understanding domain. In grades 2–3, girls scored higher than boys (t(872) = −3.17, p = 0.002, d = 0.21). A similar pattern was observed in grade 4 (t(382) = −2.16, p = 0.031, d = 0.22) and in grades 5–6 (t(768.53) = −2.35, p = 0.019, d = 0.17). Across all analysed groups, girls consistently achieved higher scores than boys; however, the effect sizes were small, indicating limited practical significance of the observed differences.

The Kruskal–Wallis test showed statistically significant differences in Knowledge and Understanding scores depending on the educational stage (

Table 4. Comparison of Knowledge and Understanding test results between grade groups—Kruskal–Wallis test (N = 2030).

DV	Grades 2–3 (n = 874)		Grade 4 (n = 384)		Grades 5–6 (n = 772)		H	df	p	η2
	Mdn	IQR	Mdn	IQR	Mdn	IQR
K	2.0	2.0	2.0	2.0	2.0	1.0	84.941	2	<0.001	0.041
U	3.0	2.0	4.0	2.0	5.0	3.0	193.288	2	<0.001	0.094
KU	5.0	4.0	6.0	3.0	7.0	2.0	213.106	2	<0.001	0.104

Abbreviations: K—subdomain Knowledge; U—subdomain Understanding; KU—Knowledge and Understanding domain.

). The median score increased with the level of education. Post hoc comparisons (adjusted for Bonferroni) confirmed significant differences between all groups. Effect sizes suggest a small-to-moderate effect for the Understanding subdomain and the total Knowledge and Understanding score and small effects for the Knowledge subdomain.

3.2. Correlations Between Knowledge and Understanding and Other CAPL-2 Domains

Spearman’s correlation coefficients between Knowledge and Understanding scores and other CAPL-2 domains (Daily Behaviour, Motivation and Confidence, Physical Competence) are presented in

Table 5. Correlations between Knowledge and Understanding test results and other CAPL-2 domains (N = 2030).

CAPL-2 Domains		Knowledge and Understanding	Knowledge	Understanding
Daily Behavior	Grades 2–3	ρ = 0.079 *	0.079 *	0.058
	Grade 4	ρ = 0.036	0.033	0.008
	Grades 5–6	ρ = 0.044	0.001	0.057
	Total	ρ = 0.049 *	0.036	0.040
Motivation and Confidence	Grades 2–3	ρ = 0.138 ***	0.059	0.149 ***
	Grade 4	ρ = 0.139 *	0.099	0.132 *
	Grades 5–6	ρ = 0.139 ***	0.014	0.172 ***
	Total	ρ = 0.105 ***	0.032	0.121 ***
Physical Competence	Grades 2–3	ρ = 0.212 ***	0.170 ***	0.183 ***
	Grade 4	ρ = 0.209 ***	0.145 *	0.192 ***
	Grades 5–6	ρ = 0.184 ***	0.042	0.215 ***
	Total	ρ = 0.280 ***	0.169 ***	0.273 ***

* p < 0.05; *** p < 0.001 (two-tailed).

. Overall, the observed correlations were statistically significant in several cases; however, their magnitude was generally small, indicating limited practical significance. This is particularly important given the large sample size, which increases the likelihood of detecting statistically significant but weak associations. No consistent or practically meaningful relationships were observed between Knowledge and Understanding and Daily Behaviour across grade groups (ρ ranging from 0.001 to 0.079). Although some correlations reached statistical significance, their effect sizes were negligible. For Motivation and Confidence, statistically significant correlations were found across all grade groups; however, these associations were weak (ρ ≈ 0.10–0.14 for the total score), indicating limited explanatory value. Importantly, this pattern was more apparent for the Understanding subdomain, while Knowledge alone showed minimal or non-significant associations. A similar pattern was observed for Physical Competence. Although correlations with Knowledge and Understanding were statistically significant, their magnitude remained small to moderate (ρ ≈ 0.17–0.28), suggesting that knowledge-related variables account for only a limited proportion of variance in this domain. Again, the Understanding subdomain showed relatively stronger (although still weak) associations than Knowledge, particularly in older students.

3.3. Grade Group as Moderators of the Relationship Between Knowledge and Understanding and Physical Competence, Daily Behaviour, and Motivation and Confidence

The results of three separate moderation analyses are presented below, in which the moderating variable was the students’ grade group. The effect of ‘Knowledge’ and ‘Understanding’, as well as their interactions with grade group, were examined for three dependent variables: Physical Competence, Motivation and Confidence, and Daily Behaviour. Sex was additionally included in all models as a covariate to control for potential differences between boys and girls. The analysis was carried out in relation to the assumptions of Piaget’s theory of cognitive development, which suggest that with age, children process information more effectively, better understand cause-and-effect relationships, and are increasingly able to apply knowledge in context. However, in the present study, these assumptions were tested empirically within the structure of the CAPL-2 domains.

For Physical Competence (

Table 6. Moderation analysis: Interaction of knowledge types and grade group in predicting the Physical Competence domain.

Predictor	B	SE	β	t	p
Intercept	14.16	0.45	—	31.23	<0.001
Knowledge	0.65	0.18	0.12	3.70	<0.001
Understanding	0.51	0.11	0.16	4.49	<0.001
Grade group
Grade 4 vs. 2–3	1.45	0.89	0.26	1.62	0.105
Grades 5–6 vs. 2–3	2.75	0.80	0.43	3.44	<0.001
Sex
girls–boys	−1.03	0.25	−0.18	−4.19	<0.001
Knowledge × Grade group
Knowledge × Grade 4	0.15	0.33	−0.03	−0.47	0.639
Knowledge × Grades 5–6	−0.64	0.27	−0.12	−2.35	0.019
Understanding × Grade group
Understanding × Grade 4	0.10	0.21	0.03	0.48	0.631
Understanding × Grades 5–6	0.17	0.12	0.08	1.62	0.105

Note. R² = 0.128, Adjusted R² = 0.124. Reference group: grades 2–3; sex coded as 1 = boys, 2 = girls.

), the model explained more than 12% of the variance. Both “Knowledge” and “Understanding” were statistically significant predictors of Physical Competence, with Understanding showing a slightly higher standardised coefficient. A significant interaction was observed between ‘Knowledge’ and age for students in grades 5–6 (β = −0.12, p = 0.019), indicating that the association between ‘Knowledge’ and Physical Competence decreased with age. No other interaction effects were statistically significant. Additionally, sex was a significant predictor, with boys demonstrating higher levels of physical competence than girls.

For motivation and confidence (

Table 7. Moderation analysis: Interaction of knowledge types and grade group in predicting the Motivation and Confidence domain.

Predictor	B	SE	β	t	p
Intercept	24.29	0.34	—	72.20	<0.001
Knowledge	0.10	0.13	0.03	0.73	0.465
Understanding	0.34	0.08	0.15	4.05	<0.001
Grade group
Grade 4 vs. 2–3	−0.57	0.67	−0.14	−0.85	0.393
Grades 5–6 vs. 2–3	−1.67	0.59	−0.36	−2.85	0.004
Sex	0.01	0.18	0.00	0.03	0.975
Girls–boys
Knowledge × Grade group
Knowledge × Grade 4	0.15	0.24	0.04	0.64	0.525
Knowledge × Grades 5–6	−0.12	0.20	−0.03	−0.61	0.542
Understanding × Grade group
Understanding × Grade 4	−0.08	0.15	−0.04	−0.54	0.588
Understanding × Grades 5–6	0.11	0.12	0.05	0.92	0.358

Note. R² = 0.039, Adjusted R² = 0.034. Reference group: grades 2–3; sex coded as 1 = boys, 2 = girls.

), the model was statistically significant; however, it explained only a very small proportion of the variance (3.9%). This level of explanatory power is negligible and does not support meaningful interpretation, although ‘Understanding’ reached a statistically significant relationship. ‘Knowledge’ was not associated with this outcome, and no interaction effect with age group was observed. Sex did not contribute significantly to the model. Taken together, these results do not provide meaningful evidence for a relationship between knowledge-related variables and Motivation and Confidence.

The model for daily physical activity behaviours (

Table 8. Moderation analysis: Interaction of knowledge types and grade group in predicting the Daily Behaviour domain.

Predictor	B	SE	β	t	p
Intercept	13.77	0.60	—	22.95	<0.001
Knowledge	0.39	0.23	0.12	1.71	0.088
Understanding	0.23	0.15	0.06	1.57	<0.116
Grade group
Grade 4 vs. 2–3	1.23	1.17	0.04	1.06	0.291
Grades 5–6 vs. 2–3	0.23	1.06	−0.09	0.22	0.83
Sex
girls–boys	−1.46	0.32	−0.21	−4.51	<0.001
Knowledge × Grade group
Knowledge × Grade 4	−0.07	0.43	−0.01	−0.16	0.873
Knowledge × Grades 5–6	−0.45	0.36	−0.07	−1.27	0.205
Understanding × Grade group
Understanding × Grade 4	−0.22	0.27	0.05	−0.81	0.420
Understanding × Grades 5–6	0.02	0.22	0.00	0.08	0.934

Note. R² = 0.016, Adjusted R² = 0.011. Reference group: grades 2–3; sex coded as 1 = boys, 2 = girls.

) explained only a negligible proportion of the variance (R² = 0.016). Neither Knowledge nor Understanding was a significant predictor of Daily Behaviour, and no interaction effect with age group was observed. Importantly, sex was a significant predictor, with boys reporting higher levels of daily physical activity. Overall, the lack of association between knowledge-related variables and Daily Behaviour is a meaningful finding, suggesting that cognitive components measured by CAPL-2 may not directly translate into everyday physical activity behaviour in this age group.

4. Discussion

The results of the study indicate that the level of Knowledge and Understanding of children in the field of physical activity increases with age and educational stage. The highest results were achieved by students in grades 5 and 6, which may be tentatively interpreted in light of Piaget’s (1936/1966, 1964/1966) theory, although the cross-sectional design does not allow for developmental conclusions. According to this theory, children move from the concrete operational stage (about 7–11 years of age) to the formal operational stage (from about 12 years of age), which is associated with an increase ability to understand relationships and interpret information in a more integrated way.

Furthermore, the differences observed between boys and girls are partly consistent with the findings of previous studies using the CAPL-2 instrument. As in studies conducted in China (Li et al., 2020), Spain (Mendoza Muñoz et al., 2024), and Pakistan (Hadier et al., 2024), boys achieved higher scores in terms of physical fitness and everyday behaviours. At the same time, the findings regarding the cognitive domain are less clear-cut. In some studies, no significant gender differences were observed in terms of knowledge and understanding (Gilic et al., 2022; Hadier et al., 2024), which is consistent with the findings of the present study in which girls achieved slightly higher scores in this domain. Such differences may stem not only from developmental factors but also from socio-cultural conditions. In many contexts, boys are more frequently encouraged to participate in sporting activities, which promotes the development of physical fitness and everyday physical behaviour, whilst girls may achieve relatively higher scores in the cognitive domain. It should be emphasised, however, that the observed differences were small, suggesting their limited practical significance.

The results indicate that the relationship between the cognitive domain and daily behaviour is more complex and ambiguous. Contrary to expectations, the analysis did not show a meaningful relationship between the Knowledge and Understanding domain and the level of Daily Behaviour. The observed correlations were negligible and inconsistent across grade groups. This pattern is consistent with previous CAPL-based findings, in which the relationship between K&U and Daily Behaviour was also weak. In the original refinement study of the CAPL model (Gunnell et al., 2018a), the association between these two domains was small (r = 0.102), while in the Gunnell et al. (2018b) study, daily step count was not significantly correlated with any CAPL domain. Similar findings suggest that the pathway from knowledge to everyday physical activity behaviour may be indirect and influenced by contextual or environmental factors rather than immediate cognitive awareness alone (Gunnell et al., 2018a; Hadier et al., 2024). A possible explanation for this lack of association relates to the role of environmental factors: children in the age group studied do not yet have autonomy in organizing their own time. Their physical activity depends to a large extent on the decisions of parents and teachers (Bandura, 2001; Biddle et al., 2004). As a result, even a higher level of knowledge and understanding may not translate directly into behaviour. Importantly, these results should be interpreted cautiously and do not necessarily contradict the theoretical assumptions of PL proposed by Whitehead (2001), which assumes mutual reinforcement of PL domains. Rather, it may indicate that the relationship between cognitive and behavioural components is indirect and mediated by contextual factors.

Further analyses show that a higher level of knowledge and understanding is significantly associated with a higher level of Physical Competence. It is possible that children who are more physically active have more opportunities to gain knowledge in practice but also that having knowledge about effective and safe exercise is conducive to the development of physical fitness and motor skills. However, these relationships were generally weak to moderate in strength. This suggests that cognitive factors explain only a limited proportion of variance in physical competencies. Notably, the Understanding subdomain demonstrated relatively stronger (although still weak) associations with Physical Competence than with Knowledge. This observation is broadly in line with earlier CAPL research, in which the cognitive domain showed more consistent links with Physical Competence than with other PL domains (Gunnell et al., 2018a, 2018b). This interpretation is also supported by cross-validation studies. In the Greek sample (Dania et al., 2020), knowledge scores appeared sensitive to pedagogical context, whereas physical competence showed clearer age- and sex-related differentiation. In the Pakistani sample (Hadier et al., 2024) the Knowledge and Understanding domain was weakly associated with the remaining PL domains, while Physical Competence showed a more robust position in the overall model. This may reflect differences in how these subdomains are operationalised within the instrument. However, given the cross-sectional design of the study, these relationships should not be interpreted as causal.

A similar pattern was observed for Motivation and Confidence. Although statistically significant associations were found, their magnitude was small. Moreover, the regression model explained only a negligible proportion of variance, which does not support the meaningful interpretation of predictive relationships. Importantly, Knowledge was not significantly associated with Motivation and Confidence, and no interaction effects with grade group were observed. Earlier Canadian analyses also reported either very weak or non-significant associations between K&U and M&C (Gunnell et al., 2018a, 2018b). Although Understanding showed a statistically significant relationship, its explanatory value remains limited. Therefore, these results should be interpreted with caution and do not provide strong evidence for a meaningful relationship between cognitive variables and motivational outcomes.

The moderation analysis showed that the relationship between Knowledge and Physical Competence decreased in the older grade group, while no consistent interaction effects were observed for Understanding. These findings provide only partial support for developmental interpretations based on Piaget’s theory and should, therefore, be treated as exploratory.

Overall, the findings suggest that although Knowledge and Understanding are related to selected domains of PL, these relationships are generally weak and context-dependent. The results do not, therefore, support strong conclusions about the role of cognitive factors as key predictors of physical literacy outcomes.

It should be emphasised that the analytical approach adopted in this study, which involves analysing the relationships between the individual domains of CAPL-2, constitutes a certain methodological simplification. Whilst it enables the empirical examination of specific relationships, it does not fully capture the integrated and dynamic nature of physical literacy as conceived by Whitehead (2001). Consequently, the results obtained should be interpreted with this limitation in mind. This reflects a broader limitation of any instruments that operationalises physical literacy through distinct measurable domains, as such tools inevitably simplify a holistic construct.

Despite the large and diverse sample and the use of a validated measurement tool, this study has some limitations that should be taken into account when interpreting the results. First, the study was cross-sectional, which makes it impossible to infer causal relationships between variables. Second, although CAPL-2 is a widely used instrument, the Knowledge and Understanding domain may capture cognitive aspects in a limited way, focusing primarily on questionnaire-based assessment. This limitation is worth noting that in light of previous validation studies, which suggested that the cognitive domain of CAPL-2 may be psychometrically less robust than other components, particularly across different cultural and linguistic context (Gunnell et al., 2018a, 2018b; Knisel et al., 2024; Pastor-Cisneros et al., 2022). Third, the children’s answers were self-reported, which may be associated with the risk of cognitive biases or a tendency to give socially desirable answers. Finally, although the sample was large and included students from different parts of Poland, the influence of local factors such as the teachers’ working style, sports infrastructure, or cultural conditions, which could have influenced the results, cannot be completely ruled out. Future research should consider longitudinal design and more comprehensive approaches to assessing the cognitive domain of PL, as well as analytical strategies that account for contextual and hierarchical factors. Importantly, the distinction between Knowledge and Understanding in this study reflects an operational separation within the CAPL-2 instrument and should not be interpreted as empirical evidence of distinct cognitive processes.

5. Conclusions

Given the descriptive and exploratory nature of the study, within the CAPL-2 framework, the Understanding subdomain showed relatively stronger (yet still weak) associations with selected outcomes. These findings should be interpreted as descriptive tendencies within the measurement model rather than as evidence of substantively different cognitive roles.

From the perspective of Piaget’s theory of cognitive development, this may suggest that physical education could be adapted to the stage of child development: younger children (7–11 years old), functioning at the stage of concrete operations, better assimilate simple cause-and-effect relationships, while older children (12+) are capable of reflecting, abstracting, and making informed health decisions. Although the observed relationships were weak, the findings may tentatively inform educational considerations, particularly in relation to supporting students’ understanding and application of knowledge in physical education contexts. It is important that educational goals refer to higher levels of cognitive taxonomy such as analysis, application, and explanation, and not just memorisation or definition.

In addition, the child’s environment—both school and family—may play a key role in transforming knowledge into lasting habits. Younger children, deprived of autonomy in time management, may need the support of adults to implement the acquired knowledge in everyday life. At school, this may involve a greater balance between motor activity and opportunities for cognitive engagement and reflection. A physical education teacher should not only be an organiser of activities but also a guide in supporting students’ understanding of health and bodily awareness.

In light of the above, these findings may have implications for the further development of educational programs and measurement tools. However, given the limited strength of the observed relationships, any such implications should be treated as tentative.

In summary, promoting physical activity among pupils may require not only increasing their knowledge but also supporting their understanding and application of that knowledge in context. Every physical education lesson can include elements that engage pupils cognitively—encouraging them to think, understand, and act. Creating space for dialogue and reflection can foster more active and conscious participation in physical education classes.

From a practical perspective, these findings also suggest that physical education curricula and core curricula should integrate cognitive engagement with opportunities for active participation, whilst recognising that the relationship between knowledge and behaviour is both indirect and context-dependent.