Physical literacy is associated with children's adherence to physical activity guidelines during COVID-19

ABSTRACT

Background: Despite the benefits of physical activity, children’s compliance with physical activity recommendations is very low and this situation worsened during the COVID-19 pandemic. Physical literacy has been theorised as a factor that can enhance and sustain children’s engagement with physical activity. This study aimed to investigate whether physical literacy is associated with children’s adherence to the physical activity guidelines during COVID-19.

Methods: Data for this cross-sectional study were collected from 648 Australian children (60% boys) aged 7–12 years (M = 10 years; SD = 1.7) and their parents/carers. Recruitment was conducted through social media and panel sampling. Self-perceived physical literacy was measured using the pictorial Physical Literacy in Children Questionnaire. Children’s adherence to physical activity guidelines was collected using a two-item questionnaire asking parents/guardians about the number of days their children were active for at least 60 min in a typical week before and during COVID-19. The association between physical literacy (predictor) and physical activity (binary outcome: meeting/not meeting guidelines before/during COVID-19) was tested with logistic regressions while adjusting for relevant confounders.

Results: Most parents spoke English at home (91%), over half had a university education (62%) and three quarters had paid employment (75%). Children’s physical literacy scores ranged from 32–120 (M = 94; SD = 16). Physical literacy overall score was associated with significantly higher odds of meeting the physical activity recommendations during COVID-19 (OR = 1.03, 95% CI [1.02; 1.05]). Each of the physical literacy subdomains (i.e. physical, psychological, social, cognitive) was also significantly associated with children’s adherence to the physical activity guidelines (ORs = 1.07–1.22).

Conclusions: These findings indicate that physical literacy is a predictor of physical activity and may act as a protective factor of such behaviour during limiting events, such as the COVID-19 pandemic. Further research could confirm this relationship using objective measures of physical activity and exploring whether interventions designed to improve physical literacy can benefit physical activity.

KEYWORDS:

• Physical literacy
• physical activity
• PL-C quest
• self-report
• pictorial scale

Background

Physical activity in childhood is associated with favourable health-related outcomes (Poitras et al. 2016), increased wellbeing (Biddle et al. 2019), and improved cognitive functions (Alvarez-Bueno et al. 2017; Pesce et al. 2023). Based on the evidence of benefits associated with an active lifestyle (Chaput et al. 2020), the World Health Organization recommends that children aged 5–12 years perform at least an average of 60 min per day of moderate- to vigorous-intensity physical activity across a week (Bull et al. 2020).

Despite this, physical inactivity is prevalent among children globally, with less than one-third of 5–17-year-old children and adolescents meeting the physical activity guidelines (Aubert et al. 2022). According to the Global Matrix 4.0, approximately 30% of children and adolescents from 57 countries met the physical activity guidelines and only 20–26% in Australia (Aubert et al. 2022). Furthermore, there has been a decline in physical activity levels among children in Europe, North America, and Asia as a result of the COVID-19 pandemic (Cachón-Zagalaz et al. 2021; Jia et al. 2021; Tulchin-Francis et al. 2021). Similarly, a significant reduction in children’s likelihood to meet the physical activity guidelines during COVID-19 was reported in Australia (Arundell et al. 2022). Hence, promoting children’s physical activity is a relevant public health priority, and understanding what factors may help to prevent its decline in children is important.

Physical literacy has been theorised to be the foundation of lifetime physical activity participation (Cairney et al. 2019; Edwards et al. 2017; Keegan et al. 2013). Physical literacy refers to people's potential to participate and learn from their physical interaction with the environment, whereby people become more competent and motivated to participate in physical activity in the long term (Keegan, Dudley, and Barnett 2019). Different definitions of physical literacy have evolved to reflect cultural and political contexts. The International Physical Literacy Association (IPLA) defined physical literacy as ‘the motivation, confidence, physical competence, knowledge and understanding to value and take responsibility for engaging in physical activities for life’ (International Physical Literacy Association 2017). The definition in use in Canada, was drawn from the IPLA definition (Keegan, Dudley, and Barnett 2019; ParticipAtion et al. 2015). In recent years, the Australian Physical Literacy Framework suggested physical literacy was a holistic concept, defined as the ‘lifelong holistic learning acquired and applied in movement and physical contexts’, comprising four domains (physical, psychological, cognitive, and social) and 30 elements (12 elements in the physical domain, seven in the psychological and cognitive domains, and four in the social domain) (Australian Sports Commission 2019).

To understand the relationship between children’s physical activity and physical literacy, we need to be able to assess physical literacy. According to a recent systematic review (Shearer et al. 2021), only a few tools based on the Canadian interpretation were explicitly labelled as physical literacy assessments – e.g. Canadian Assessment of Physical Literacy (CAPL-2) (Longmuir et al. 2018; Tremblay et al. 2018); Physical Literacy Assessment for Youth (PLAY) Tools (Caldwell et al. 2020). CAPL-2 measures the physical literacy of children (aged 8–12 years) mainly using objective assessment of their physical competence, such as health-related fitness and movement skill measures (Longmuir et al. 2018). The PLAY Tools, assess children’s physical literacy level in a few ways, e.g. by objective assessment of children’s motor competence (PLAYfun) and self-evaluation (PLAYself) (Caldwell et al. 2020). While some items in the CAPL-2 and the PLAY Tools align with the Australian definition, almost half of the elements in the Australian Physical Literacy Framework are not covered by either tool. Therefore, the Physical Literacy in Children Questionnaire (PL-C Quest) was developed to produce comprehensive evidence of children’s physical literacy, in line with the Australian Physical Literacy Framework (Barnett et al. 2020).

The primary-school age (5–12 years) is seen as a crucial stage for the development of physical literacy attributes necessary for long-term physical activity participation, health, and wellbeing (Belanger et al. 2018). Recent studies in school-aged children provide supporting evidence of the longitudinal association between physical activity and physical literacy, as well as the link between physical literacy and health indicators such as lower body fat percentage and improved cardiovascular fitness (Caldwell et al. 2020; Nezondet et al. 2023). Despite the emerging evidence on the correlates of physical literacy, none of the available studies have explored how physical literacy measured using the PL-C Quest – i.e. the only instrument currently available to align with the holistic Australian Physical Literacy Framework – and children’s adherence to physical activity guidelines. In addition, although some studies documented declines in children’s physical activity and physical literacy during COVID-19 (Houser et al. 2022), there is currently limited evidence on the role that physical literacy can play as a protective factor from physical activity decline. For these reasons, the primary aim of our study was to investigate whether there is an association between physical literacy (measured using the PL-C Quest) and children’s adherence to physical activity guidelines before/during COVID-19.

Methods

Design and procedure

Data for this cross-sectional study were collected via social media (August to November 2020) and an online panel service via Qualtrics (November 2020 to January 2021). The study was approved by the Faculty of Health Human Ethics Advisory Group in Deakin University (Reference number: HEAG-H 82_2019). Parents consented online, provided information on demographic characteristics and their child’s physical activity, and children completed the physical literacy survey (the PL-C Quest). Demographic information included the child’s date of birth, sex, language at home, parental residing state, parental education, and parental employment.

The physical literacy data were used in a previous study to assess the validity of the instrument used to measure physical literacy (Barnett et al. 2022). Overall, 669 children (aged 7–12 years) living in Australia were included in the validity study (Barnett et al. 2022). The current study builds upon Barnett et al. (2022) by specifically exploring the relationship between perceived physical literacy and physical activity. A total of 648 children had complete physical literacy and physical activity data and were included in the present study.

Measurement tools

Physical activity

Physical activity levels were measured by a proxy-reported two-item questionnaire (pre- and during COVID-19) completed by parents/carers. Proxy report is advantageous for assessing physical activity in 7–12-year-olds, as it may help to reduce age-related subject bias and recall error within children’s self-reported surveys (Hidding et al. 2018; Telford et al. 2004). Physical activity was defined as any activity that increases heart rate and makes children get out of breath some of the time. Examples were provided before the item administration. Specifically, parents were asked ‘over the past 7 days on how many days was your child physically active for a total of at least 60 min per day?’ (Item 1) and ‘over a typical or usual week (i.e. before COVID-19 during a typical school term) on how many days was your child physically active for a total of at least 60 min per day?’ (Item 2). The original item (Item 1) has high criterion validity (i.e. Rho = 0.49, 95% Confidence Intervals (CI) (0.32–0.53) and 88.2% agreement when assessed with accelerometer data for average 7-days moderate-to-vigorous physical activity) (Ridgers et al. 2012).

Physical literacy

Children's perceived physical literacy was assessed by the PL-C Quest, a self-report pictorial questionnaire (Barnett et al. 2020). In line with the Australian Physical Literacy Framework, the tool includes 30 items under four domains (physical, psychological, social, and cognitive) (Barnett et al. 2020). The score for each individual physical literacy item ranges from one (i.e. low proficiency) to four (i.e. high proficiency). An example of the item ‘Moving with equipment’ (physical domain) and the associated scoring is illustrated in Figure 1.

Figure 1. Example of Physical Literacy in Children Questionnaire item relating to ‘Moving with Equipment’, within the physical domain. The scoring assigned to the participant’s response range between 1 (far right option indicating low proficiency at a skill) to 4 (far left option indicating high proficiency at a skill). Note that scores are not visible during administration.

The sum of each item scores under a domain constitutes the domain score and the sum of domain scores is the overall score. The overall physical literacy score ranges from 30 to 120. A recent review of validity of child and youth physical literacy instruments showed the PL-C Quest as one of the instruments with more validity evidence (Barnett et al. 2023). Specifically, the PL-C Quest has evidence of test content and response processes in a range of Australian children including Indigenous (Barnett et al. 2020; De Silva et al. 2023). Desirable test-retest reliability over an average of 15.8 days (Intraclass Correlation Coefficient = 0.83 for overall physical literacy) and high internal consistency in Australian children (Cronbach alpha = 0.94) have also been reported (Barnett et al. 2022). Further, the construct validity of the PL-C Quest shows a good ability to measure children’s physical literacy, with most items loading on their respective domains close or above the standard of λ = 0.45, apart from two items (Moving with Equipment and Rules) that loaded at λ = 0.40 (Barnett et al. 2022). Additionally, the overall model showed excellent fit (SRMR  = 0.043, CFI  = 0.92, TLI  = 0.90, RMSEA  = 0.04) (Barnett et al. 2022).

Data cleaning and management

Data were exported from Qualtrics into an Excel spreadsheet. Microsoft Excel 365 and Stata (IC 17.0, StataCorp, Texas USA) were used for data cleaning and management. Overall, 648 children had complete physical literacy and physical activity data.

For demographic details, child age was calculated as decimal years. Only boys and girls were included in sex-based comparisons (five children were excluded as gender was stated as ‘other’ or ‘prefer not to answer/say’). Other variables were treated as follows: language at home (English versus non-English), parental education [tertiary education versus the rest (technical/trade certificate/apprenticeship, completed high school and some high school or less)], and parental employment [paid employment versus the rest (unpaid employment, retired, unemployed, home-duties full time, student and other)].

A dichotomous variable was created to identify children who were active for at least 60 min each day of the week (1 = meeting the physical activity guidelines) and those who were not (0 = not meeting the physical activity guidelines). Each child had two variables to indicate their adherence to the physical activity guideline, before and during COVID-19. The physical literacy score was treated as a continuous variable.

Data analysis

Descriptive statistics were generated for demographic information, adherence to physical activity recommendations, and physical literacy. Preliminary analyses were conducted to assess differences in adherence to physical activity recommendations and physical literacy based on children’s age, sex, parental education (tertiary vs not), parental employment (paid employment vs other), and language spoken at home (English vs other), using t-tests or χ² test depending on the nature of the data (continuous/categorical).

Separate logistic regressions were used to investigate whether physical literacy overall and subdomain scores (predictor) are independently associated with meeting the physical activity guidelines (i.e. before or during COVID-19, outcomes), while adjusting for children’s sex, age and any potential cofounders found to be associated with the outcomes of interest (i.e. adherence to physical activity recommendations) at p < 0.10 in the preliminary analyses. Additional independent logistic regressions between physical literacy scores (predictor) and physical activity during COVID-19 (outcome), while correcting for the confounding effect of physical activity before COVID-19 and the other relevant confounders included in the previous models, were fitted to examine the potential for physical literacy to be a protective factor against a decline in physical activity during COVID-19. To account for the increase probability of observing significant results due to multiple statistical comparisons (i.e. 15 logistic regressions), potentially leading to false positive results or Type I errors, we conducted a Bonferroni correction of the alpha level (i.e. α/k = 0.05/15 = 0.0033).

Results

Participants

A total of 648 Australian children were included (M_age = 10.1 ± 1.7 years; 60% boys). Approximately two-thirds of parents (61.7%) completed tertiary education. Three quarters of parents (75.0%) were in paid employment. Most of the parents (91.2%) spoke English at home. Participants from all Australian states and major territories were included. Parental characteristics are detailed in Table 1.

Table 1. Parental demographic characteristics.

	N	Percentage
Education
University or tertiary education	400	61.70
Technical/trade certificate/apprenticeship	102	15.74
Completed high school	98	15.12
Some high school or less	48	7.41
Employment
Paid employment	486	75.00
Unpaid employment	22	3.40
Home-duties full time	76	11.73
Unemployed	40	6.17
Student	13	2.01
Retired	9	1.39
Other^a	2	0.31
English at home
Yes	591	91.20
No	57	8.80
State of residence (Australia)
NSW	222	34.30
VIC	196	30.30
QLD	114	17.60
WA	52	8.00
SA	43	6.60
TAS	9	1.40
ACT	8	1.20
NT	4	0.60

Notes: Australian Capital Territory (ACT); New South Wales (NSW); Northern Territory (NT); Queensland (QLD); Victoria (VIC); Western Australia (WA); South Australia (SA); Tasmania (TAS). ^aUnknown.

Physical activity level and adherence to physical activity guidelines

The average number of days children were active for at least 60 min was 4.70 days (SD = 1.94) before COVID-19 and 4.47 days (SD = 1.98) during COVID-19. The proportion of children who were reported to meet the physical activity guidelines before COVID-19 (n = 170; 26.2%) was slightly higher than during COVID-19 (n = 160; 24.7%). A large proportion of the whole sample (n = 451; 70%) were reported to not meet the guidelines at either time points. Just over three quarters of children meeting the physical activity guidelines before COVID-19 also met them during COVID-19 (n = 133; 78.2%).

Preliminary analyses showed significant differences by parental occupation (i.e. paid employment vs other) in relation to meeting the physical activity guidelines. None of the other investigated factors appeared significant related to children’s adherence to the physical activity guidelines (see Table 2).

Table 2. Adherence to physical activity stratified by participants’ characteristics.

Characteristic	Not meeting PA guidelines		Meeting PA guidelines		Overall
	n (%)	Mean ± SD	n (%)	Mean ± SD	n (%)	Mean ± SD	t / χ^2	p-value
Before COVID-19
Age	478	10.2 ± 1.7	170	10.0 ± 1.7	648	10.1 ± 1.7	1.40	0.162
Sex^a							1.75	0.186
Boys	289 (61.1)		94 (55.3)		383 (59.6)
Girls	184 (38.9)		76 (44.7)		260 (40.4)
Education ^b							1.62	0.202
Tertiary education	302 (63.2)		98 (57.7)		400 (61.7)
Non-tertiary education	176 (36.8)		72 (42.3)		248 (38.3)
Employment							5.62	0.018
Paid employment	370 (77.4)		116 (68.2)		486 (75.0)
Unpaid employment / unemployed ^c	108 (22.6)		54 (31.8)		162 (25.0)
English at home							1.55	0.213
Speak English at home	432 (90.4)		159 (93.5)		591 (91.2)
Do not speak English at home	46 (9.62)		11 (6.47)		57 (8.8)
During COVID-19
Age	488	10.2 ± 1.7	160	10.0 ± 1.7	648	10.1 ± 1.7	0.97	0.333
Sex^a							0.97	0.324
Boys	293 (60.7)		90 (56.3)		383 (59.6)
Girls	190 (39.3)		70 (43.7)		260 (40.4)
Education^b							0.80	0.372
Tertiary education	306 (62.7)		94 (58.7)		400 (61.7)
Non-tertiary education	182 (37.3)		66 (41.3)		248 (38.3)
Employment							7.48	0.006
Paid employment	379 (77.7)		107 (66.9)		486 (75.0)
Unpaid employment / unemployed^c	109 (22.3)		53 (33.1)		162 (25.0)
English at home							1.72	0.190
Speak English at home	441 (90.4)		150 (93.8)		591 (91.2)
Do not speak English at home	47 (9.6)		10 (6.2)		57 (8.8)

Notes: Physical activity (PA); standard deviation (SD); p-values indicate the statistical significance of differences between groups, assessed with t-test or χ² test depending on the nature of the variables (categorical/continuous); significant differences are marked in bold. ^aChildren whose gender was reported to be ‘other’ or ‘prefer not to answer/say’ were not included (n = 5); ^bTertiary education refers to university or tertiary education, whereas non-tertiary education included technical/trade certificate/apprenticeship, completed high school and some high school or less; ^cIncluding unpaid employment, retired, unemployed, home-duties full time, student and other.

Physical literacy

The average score of children’s overall physical literacy was 93.7 ± 16.4. Boys’ perception of their own skills related to the physical domain appeared significantly higher than girls’. No sex-based differences emerged in relation to the other physical literacy domains. Overall and sub-domain physical literacy scores for the whole sample and stratified by sex are presented in Table 3.

Table 3. Physical literacy total and subdomain scores stratified by sex.

	Score range		Boys (n = 383)	Girls (n = 260)	Overall (n = 643)^a
Physical Literacy	Possible	Actual	Mean ± SD	Mean ± SD	Mean ± SD	T-statistic	P-value
Overall score	30–120	32–120	94.6 ± 16.4	92.7 ± 16.2	93.8 ± 16.4	1.44	0.152
Physical domain	12–48	12–48	37.1 ± 7.2	35.5 ± 7.2	36.4 ± 7.2	2.76	0.006
Psychological domain	7–28	7–28	21.9 ± 4.5	21.6 ± 4.3	21.8 ± 4.4	0.92	0.356
Social domain	4–16	4–16	13.2 ± 2.7	13.4 ± 2.6	13.2 ± 2.7	−0.92	0.360
Cognitive domain	7–28	7–28	22.4 ± 4.5	22.3 ± 4.3	22.4 ± 4.4	0.46	0.645

Note: Standard deviation (SD). ^aOnly male and female were included in this comparison; children whose gender was reported to be ‘other’ or ‘prefer not to say’ (n = 5) were not included.

Amongst the other factors, significant differences in physical literacy scores were also noted in relation to parental education (tertiary vs non-tertiary education) and parental occupation (paid occupation vs other). Physical literacy overall and subdomain scores were higher for children whose parents completed tertiary education compared to those who did not. Similar differences were noted in relation to parental occupation, with children of parents in paid employment showing higher scores than those in other circumstances (i.e. unpaid employment, retired, unemployed, home-duties full time, student and other). Detailed results are presented in Table 4.

Table 4. Physical literacy total and subdomain scores stratified by parental education and occupation.

	Parental Education^a			Parental Occupation			Overall (n = 648)
Physical Literacy	Tertiary (n = 400)	Non-tertiary (n = 248)	P-value	Paid employment (n = 486)	Other^b (n = 162)	P-value	Mean ± SD
Overall score	96.2 ± 15.5	89.7 ± 17.1	<0.0001	94.7 ± 16.0	90.7 ± 17.2	0.009	93.7 ± 16.4
Physical domain	37.2 ± 6.9	35.1 ± 7.5	0.0003	36.9 ± 7.0	34.9 ± 7.5	0.003	36.4 ± 7.2
Psychological domain	22.4 ± 4.2	20.8 ± 4.6	<0.0001	22.0 ± 4.3	21.1 ± 4.7	0.030	21.8 ± 4.4
Social domain	13.6 ± 2.4	12.6 ± 3.0	<0.0001	13.3 ± 2.7	13.1 ± 2.7	0.400	13.2 ± 2.7
Cognitive domain	23.0 ± 4.3	21.2 ± 4.5	<0.0001	22.6 ± 4.3	21.6 ± 4.7	0.026	22.3 ± 4.4

Note: Presented values indicate mean and standard deviation (M ± SD) of physical literacy scores.

^a Tertiary education refers to university or tertiary education, whereas non-tertiary education included technical/trade certificate/apprenticeship, completed high school and some high school or less; ^bIncluding unpaid employment, retired, unemployed, home-duties full time, student and other.

Association between physical literacy and adherence to physical activity guidelines

Children’s physical literacy overall and subdomain scores stratified by their adherence with the physical activity guidelines are presented in Table 5.

Table 5. Physical literacy overall and subdomain stratified by adherence with physical activity guidelines before and during COVID-19.

Physical Literacy	Not meeting PA guidelines		Meeting PA guidelines		Overall		T-statistic	P-value
	n	Mean ± SD	n	Mean ± SD	n	Mean ± SD
Before COVID-19
Physical literacy score	478	92.3 ± 16.9	170	97.7 ± 14.2	648	93.7 ± 16.4	−4.08	0.0001
Physical domain	478	35.8 ± 7.3	170	38.0 ± 6.6	648	36.4 ± 7.2	−3.55	0.0004
Psychological domain	478	21.4 ± 4.6	170	22.7 ± 4.0	648	21.8 ± 4.4	−3.45	0.0006
Social domain	478	13.0 ± 2.8	170	13.9 ± 2.2	648	13.9 ± 2.7	−4.27	< 0.0001
Cognitive domain	478	22.0 ± 4.6	170	23.1 ± 3.9	648	22.3 ± 4.4	−3.05	0.0025
During COVID-19
Physical literacy score	488	91.9 ± 16.9	160	99.1 ± 12.6	648	93.7 ± 16.4	−5.40	< 0.0001
Physical domain	488	35.7 ± 7.3	160	38.5 ± 6.5	648	36.4 ± 7.2	−4.46	< 0.0001
Psychological domain	488	21.4 ± 4.6	160	23.0 ± 3.8	648	21.8 ± 4.4	−4.44	< 0.0001
Social domain	488	12.9 ± 2.8	160	14.1 ± 2.0	648	13.2 ± 2.7	−5.59	< 0.0001
Cognitive domain	488	21.9 ± 4.6	160	23.6 ± 3.8	648	22.3 ± 4.4	−4.55	< 0.0001

Note: Physical activity (PA); standard deviation (SD).

Logistic regressions showed that a higher physical literacy overall score was associated with higher likelihood of meeting the physical activity guidelines, either before (OR = 1.02, 95% CI [1.01, 1.04], p < 0.001) or during COVID-19 (OR = 1.03, 95% CI [1.02, 1.05], p < 0.001), while adjusting for the potential confounding effect of children’s sex and age and parental occupation. The predicted probability of meeting the physical activity guidelines before/during COVID-19 by physical literacy score is presented in Figure 2.

Figure 2. Predicted probability of meeting the physical activity guidelines before (A) or during (B) COVID-19 according to children’s physical literacy score.

In addition, physical literacy remained positively associated with the adherence to physical activity guidelines during COVID-19, even after adjusting for children’s adherence to the guidelines before COVID-19. For each unit of increase in physical literacy overall score, the odds of meeting the physical activity guidelines increased by 3% (OR = 1.03, 95% CI [1.01, 1.05], p = 0.001).

Similar significant positive associations were also found between each physical literacy domains and adherence to the physical activity guidelines before or during COVID-19 (Physical: ORs range = 1.05–1.07; Psychological: ORs range = 1.08–1.10; Social: ORs range = 1.16–1.22; Cognitive: ORs range = 1.07–1.12), while adjusting for relevant confounders. Detailed results of the regression models are presented in Table 6.

Table 6. Associations between physical literacy and meeting physical activity guidelines before and during COVID-19.

	Meeting PA guidelines before COVID-19				Meeting PA guidelines during COVID-19				Meeting PA guidelines during COVID-19
			95% CI				95% CI				95% CI
Predictors	OR	p-value	LL	UL	OR	p-value	LL	UL	OR	p-value	LL	UL
Physical literacy (overall score)	1.02	< 0.001	1.01	1.04	1.03	< 0.001	1.02	1.05	1.03	0.001	1.01	1.05
Girls^a	1.23	0.261	0.86	1.78	1.17	0.405	0.80	1.71	1.01	0.979	0.58	1.75
Child age	0.93	0.164	0.83	1.03	0.94	0.294	0.84	1.05	1.00	0.976	0.85	1.17
Paid employment^b	0.58	0.009	0.39	0.87	0.52	0.002	0.34	0.78	0.58	0.079	0.31	1.07
Meeting PA guidelines before COVID-19									57.45	< 0.001	33.24	99.31
Physical domain	1.05	< 0.001	1.02	1.08	1.07	< 0.001	1.04	1.10	1.05	0.008	1.01	1.10
Girls^a	1.28	0.190	0.89	1.84	1.22	0.293	0.84	1.79	1.05	0.851	0.61	1.83
Child age	0.93	0.199	0.84	1.04	0.95	0.367	0.85	1.06	1.01	0.918	0.86	1.18
Paid employment^b	0.58	0.008	0.39	0.87	0.52	0.002	0.35	0.78	0.59	0.089	0.32	1.08
Meeting PA guidelines before COVID-19									56.98	< 0.001	33.13	97.98
Psychological domain	1.08	0.001	1.03	1.13	1.10	< 0.001	1.05	1.15	1.09	0.010	1.02	1.16
Girls^a	1.21	0.295	0.84	1.75	1.15	0.463	0.79	1.67	0.98	0.954	0.57	1.71
Child age	0.93	0.169	0.83	1.03	0.94	0.309	0.85	1.05	1.00	0.993	0.85	1.18
Paid employment^b	0.60	0.012	0.40	0.90	0.54	0.003	0.36	0.81	0.61	0.112	0.33	1.12
Meeting PA guidelines before COVID-19									57.50	< 0.001	33.43	98.93
Social domain	1.16	< 0.001	1.07	1.25	1.22	< 0.001	1.12	1.32	1.20	0.003	1.07	1.35
Girls^a	1.16	0.425	0.81	1.67	1.08	0.674	0.74	1.58	0.91	0.740	0.52	1.58
Child age	0.93	0.159	0.83	1.03	0.94	0.293	0.84	1.05	1.00	0.957	0.85	1.17
Paid employment^b	0.62	0.017	0.41	0.92	0.55	0.004	0.37	0.83	0.61	0.108	0.33	1.11
Meeting PA guidelines before COVID-19									57.34	< 0.001	33.23	98.96
Cognitive domain	1.07	0.003	1.02	1.11	1.11	< 0.001	1.06	1.16	1.12	0.001	1.05	1.20
Girls^a	1.20	0.328	0.83	1.72	1.13	0.511	0.78	1.65	0.96	0.871	0.55	1.66
Child age	0.93	0.172	0.83	1.03	0.94	0.294	0.85	1.05	0.99	0.878	0.84	1.16
Paid employment^b	0.60	0.013	0.41	0.90	0.53	0.002	0.36	0.80	0.57	0.075	0.31	1.06
Meeting PA guidelines before COVID-19									60.25	< 0.001	34.69	104.64

Notes: Physical activity (PA); Odd ratio (OR); Confidence intervals (CI); Lower limit (LL); Upper limit (UL). Significant associations adjusted for multiple comparisons (i.e. Bonferroni correction of the alpha level: α/k = 0.05/15 = 0.0033) are marked in bold. ^aReference level: Boys; ^bReference level: Other (i.e. including unpaid employment, retired, unemployed, home-duties full time, student and other).

Discussion

The aim of this study was to assess the relationship between physical literacy and children’s physical activity before/during COVID-19. To the best of our knowledge, this study is the first to empirically assess the relationship between physical literacy assessed using the PL-C Quest (i.e. the only measure currently aligned with the holistic Australian Physical Literacy Framework) and adherence to physical activity guidelines in primary-school-aged children. We found that children with a higher physical literacy score had a higher odds of meeting physical activity guidelines before and during COVID-19. Also, our additional analysis suggested that the strength of the association was maintained even after adjusting for whether children met the physical activity recommendations before COVID-19, suggesting that physical literacy may have acted as a protective factor against physical activity decline during the pandemic. For every one unit of increase in the overall physical literacy, children’s odds of meeting physical activity guidelines can increase by 3%. To understand what this finding means in real terms, if a child’s physical literacy increases by four units (out of a potential 120), their odds of meeting physical activity guidelines may increase by 12%. This could be seen as having potential for meaningful impact considering the high prevalence of children’s inactivity (Guthold et al. 2020).

The positive associations were also found in relation to each of the physical literacy domains, with greater magnitude of effects observed for each of the physical literacy domains compared to the overall score. This appears to suggest that increasing the score of a single physical literacy domain has a potential greater impact on children’s physical activity than the overall score. For example, one unit increase in the social domain score is associated with a 16–22% greater odds to meet the physical activity guidelines before or during COVID-19, whereas one unit increase in the overall physical literacy score corresponds to 2–3% increased odds of meeting the guidelines. However, it is important to note that the scale of the predictor variable influences the magnitude of effect of a predictor variable in logistic regression, which explains why greater effects were noted for the social domain (possible scale range = 4–16) than the overall physical literacy construct (possible score range = 30–120).

Combined with previous research on the relationship between physical literacy and physical activity, our research provides additional supporting evidence on the important role that physical literacy can play in physical activity promotion. In general, our finding aligns with previous research that examined physical literacy and physical activity although specific comparison might be difficult in most cases due to differences in the measurement employed by the different studies to measure physical literacy and/or physical activity, and in some cases due to the different age of the target population (i.e. children vs. adolescents). To our knowledge, only one other study, by Belanger et al. (2018), explicitly assessed the association between physical literacy and children’s adherence to physical activity guidelines. This study was based on the Canadian physical literacy concept, measured in 2956 children aged 8–12 years using the CAPL-2 (a measure that focuses on behavioural, physical, cognitive and psychological elements of physical literacy) (Belanger et al. 2018). In that study, children with higher physical literacy scores were also more likely to meet physical activity recommendations (≥12,000 steps; ≥ 6 days/week) compared to those with lower scores (Belanger et al. 2018). However, the association was much stronger than in the current study (OR = 1.2 − 2.1 across the domains versus our finding of OR = 1.03) (Belanger et al. 2018). One explanation could be that the two studies adopted different physical literacy definitions and physical literacy measurements. The CAPL-2 was based on the International Physical Literacy Association definition and the Canadian interpretation (Longmuir et al. 2018). The CAPL-2 does not include a social domain, which is part of the Australian Framework of Physical Literacy, so the results are not directly comparable (Australian Sports Commission 2019). Also, CAPL-2 utilises motor skills and fitness measurement in the physical competence domain, whereas the PL-C Quest uses self-report measurements across all domains (Barnett et al. 2022; Caldwell et al. 2020).

Other studies assessed the relationship between physical literacy and minutes in physical activity (Caldwell et al. 2020; Choi et al. 2018). Caldwell et al.’s (2020) 2-year longitudinal cohort study including 222 children between 8 and 13 years of age, found that higher physical literacy (measured using the PLAY Tools) was associated with higher pedometer-measured moderate-to-vigorous intensity physical activity in children. While the PLAY Tools study showed a positive relationship between physical literacy and physical activity, there are potential limitations with data analysis (Caldwell et al. 2020). Each PLAY instrument collects child data on physical literacy using a different medium. PLAYfun assesses children’s motor skill level by teachers’ observation, PLAYself assess child perceptions of physical literacy and PLAYparent assesses parent report of child physical literacy (Caldwell et al. 2020). Thus, PLAYself and PLAYparent may be correlated results as they arguably measure the same concept. As such, it may not be appropriate to include both the PLAYself and the PLAYparent results in the physical literacy composite score when assessing its association or correlation with physical activity, as predicting variables are assumed to be independent of each other in regression analyses. Thus, the strength of the relationship showed in this study may not be an appropriate reference for the physical literacy and physical activity relationship.

Choi et al.’s (2018) study in 1945 adolescents aged between 12 and 18 years, also found physical literacy [measured using the Perceived Physical Literacy Instrument (PPLI) – a self-reported assessment of the children’s perception of physical literacy, social and cognitive elements (Sum et al. 2016)] was positively associated with minutes in recreational physical activity (measured with the International Physical Activity Questionnaire for Adolescents). The PPLI used in Choi et al.’s (2018) study is a self-report instrument used to assess perceptions around three sub-constructs of physical literacy, that were categorised as follows: sense of self and self-confidence; self-expression and communication with others; and knowledge and understanding. Although the PL-C Quest and the PPLI are both self-report physical literacy instrument and share similarities in relation to the way responses are collected using a Likert scale, many aspects between the two instruments differ. For example, the PL-C Quest presents each of its 30 items as a dual scenario of a fantastical character performing a related activity accompanied by a written description of the represented concept; response options are anchored to a 4-point Likert scale and relate to how closely the participant relates with one scenario or the other. On the contrary, the PPLI presents each of the nine items as written statements and asks the participant to indicate agreement/disagreement on a 5-point Likert scale. In addition the participants in Choi et al.’s (2018) study were adolescents, which further limits the potential comparison with our study (i.e. only including children).

Strength and limitations

The main strengths of the current study are the large sample size with geographical, parental education and parental employment diversity. Another strength is that our questionnaires were administered online, thus the social distancing restrictions in most areas of Australia did not limit the application of the questionnaires. A limitation is subjective physical activity measurement. Although subjective questionnaires may have lower validity compared to objective measurements, Dollman, Norton, and Norton (2005) suggest that survey questionnaires can be suitable for large descriptive studies due to lower cost and lower participant burden. Moreover, the validity of our questionnaires is adequate (Barnett et al. 2022; Ridgers et al. 2012). The cross-sectional design limits our ability to determine the directionality of the associations. Children’s physical activity was proxy-reported by parents/carers, which is another limitation of our study. The findings may not be generalised to all Australian primary-school-aged children due to the potential selection bias due to the convenient sampling method used for the study. Further, we acknowledge that the varying severity and duration of COVID-19 restrictions across Australian States and Territories during the survey period pose a limitation to our study, as we do not have detailed data accounting for the specific circumstances affecting respondents at the time of survey completion. However, given the evidence regarding the relationship between the physical literacy and physical activity is still emerging, our study has added evidence to current research.

Conclusion

Our study investigated the possible association between physical literacy and physical activity during COVID-19. It is one of few studies which has examined the relationship between physical literacy and children’s adherence to physical activity guidelines. Our results suggested that for every unit of increase in the physical literacy score, children’s odds of meeting the guidelines would increase by 3%. This finding provides educators and policymakers with the potential to focus on building and developing children’s physical literacy as a mechanism for increasing physical activity, through holistic and inclusive approaches aimed at improving the physical (e.g. fitness, motor skills) as well as the non-physical skills (e.g. collaboration, confidence, knowledge) related to lifelong physical activity participation. The PL-C Quest may be used to identify individual areas that require improvement, setting individualised goals, tailoring instructions, and monitoring progress. Further research should explore how physical literacy relates with physical activity, and other physical, social, psychological and cognitive skills, extending beyond self-report measures – where possible, e.g. using accelerometers, fitness / motor skill / cognitive assessment batteries – and employing longitudinal and experimental studies, to investigate if changes in one variable can generate effects on the other (causality).

Ethics approval and consent to participate

A low-risk ethics application was approved by the Faculty of Health Human Ethics Advisory Group in Deakin University (Reference number: HEAG-H 82_2019). Informed consent was obtained from parents or guardians of children before entering surveys.

Acknowledgements

This work has been based on the ‘Physical Literacy in Children Questionnaire (version 1)’, which is the copyright of the Australian Sports Commission. More information regarding the physical literacy tool can be found at https://www.sportaus.gov.au/physical_literacy/resources. We sincerely thank the parents and children who consented to take part in the study and contributed with their data. LB and EM led the conception and design of the project and data acquisition. JX contributed to the development of research questions and aims for this study. JX and EM led the data cleaning, management, analysis, and synthesis of results with support from LB. JX wrote the first manuscript draft. EM led the revision and improvement of the following manuscript drafts and statistical analyses. All authors reviewed the manuscript, provided intellectual contributions, and approved its final version.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The datasets for this study are available upon reasonable request submitted to the Faculty of Health Human Ethics Advisory Group in Deakin University (Reference number: HEAG-H 82_2019).

Additional information

Funding

The study received funding from the Institute for Physical Activity and Nutrition (IPAN), Deakin University.