The impact of different types of screen time on school-aged children’s quality of life and participation

Abstract

Introduction

This study investigated the impact of different types of screen time on quality of life and participation in school-aged children.

Method

Twenty-nine typically developing children (mean age: 9.34 $\pm$ 1.37, range: 8–14 years) and 20 of their parents completed the Children Screen Time Use Report, the KIDSCREEN-52 and the Participation and Environment Measure for Children and Youth. Spearman rho correlations and linear regressions with bootstrapping were used to assess associations between screen time, health-related quality of life and activity participation.

Results

Regression analysis identified that passive screen time was a significant predictor of various KIDSCREEN-52 domains: physical wellbeing (b = −0.445, SE B = 0.008, p = .016), autonomy (b = −0.445, SE B = 0.011, p = .016) and social acceptance (b = 0.447, SE B = 0.007, p = .048). Educational, social and passive screen time were positively correlated with increased participation in home, community and school activities, respectively.

Conclusion

Findings suggest that passive screen time was mostly predictive of lower quality of life levels. However, passive, educational and social screen time positively correlate with participation in typically developing school-aged children. These findings assist occupational therapists to understand the impact of screen time on health and participation in children.

Introduction

In the contemporary world, children grow up surrounded by technology (Stiglic and Viner, 2019) and The Royal College of Paediatrics and Child Health (RCPCH) has recommended that British children limit their screen time (ST) use to two hours per day (RCPCH, 2019). Increased ST use has been linked with poorer health outcomes and reduced physical activity participation, and as new uses of ST emerge, the RCPCH has stated that further research on this topic is warranted (RCPCH, 2019). In Australia, the Department of Health (2019) recommends similar ST guidelines; however, it has been found that Australian children engage in more than 10 hours of ST per week (Australian Bureau of Statistics [ABS], 2019). With the spread of the Covid-19 pandemic around the world and the resultant move to the delivery of tele-education and many other regular and routine activities for children online, the amount of daily ST that children engaged in increased exponentially (Lee et al., 2022; McArthur et al., 2021; Toombs et al., 2022). The purpose of this study was to examine the associations between different types of ST and participation and health-related quality of life (HRQoL) in a sample of typically developing school-aged Australian children.

Literature review

ST refers to the use or viewing of anything with a screen with various types of ST requiring different levels of cognitive and physical engagement from the user (Norozi et al., 2020; Sweetser et al., 2012). Sanders et al. have categorized four types of ST in Australian children: ‘educational ST (e.g. computer use for homework), social ST (e.g. social media), interactive ST (e.g. video games) and passive ST (e.g. television)’ (Sanders et al., 2019: p. 2).

Previous research has investigated the impacts of ST on health and wellbeing in children using the HRQoL which is a multi-dimensional concept that encompasses physical, mental and social domains of health (Centers for Disease Control and Prevention [CDC], 2020). For example, higher levels of television-viewing time have been associated with higher adiposity and poorer nutritional patterns (Delgado-Floody et al., 2020; Stiglic and Viner, 2019). Studies have also found correlations between screen time and poorer sleep quality, lower self-esteem in young girls and worse self-reported HRQoL in children (Mireku et al., 2019; Racine et al., 2011; Stiglic and Viner, 2019).

Nevertheless, some studies have reported positive associations between specific types of screen time and HRQoL. For instance, while total ST is associated with lower HRQoL, educational and interactive ST has been positively associated with children’s educational outcomes (Sanders et al., 2019). Co-viewing television with family is also reported to be positively associated with mental wellbeing in school-aged children (McDade-Montez et al., 2015). Many studies have tended to focus on a single type of ST, notably television watching, but there is only sparse research that explores the positives and negatives of different ST types (Stiglic and Viner, 2019).

ST has also been linked to changes in activity participation among children. Participation in meaningful activities is important for children as they provide opportunities to learn new skills, thereby promoting health, skill development, academic learning and mental wellbeing (Rosenberg et al., 2019). The ‘digital Goldilocks hypothesis’ suggests that moderate ST can provide children with valuable connections to social and leisure pursuits, but over-participation can displace meaningful activities (Przybylski and Weinstein, 2017). Therefore, ST has the potential to foster increased and decreased participation in meaningful activities. While higher ST use has been linked to lower physical activity participation, which in turn affects HRQoL, video games that incorporate physical movement as part of gameplay can elicit moderate and vigorous-intensity physical activity, and are viewed by users as enjoyable (Norozi et al., 2020; Stiglic and Viner, 2019; Zhang et al., 2017). Furthermore, studies suggest that when adjusted for sociodemographic variables, ST’s correlation to displaced recreational activity participation in children no longer exists (Lees et al., 2020). More research, however, on the associations between ST and participation is needed (Fairlie and Kalil, 2017).

Occupational therapists have a vested interest in children’s participation and occupational engagement within different environments to promote health and wellbeing (Law et al., 1996). As children grow up in the digital age, it is important to establish baseline knowledge and understanding about this relationship to ensure children continue to participate successfully in daily occupations within the home, school and community environment. The extent of the relationship between ST, HRQoL and participation has not been investigated in an Australian context. Furthermore, given the limited evidence base regarding this topic in the United Kingdom, and similar ST use trends between British and Australian children, further investigation of different types of ST is warranted (ABS, 2019; RCPCH, 2019).

Research questions

• 1. Are educational, social, interactive or passive screen times associated with children’s health-related quality of life?

• 2. Are educational, social, interactive or passive screen times associated with children’s participation in home, school and community activities?

Method

Design

This study used a cross-sectional survey design approach.

Participants

Participants were recruited using convenience sampling but due to restrictions during the Covid-19 pandemic, snowball sampling was used to recruit additional participants. The total sample included 29 child participants and 20 parents/caregivers. The inclusion criteria required children to be aged 8–14 years, parents/caregivers had to provide written consent, children had to provide written and verbal assent and children did not have a known intellectual, developmental, physical or mental disability or illness (based on parental/caregiver report).

Instrumentation

Demographic questionnaire

A demographic questionnaire was used to record participants’ age, gender, school grade, family structure, before and after school care, parents’ employment and the impact of Covid-19 restrictions on time spent between parents and children. The questionnaire was trialled with a group of four parents and two occupational therapists, whose feedback was sought before it was used in the study.

Children Screen Time Use Report (CSTUR)

The CSTUR was developed by the study authors to record child participants’ ST on two typical weekdays and two weekend days over 1 week. The CSTUR measures four types of ST: educational ST, social ST, interactive ST and passive ST Definitions of the different screen types were provided to support participants’ completion of the report. The tool’s contents used screen categories reported previously in the Australian child population (Sanders et al., 2019). The CSTUR was trialled with a group of four parents and two occupational therapists whose feedback was sought. The CSTUR’s face and content validity appeared favourable. Children were asked to report time spent in each category of ST in hours and minutes, and parents/caregivers also completed the CSTUR based on their own perception of their child’s ST use. The CSTUR took participants approximately 15 min to complete.

KIDSCREEN-52

The KIDSCREEN-52 measures children’s HRQoL in 10 domains: physical wellbeing, psychological wellbeing, moods and emotions, self-perception, autonomy, parent relations and home life, social support and peers, school environment, social acceptance and financial resources (Ravens-Sieberer et al., 2008). Using a 5-point Likert scale, children rate the frequency of behaviours and attitudes of 52 items. Higher scores are indicative of higher HRQoL. Many studies have used the self-report KIDSCREEN-52 and the instrument has established internal consistency and convergent and discriminant validity against the Pediatric Quality of Life Inventory (Delgado-Floody et al., 2020; Mireku et al., 2019; Taliep and Florence, 2012; Zhu et al., 2019). It took participants 20 min to complete the KIDSCREEN-52.

Participation and Environment Measure for Children and Youth (PEM-CY)

The PEM-CY is a parent/caregiver self-report instrument that measures children’s activity participation in three domains: home, school and community (Coster et al., 2011). In total, the PEM-CY measures 25 types of activity. Each activity requires parents/caregivers to report on the frequency of their child’s participation in the last 4 months, how involved their child is in the activity and whether the parent would like to see changes in participation for each activity. Parents/caregivers report on questions using 7-, 5- and 6-point Likert scale responses. Likewise, parents/caregivers are also asked to identify features of the environment that facilitate or hinder participation, using a 4-point Likert scale. The PEM-CY has moderate to good internal consistency (>0.59), test–retest reliability (>0.58) and excellent validity (Coster et al., 2011; Jeong et al., 2016). The PEM-CY took parents/caregivers 30 min to complete.

Data analysis

The Statistical Package for the Social Sciences (SPSS), version 27 (IBM Corp, 2020), was used to store and analyse the data and participants’ demographic data was calculated using descriptive statistics. An independent t-test was used to determine if there were significant differences between child-reported and parent-reported ST. The CSTUR scores for each type of ST were correlated with the subscales of the KIDSCREEN-52 and PEM-CY using Spearman rho correlations to measure the associations between ST, HRQoL and participation.

Linear regression analyses were completed to determine if a child’s ST (independent variable) was predictive of their HRQoL and participation (dependent variables). Only dependent variables that were significantly correlated with the independent variables were included in the regression models. Bootstrapping is an iterative resampling method that infers a population from the sample to increase the accuracy of the confidence interval estimate (Chernick, 2008). Bootstrapping was used to increase the study sample to 1000 for all statistical analyses to ensure independent t-tests, correlations and regression models were statistically sound.

Procedure

Ethical approval for this research study was approved by the Monash University Human Research Ethics Committee (MUHREC) (Project Number: 26,607). Likewise, ethics approval from the Victorian Department of Education and Training was sought as the project was completed in a school setting (Project Number: 2021_004,351, Approval date: 15/02/2021).

Participants were recruited from a metropolitan public primary school in Melbourne. Grade 3 to Grade 6 classes were approached with an invitation to participate in the study. All parents/caregivers of children who met the inclusion criteria were also invited to participate in the study. Parents/caregivers and child participants were informed of the study’s purpose, procedures and the voluntary nature of participation. Five participants were excluded from the school sample because they did not meet the inclusion criteria, resulting in a total of 24 recruits from the school. Using snowball sampling, seven additional children were recruited through the promotion of a brief description about the study to parents in the local community. Two participants were excluded from the snowball sample because they did not meet the inclusion criteria.

The total response rate was 12.9%. After screening against the inclusion criteria, only 10.4% of the sample initially approached were included in the study. All 29 child participants were included in the study, but only 20 parents/caregivers were included in the data analysis as nine parents/caregivers did not complete and return the parent/caregiver-reported CSTUR and PEM-CY to the investigators, despite three follow-up attempts. The response rate for parents/caregivers was 7.1%.

Results

Participant demographics

The study included 29 typically developing school-aged children with a mean age of 9.34 $\pm$ 1.37 years and 20 parents/caregivers. There was nearly an even number of female and male child participants (15 females and 14 males). Participants predominantly consisted of children who were in the fourth grade, spoke English as their primary language at home, did not attend before or after school care and had parents who worked part-time and were not affected by the Covid-19 lockdown restrictions. The full demographic findings are reported in Table 1.

Table 1.

Demographic data (n = 29).

Participant characteristics	Frequency (n)	Percent (%)
Gender
Male	14	48.3
Female	15	51.7
School grade
Grade 3	8	27.6
Grade 4	14	48.3
Grade 5	1	3.4
Grade 6	5	17.2
Grade 8	1	3.4
Main language spoken at home
English	26	89.7
Other	3	10.3
Parent living situation with child
Full time	29	100
Parent employment
No work	1	3.4
Full time	12	41.4
Part time	16	55.2
Child before/afterschool care
No	22	75.9
Sometimes	7	24.1
Lockdown restriction impact on time spent with child
Yes	9	31
No	20	69

Child- and parent-reported screen time differences

An independent-samples t-test with a bootstrapped sample of 1000 was conducted to examine whether any differences existed between children’s own self-reported ST use and their ST use as reported by their parents. There was a significant difference in educational ST on a weekday reported by children (M = 41.69, SD = ± 41.52) and parents (M = 19.37, SD = ± 22.13; t = 2.44, p = 0.02). No other statistically significant differences in other types of ST scores were found.

Interactive and passive ST were the highest reported ST type for children by both child and parent respondent groups. All ST types increased on weekends except for educational ST Table 2 reports the full details of the child- and parent-reported differences in children’s ST use.

Table 2.

Child and parent differences in measures of child screen time-use (child: n = 29, parent: n = 20; bootstrapped sample of 1000).

						Before bootstrapping a				After bootstrapping b
	Child		Parent		Cohen’s D	df	t	p	95% CI [LL, UL]	t	p	95% CI [LL, UL]
	M	SD	M	SD
Educational ST/weekday	41.690	41.519	19.368	22.131	35.227	46	2.147	0.037	[1.393, 43.250]	2.436	0.024	[6.010, 41.854]
Educational ST/weekend	5.276	10.661	7.895	16.859	13.431	46	−0.661	0.512	[−10.598, 5.361]	−0.612	0.577	[−11.775, −5.450]
Social ST/weekday	28.397	39.937	24.737	55.565	46.679	46	0.266	0.792	[−24.073, 31.392]	0.247	0.813	[−31.480, 29.849]
Social ST/weekend	28.569	39.326	36.316	51.924	44.681	46	−0.587	0.560	[−34.292, 18.798]	−0.544	0.599	[−39.358, 18.409]
Interactive ST/weekday	76.345	90.242	51.579	72.549	83.765	46	1.002	0.322	[−25.000, 74.532]	1.048	0.318	[−22.821, 71.093]
Interactive ST/weekend	83.293	117.128	83.421	103.468	111.981	46	−0.004	0.997	[−66.657, 66.401]	−0.004	−	[−61.275, 58.475]
Passive ST/weekday	50.483	49.608	40.658	45.161	47.917	46	0.695	0.491	[−18.643, 38.293]	0.718	0.469	[−20.248, 38.392]
Passive ST/weekend	84.569	59.172	100.536	74.417	65.561	46	−0.825	0.414	[−54.908, 22.993]	−0.813	0.416	[−59.766, 24.386]

Note. M: mean; SD: standard deviation; p: significance (2-tailed); CI: confidence interval; LL: lower limit; UL: upper limit.

^aValues based on ‘equal variances assumed’ as all Levene’s Test for Equality of Variances had more than 0.05 significance.

^bBootstrapping specifications: (a) sampling method: simple; (b) number of samples: 1000; (c) CI level: 95%; and (4) CI type: BCa.

Associations between different types of ST with HRQoL

Parent-reported screen time

There was a significant negative correlation between educational ST on a weekday with the KIDSCREEN-52 moods and emotions (rho = −0.519, p = .019) domain. Interactive ST on a weekday was significantly negatively correlated with the KIDSCREEN-52 financial resources (rho = −0.461, p = .041), social support and peers (rho = −0.506, p = .023) and the school environment (rho = −0.538, p = .014) domains. Interactive ST on a weekend was significantly negatively correlated with the KIDSCREEN-52 parent relations and home life (rho = −0.444, p = .05), financial resources (rho = −0.459, p = .042) and the school environment (rho = −0.462, p = .04) domains. However, passive ST on a weekend was significantly positively correlated with the KIDSCREEN-52 social acceptance (rho = 0.523, p = .018) domain.

Child-reported screen time

There were statistically significant negative correlations between interactive ST on a weekday and four KIDSCREEN-52 domains: moods and emotions (rho = −0.463, p = 0.011), financial resources (rho = −0.418, p = .024), social support and peers (rho = −0.496, p = 0.006) and the school environment (rho = −0.534, p = .003). There was a significant negative correlation between interactive ST on a weekend with the KIDSCREEN-52 parent relations and home life (rho = −0.505, p = .005), social support and peers (rho = −0.456, p = 0.13) and the school environment (rho = −0.463, p = 0.12) domains.

Higher passive hours on a weekday were significantly correlated with lower scores on the KIDSCREEN-52 social support and peers (rho = −0.046, p = .012) and the school environment (r = −0.456, p = .013) domains. Higher passive ST on a weekend was significantly correlated with lower scores on the KIDSCREEN-52 physical wellbeing (rho = −0.374, p = .046), autonomy (rho = −0.462, p = .012), social support and peers (rho = −0.494, p = .006) and the school environment (rho = −0.564, p = .001) domains. There were no statistically significant correlations between educational and social ST on a weekday or weekend with any of the KIDSCREEN-52 domains.

Associations between different types of ST on children’s participation

Parent-reported screen time

Higher educational ST on a weekday was significantly correlated with higher scores in the PEM-CY home participation frequency (rho = 0.531, p = .019) subscale. Higher educational ST on a weekend was significantly correlated with higher scores on the PEM-CY home participation frequency (rho = 0.500, p = .029), home involvement (rho = 0.464, p = .045) and home environment support (rho = 0.580, p = .009) subscales. Social ST on a weekday was significantly positively correlated with the PEM-CY home participation frequency (rho = 0.481, p = .037) and home environment support (rho = 0.582, p = .009) subscales. Higher social ST on a weekend was statistically significantly correlated with an increased desire for home participation change by parents (rho = 0.460, p = .048) on the PEM-CY. Passive ST on a weekend was also significantly correlated with an increased desire for home participation change (rho = 0.495, p = .031) on the PEM-CY.

Child-reported screen time

Educational ST on a weekend was significantly positively correlated with the PEM-CY home environmental support subscale (rho = 0.471, p = .042). Social ST on a weekday was significantly positively correlated with the PEM-CY home environmental support (rho = 0.547, p = .015), school environmental support (rho = 0.715, p = .001) and community environmental support (rho = 0.459, p = .048) subscales. Social ST on a weekend was significantly positively correlated with the PEM-CY community participation frequency subscale (rho = 0.468, p = .043) while passive ST on a weekday was significantly positively associated with the PEM-CY school participation frequency subscale (rho = 0.475, p = .040). However, passive ST on a weekend was significantly negatively correlated with the PEM-CY school environmental support subscale (rho = −0.496, p = .031).

Table 3 shows the CSTUR screen time types that were significantly correlated with KIDSCREEN-52 and PEM-CY subscales after bootstrapping for 1000 samples. The full table reporting all significant and non-significant correlation results is available from the author upon request.

Table 3.

CSTUR screen time types that were significantly correlated with KIDSCREEN-52 and PEM-CY subscales (child: n = 29, parent: n = 20; bootstrapped sample of 1000).

	Physical wellbeing	Psychological wellbeing	Moods and emotions	Self-perception	Autonomy	Parent relations and home life	Financial resources	Social support and peers	School environment	Social acceptance (bullying)
Average parent-reported educational hours/weekday
rho	0.071	0.056	−0.519*	−0.275	0.013	−0.194	−0.183	−0.247	−0.264	−0.003
p	.766	.813	.019	.241	.958	.411	.440	.294	.260	.99
Bootstrap a
Bias	−0.016	−0.017	0.015	0.014	0.009	0.006	0.013	0.006	0.004	0.023
SE	0.233	0.213	0.150	0.217	0.254	0.228	0.219	0.212	0.222	0.240
BCa 95% CI [LL, UL]	[−0.383, 0.457]	[−0.332, 0.396]	[−0.755, −0.163]	[−0.616, 0.250]	[−0.488, 0.582]	[−0.627, 0.312]	[−0.591, 0.295]	[−0.674, 0.228]	[−0.684, 0.182]	[−0.473, 0.525]
Average parent-reported interactive hours/weekday
rho	−0.313	−0.23	−0.297	−0.109	−0.217	−0.286	−0.461*	−0.506*	−0.538*	−0.147
p	.179	.329	.203	.649	.357	.221	.041	.023	.014	.536
Bootstrap a
Bias	0.019	0.027	0.014	0.007	0.012	0.012	0.009	0.015	0.02	−0.008
SE	0.218	0.229	0.209	0.227	0.208	0.226	0.179	0.171	0.16	0.214
BCa 95% CI [LL, UL]	[−0.625, 0.154]	[−0.673, 0.373]	[−0.712, −0.231]	[−563, 0.376]	[−0.603, 0.210]	[−0.675, 0.183]	[−0.735, −0.111]	[−0.785, −0.133]	[−0.782, −0.139]	[−0.549, 0.265]
Average parent-reported interactive hours/weekend
rho	−0.100	−0.226	−0.363	−0.182	−0.225	−0.444*	−0.459*	−0.394	−0.462*	−0.114
p	.675	.339	.115	.442	.340	.050	.042	.086	.040	.633
Bootstrap a
Bias	0.007	0.024	0.018	0.014	0.018	0.015	0.018	0.016	0.016	0
SE	0.223	0.243	0.230	0.216	0.225	0.217	0.184	0.199	0.219	0.226
BCa 95% CI [LL, UL]	[−0.482, 0.318]	[−0.647, 0.313]	[−0.829, 0.249]	[−0.585, 0.298]	[−0.662, 0.298]	[−0.787, 0.069]	[−0.751, 0.006]	[−0.726, −0.067]	[−0.843, 0.031]	[−0.566, 0.382]
Average parent-reported passive hours/weekend
rho	−0.092	−0.129	0.377	0.211	0.124	−0.120	0.318	0.152	0.248	0.523*
p	.698	.587	.102	.372	.603	.613	.171	.522	.291	.018
Bootstrap a
Bias	0.020	0.014	−0.020	−0.017	−0.025	0.010	−0.025	−0.005	−0.019	−0.029
SE	0.226	0.222	0.208	0.239	0.264	0.208	0.198	0.224	0.221	0.187
BCa 95% CI [LL, UL]	[−0.506, 0.404]	[−0.463, 0.311]	[−0.056, 0.693]	[−0.233, 0.582]	[−0.220, 0.524]	[−0.561, 0.394]	[−0.087, 0.631]	[−0.240, −0.511]	[−0.183, −0.594]	[−0.139, 0.790]
Average child-reported interactive hours/weekday
rho	−0.175	−0.159	−0.463*	−0.156	−0.360	−0.339	−0.418*	−0.496**	−0.534**	−0.095
p	.363	.410	.011	.419	.055	.072	.024	.006	.003	.624
Bootstrap a
Bias	−0.001	0.004	0.006	0.001	0.003	−0.001	0.007	0.014	0.014	0.003
SE	0.175	0.164	0.163	0.196	0.183	0.176	0.163	0.134	0.143	0.195
BCa 95% CI [LL, UL]	[−0.49, 0.174]	[−0.464, 0.179]	[−0.733, −0.126]	[−0.532, 0.247]	[−0.681, 0.041]	[−0.661, 0.019]	[−0.731, −0.032]	[−0.725, −0.193]	[−0.771, −0.199]	[−0.468, 0.306]
Average child-reported interactive hours/weekend
rho	−0.149	−0.155	−0.223	−0.185	−0.268	−0.273	−0.505**	−0.456*	−0.463*	−0.201
p	.441	.421	.246	.338	.16	.151	.005	.013	.012	.295
Bootstrap a
Bias	0.001	0.004	0.005	0.003	0.012	−0.001	0.011	0.012	0.016	0
SE	0.185	0.171	0.203	0.187	0.173	0.194	0.140	0.164	0.164	0.183
BCa 95% CI [LL, UL]	[−0.509, 0.238]	[−0.475, 0.224]	[−0.593, 0.206]	[−0.559, 0.184]	[−0.573, 0.111]	[−0.62, 0.139]	[−0.758, −0.115]	[−0.731, −0.098]	[−0.746, −0.067]	[−0.533, 0.145]
Average child-reported passive hours/weekday
rho	−0.323	−0.138	−0.058	0.075	−0.184	−0.019	−0.29	−0.460*	−0.456*	−0.118
p	.087	.475	.766	.699	.34	.922	.126	.012	.013	.542
Bootstrap a
Bias	0.009	0.012	0.007	−0.001	0.012	0.003	0.01	0.019	0.015	0.005
SE	0.157	0.173	0.185	0.189	0.187	0.185	0.183	0.151	0.154	0.176
BCa 95% CI [LL, UL]	[−0.621, 0.023]	[−0.496, 0.271]	[−0.43, 0.337]	[−0.314, 0.442]	[−0.506, 0.173]	[−−0.368, 0.378]	[−0.638, 0.126]	[−0.707, −0.089]	[−0.688, −0.119]	[−0.436, 0.222]
Average child-reported passive hours/weekend
rho	−0.374*	−0.150	−0.104	−0.048	−0.462*	−0.049	−0.316	−0.494**	−0.564**	−0.097
p	.046	.438	.590	.804	.012	.801	.095	.006	.001	.617
Bootstrap a
Bias	0.007	0.008	0.003	0.001	0.015	0	0.003	0.018	0.012	0.002
SE	0.173	0.179	0.181	0.192	0.158	0.186	0.177	0.158	0.143	0.194
BCa 95% CI [LL, UL]	[−0.68, −0.008]	[−0.52, 0.27]	[−0.49, 0.285]	[−0.428, 0.329]	[−0.73, −0.083]	[−0.424, 0.336]	[−0.66, 0.067]	[−0.752, −0.133]	[−0.807, −0.173]	[−0.449, 0.272]
	Home participation frequency	Home involvement	Parents' desired change in home participation	Home environment support	School participation frequency	School involvement	Parents' desired change in school participation	School environment support	Community participation frequency	Community involvement	Parents' desired change in community participation	Community environment support
Average parent-reported educational hours/weekday
rho	0.531*	0.097	0.105	0.442	0.410	−0.216	−0.015	0.205	0.106	−0.211	−0.016	−0.147
p	.019	.693	.668	.058	.081	.374	.952	.399	.666	.385	.949	.549
Bootstrap a
Bias	−0.021	−0.018	−0.003	−0.011	−0.023	0.008	0.008	0.005	0.003	0.016	−0.007	0.013
SE	0.173	0.292	0.260	0.210	0.243	0.239	0.287	0.227	0.252	0.258	0.250	0.239
BCa 95% CI [LL, UL]	[0.151, 0.781]	[−0.614, 0.575]	[−0.379, 0.541]	[−0.007, 0.772]	[−0.118, 0.795]	[−0.698, 0.359]	[−0.537, 0.563]	[−0.257, 0.641]	[−0.405, 0.619]	[−0.707, 0.367]	[−0.474, 0.423]	[−0.590, 0.332]
Average parent-reported educational hours/weekend
rho	0.500*	0.464*	0.162	0.580**	0.216	−0.176	−0.058	0.128	−0.089	−0.186	0.204	−0.006
p	.029	.045	.507	.009	.375	.471	.814	.602	.717	.447	.401	.979
Bootstrap a
Bias	−0.011	−0.022	0	−0.020	−0.010	0.005	0.001	0.005	0.013	0.006	−0.013	0.012
SE	0.165	0.206	0.197	0.134	0.219	0.261	0.302	0.278	0.194	0.202	0.267	0.199
BCa 95% CI [LL, UL]	[0.158, 0.755]	[−0.057, 0.746]	[−0.273, 0.584]	[0.320, 0.777]	[−0.289, 0.646]	[−0.659, 0.447]	[−0.623, 0.479]	[−0.450, 0.614]	[−0.475, 0.339]	[−0.559, 0.191]	[−0.373, 0.641]	[−0.404, 0.389]
Average parent-reported social hours/weekday
rho	0.481*	0.37	−0.041	0.582**	0.196	−0.027	0.02	0.399	0.11	0.054	−0.184	0.169
p	.037	.119	.866	.009	.421	.914	.934	.090	.653	.828	.452	.490
Bootstrap a
Bias	−0.013	−0.011	0.003	−0.008	−0.01	0.002	0.007	−0.01	−0.005	0.007	0.006	0.002
SE	0.198	0.271	0.249	0.157	0.244	0.229	0.280	0.218	0.217	0.264	0.270	0.222
BCa 95% CI [LL, UL]	[0.008, 0.772]	[−0.308, 0.801]	[−0.487, 0.413]	[0.209, 0.831]	[−0.306, 0.629]	[−0.484, 0.448]	[−0.496, 0.602]	[−0.093, 0.765]	[−0.32−, 0.529]	[−0.476, 0.572]	[−0.655, 0.364]	[−0.257, 0.575]
Average parent-reported social hours/weekend
rho	0.367	0.058	0.460*	0.214	−0.001	−0.041	0.133	0.442	0.286	0.118	−0.221	0.172
p	.123	.814	.048	.378	.996	.868	.588	.058	.235	.629	.363	.482
Bootstrap a
Bias	−0.019	−0.002	−0.023	0.003	0.003	0.019	−0.018	−0.01	−0.004	0.004	0.005	0.006
SE	0.214	0.273	0.163	0.247	0.213	0.240	0.255	0.222	0.245	0.241	0.245	0.269
BCa 95% CI [LL, UL]	[−0.089, 0.674]	[−0.584, 0.578]	[0.123, 0.698]	[−0.316, 0.656]	[−0.409, 0.423]	[−0.566, 0.490]	[−0.341, 0.558]	[−0.148, 0.812]	[−0.304, 0.780]	[−0.468, 0.620]	[−0.675, 0.269]	[−0.337, 0.666]
Average parent-reported passive hours/weekend
rho	0.055	−0.183	0.495*	−0.311	0.32	−0.286	0.118	−0.433	0.372	−0.332	0.015	−0.318
p	.822	.453	.031	.195	.182	.236	.631	.064	.117	.164	.953	.185
Bootstrap a
Bias	−0.011	0.013	−0.018	0.005	−0.004	0.01	−0.006	0.002	−0.035	0.005	0.012	−0.003
SE	0.276	0.229	0.178	0.270	0.241	0.246	0.249	0.249	0.225	0.247	0.255	0.253
BCa 95% CI [LL, UL]	[−0.442, 0.547]	[−0.638, 0.309]	[0.133, 0.765]	[−0.747, 0.180]	[−0.209, 0.732]	[−0.675, 0.245]	[−0.384, 0.583]	[−0.831, 0.123]	[−0.165, 0.697]	[−0.744, 0.169]	[−0.504, 0.508]	[−0.752, 0.184]
Average child-reported educational hours/weekend
rho	0.277	0.198	−0.014	0.471*	0.278	0.094	−0.05	0.164	0.16	0.172	0.169	0.142
p	.251	.417	.955	.042	.250	.703	.838	.502	.513	.482	.489	.562
Bootstrap a
Bias	−0.002	−0.019	0.012	−0.010	−0.016	−0.007	0.009	0.005	0.010	−0.012	−0.013	0.005
SE	0.218	0.250	0.272	0.174	0.254	0.274	0.268	0.224	0.239	0.216	0.234	0.224
BCa 95% CI [LL, UL]	[−0.155, 0.675]	[−0.294, 0.602]	[−0.611, 0.621]	[0.087, 0.786]	[−0.280, 0.716]	[−0.534, 0.659]	[−0.509, 0.442]	[−0.281, 0.597]	[−0.320, 0.626]	[−0.234, 0.549]	[−0.364, 0.571]	[−0.334, 0.568]
Average child-reported social hours/weekday
rho	0.307	0.177	−0.109	0.547*	−0.235	−0.111	0.037	0.715**	0.334	−0.001	−0.332	0.459*
p	.201	.468	.656	.015	.332	.650	.882	.001	.163	.996	.165	.048
Bootstrap a
Bias	−0.007	−0.003	0.005	−0.016	0.012	0.015	−0.005	−0.031	−0.010	0.002	0.012	−0.012
SE	0.198	0.227	0.246	0.194	0.225	0.234	0.24	0.118	0.241	0.222	0.218	0.221
BCa 95% CI [LL, UL]	[−0.073, 0.625]	[−0.294, 0.614]	[−0.579, 0.426]	[0.145, 0.809]	[−0.607, 0.209]	[−0.548, 0.441]	[−0.452, 0.529]	[0.497, 0.828]	[−0.196, 0.766]	[−0.481, 0.494]	[−0.748, 0.176]	[0.012, 0.809]
Average child-reported social hours/weekend
rho	0.316	−0.323	0.299	0.213	0.276	−0.016	−0.250	0.325	0.468*	−0.306	−0.119	0.254
p	.188	.178	.213	.381	.253	.947	.302	.175	.043	.202	.628	.294
Bootstrap a
Bias	−0.006	0.003	−0.009	−0.005	0.005	0.018	−0.012	−0.006	−0.011	0.017	0	−0.004
SE	0.211	0.259	0.218	0.228	0.229	0.236	0.243	0.220	0.196	0.214	0.244	0.218
BCa 95% CI [LL, UL]	[−0.140, 0.699]	[−0.763, 0.178]	[−0.256, 0.690]	[−0.234, 0.587]	[−0.249, 0.699]	[−0.485, 0.499]	[−0.629, 0.176]	[−0.173, 0.739]	[−0.047, 0.822]	[−0.708, 0.223]	[−0.577, 0.397]	[−0.177, 0.636]
Average child-reported passive hours/weekday
rho	0.202	−0.279	0.117	0.334	0.475*	0.035	−0.323	0.103	0.231	0.144	0.094	0.068
p	.407	.248	.634	.162	.04	.887	.178	.675	.341	.555	.703	.782
Bootstrap a
Bias	0.002	0.002	0.004	−0.015	−0.017	0.009	0	0.009	0.011	0.003	−0.014	0.011
SE	0.232	0.212	0.241	0.233	0.194	0.25	0.225	0.250	0.227	0.242	0.266	0.256
BCa 95% CI [LL, UL]	[−0.302, 0.678]	[−0.656, 0.169]	[−0.466, 0.644]	[−0.087, 0.689]	[0.022, 0.778]	[−0.510, 0.592]	[−0.657, 0.062]	[−0.437, 0.634]	[−0.308, 0.692]	[−0.321, 0.580]	[−0.435, 0.581]	[−0.446, 0.569]
Average child-reported passive hours/weekend
rho	−0.114	−0.267	0.158	−0.245	0.443	0	−0.339	−0.496*	0.176	0.031	0.213	−0.356
P	.643	.268	.519	.312	.058	.999	.156	.031	.47	.901	.382	.134
Bootstrap a
Bias	0.01	−0.001	−0.003	−0.003	−0.016	−0.002	0.006	0.031	0	0	−0.011	0.023
SE	0.254	0.247	0.218	0.243	0.214	0.233	0.219	0.188	0.224	0.254	0.279	0.177
BCa 95% CI [LL, UL]	[−0.612, 0.465]	[−0.670, 0.221]	[−0.314, 0.597]	[−0.677, 0.193]	[−0.007, 0.778]	[−0.474, 0.457]	[−0.687, 0.143]	[−0.831, 0.030]	[−0.304, 0.596]	[−0.461, 0.504]	[−0.331, 0.719]	[−0.686, 0.115]

Note. CSTUR: Children’s Screen Time Use Report; PEM-CY: Participation and Environment Measure for Children and Youth; rho: Spearman’s correlation coefficient; p: significance (2-tailed); SE: standard error; BCa: bias-corrected and accelerated; CI: confidence interval; LL: lower limit; UL: upper limit.

*Correlation is significant at the 0.05 level (2-tailed). **Correlation is significant at the 0.01 level (2-tailed).

^aBootstrapping specifications: (a) sampling method: simple; (b) number of samples: 1000; (c) CI level: 95%; and (4) CI type: BCa.

Predictors of HRQoL and participation

The child- and parent/caregiver-reported types of ST were the independent variables whereas the KIDSCREEN-52 domains and PEM-CY subscales were the dependent variables. No statistically significant regression models were found with the PEM-CY subscales. Six regression models identified significant predictors of children’s HRQoL (as measured by the KIDSCREEN-52 domains). Table 4 reports all regression statistics.

Table 4.

Linear regression analyses between CSTUR scores and KIDSCREEN-52 and PEM-CY subscales (child: n = 29, parent: n = 20; bootstrapped sample of 1000).

	Before bootstrapping						After bootstrapping b
	B a	SE B	b	t	p	Bias	SE B	p	BCa 95% CI [LL, UL]
Dependent variable: KIDSCREEN-52 physical wellbeing
Independent variables c
Constant	21.014	0.833		25.236	.000	0.025	0.818	.001	[19.475, 22.671]
Average child-reported passive minutes/weekend	−0.021	0.008	−0.445	−2.583	.016	0.000	0.008	.012	[−0.036, −0.007]
Dependent variable: KIDSCREEN-52 moods and emotions
Independent variables d
Constant	30.348	0.738		41.125	.000	0.075	0.747	.001	[28.828, 32.005]
Average child-reported interactive minutes/weekday	−0.014	0.006	−0.394	−2.226	.035	−0.001	0.008	.065	[−0.030, −0.003]
Independent variables e
Constant	29.401	1.202		24.465	.000	0.039	1.133	.001	[26.711, 31.675]
Average parent-reported educational minutes/weekday	0.005	0.017	0.063	0.269	.791	−0.002	0.019	.792	[−0.044, 0.034]
Dependent variable: KIDSCREEN-52 autonomy
Independent variables f
Constant	20.283	1.042		19.473	.000	0.071	1.110	.001	[18.183, 22.470]
Average child-reported passive minutes/weekend	−0.026	0.010	−0.445	−2.579	.016	−0.001	0.011	.022	[−0.050, −0.008]
Dependent variable: KIDSCREEN-52 parent relations and home life
Independent variables g
Constant	24.999	0.711		35.138	.000	0.168	0.786	.001	[23.065, 26.961]
Average child-reported interactive minutes/weekend	−0.001	0.005	−0.047	−0.246	.807	−0.003	0.008	.855	[−0.018, 0.005]
Independent variable h
Constant	26.119	0.841		31.049	.000	0.091	0.879	.001	[24.280, 28.031]
Average parent-reported interactive minutes/weekend	−0.007	0.007	−0.239	−1.044	.310	−0.003	0.012	.488	[−0.034, 0.006]
Dependent variable: KIDSCREEN-52 financial resources
Independent variable i
Constant	11.257	0.765		14.721	.000	0.016	0.714	.001	[9.809, 12.606]
Average child-reported interactive minutes/weekday	−0.013	0.007	−0.365	−2.036	.052	0.000	0.005	.007	[−0.023, −0.005]
Independent variable j
Constant	12.060	0.908		13.282	.000	0.044	0.868	.001	[9.839, 14.160]
Average parent-reported interactive minutes/weekday	−0.023	0.011	−0.464	−2.224	.039	−0.001	0.008	.007	[−0.040, −0.013]
Dependent variable: KIDSCREEN-52 social support and peers
Independent variable k
Constant	26.453	1.201		22.022	.000	0.232	1.272	.001	[23.451, 30.027]
Average child-reported interactive minutes/weekday	−0.006	0.013	−0.148	−0.511	.614	−0.007	0.024	.735	[−0.058, 0.017]
Average child-reported interactive minutes/weekend	0.000	0.010	0.008	0.028	.978	0.004	0.016	.985	[−0.036, 0.043]
Average child-reported passive minutes/weekday	−0.004	0.025	−0.055	−0.172	.865	−0.005	0.024	.863	[−0.049, 0.028]
Average child-reported passive minutes/weekend	−0.029	0.019	−0.429	−1.541	.136	0.002	0.015	.060	[−0.055, 0.019]
Independent variable l
Constant	24.899	1.035		24.057	.000	0.168	1.065	.001	[22.689, 27.519]
Average parent-reported interactive minutes/weekday	−0.016	0.012	−0.307	−1.369	.188	−0.007	0.019	.243	[−0.054, −0.003]
Dependent variable: KIDSCREEN-52 school environment
Independent variable m
Constant	27.111	1.051		25.786	.000	0.359	1.228	.001	[24.956, 31.575]
Average child-reported interactive minutes/weekday	−0.008	0.011	−0.208	−0.747	.463	−0.007	0.021	.625	[−0.059, 0.010]
Average child-reported interactive minutes/weekend	−0.008	0.009	−0.252	−0.886	.384	0.008	0.018	.686	[−0.040, 0.087]
Average child-reported passive minutes/weekday	0.014	0.022	0.189	0.620	.541	−0.009	0.029	.656	[−0.039, 0.042]
Average child-reported passive minutes/weekend	−0.026	0.016	−0.432	−1.613	.120	0.000	0.021	.276	[−0.069, 0.023]
Independent variable n
Constant	26.579	0.865		30.727	.000	−0.028	0.856	.001	[24.800, 28.070]
Average parent-reported interactive minutes/weekday	−0.013	0.024	−0.270	−0.542	.595	−0.003	0.021	.437	[−0.061, 0.018]
Average parent-reported interactive minutes/weekend	−0.010	0.017	−0.309	−0.620	.544	0.002	0.018	.536	[−0.038, 0.028]
Dependent variable: KIDSCREEN-52 social acceptance (bullying)
Independent variable o
Constant	11.719	0.798		14.683	.000	−0.017	0.890	.001	[9.790, 13.419]
Average parent-reported passive minutes/weekend	0.014	0.007	0.447	2.120	.048	0.001	0.007	.042	[0.002, 0.028]
Dependent variable: PEM-CY home participation frequency
Independent variable p *
Constant	5.914	0.148		39.859	.000	−0.019	0.180	.001	[5.566, 6.190]
Average parent-reported educational minutes/weekday	0.005	0.006	0.228	0.847	.410	0.002	0.008	.360	[−0.005, 0.037]
Average parent-reported educational minutes/weekend	0.008	0.008	0.248	1.011	.327	−0.002	0.010	.104	[−0.014, 0.016]
Average parent-reported social minutes/weekday	0.001	0.002	0.154	0.628	.539	0.000	0.007	.232	[−0.020, 0.036]
Dependent variable: PEM-CY home involvement
Independent variable q **
Constant	4.040	0.108		37.404	.000	−0.002	0.105	.001	[3.822, 4.232]
Average parent-reported educational minutes/weekend	0.012	0.006	0.452	2.089	.052	0.001	0.006	.027	[−0.001, 0.029]
Dependent variable: PEM-CY home participation desired change
Independent variable r
Constant	26.087	8.615		3.028	.008	−1.439	8.135	.007	[12.126, 37.136]
Average parent-reported social minutes/weekend	0.105	0.099	0.232	1.055	.307	0.072	0.178	.317	[−0.034, 1.111]
Average parent-reported passive minutes/weekend	0.122	0.066	0.404	1.838	.085	0.004	0.047	.020	[0.010, 0.225]
Dependent variable: PEM-CY home environment support
Independent variable s
Constant	90.640	1.899		47.722	.000	−0.149	2.034	.001	[87.597, 93.760]
Average child-reported educational minutes/weekend	0.231	0.118	0.427	1.959	.067	0.025	0.236	.017	[0.001, 0.853]
Average child-reported social minutes/weekday	0.038	0.037	0.225	1.031	.317	0.013	0.053	.492	[−0.048, 0.224]
Independent variable t ***
Constant	91.862	1.649		55.708	.000	−0.058	1.750	.001	[88.205, 95.224]
Average parent-reported educational minutes/weekend	0.172	0.091	0.428	1.890	.076	−0.011	0.211	.053	[−0.001, 0.373]
Average parent-reported social minutes/weekday	−0.004	0.028	−0.035	−0.155	.879	0.036	0.112	.896	[−0.044, 0.363]
Dependent variable: PEM-CY school participation frequency
Independent variable u
Constant	4.522	0.294		15.371	.000	0.024	0.248	.001	[4.050, 5.063]
Average child-reported passive minutes/weekday	0.006	0.004	0.343	1.548	.139	0.000	0.004	.118	[−0.003, 0.014]
Dependent variable: PEM-CY school environmental support
Independent variable v
Constant	97.991	2.169		45.186	.000	−0.541	2.297	.001	[93.601, 100.888]
Average child-reported social minutes/weekday	−0.014	0.030	−0.105	−0.470	.644	0.016	0.057	.780	[−0.081, 0.135]
Average child-reported passive minutes/weekend	−0.035	0.019	−0.415	−1.860	.080	0.003	0.016	.058	[−0.087, 0.006]
Dependent variable: PEM-CY community participation frequency
Independent variable w
Constant	3.572	0.241		14.838	.000	−0.001	0.242	.001	[3.044, 4.046]
Average child-reported social minutes/weekend	0.005	0.005	0.240	1.050	.308	0.001	0.008	.179	[−0.002, 0.016]
Dependent variable: PEM-CY community environmental support
Independent variable x
Constant	94.908	1.656		57.328	.000	−0.230	1.710	.001	[91.083, 97.770]
Average child-reported social minutes/weekday	0.003	0.035	0.023	0.096	.924	0.018	0.057	.932	[−0.068, 0.188]

Note. CSTUR: Children’s Screen Time Use Report; PEM-CY: Participation and Environment Measure for Children and Youth; B: unstandardized beta coefficient; SE B: standard error for the unstandardized beta; b: standardized beta; t: the t-test statistic; p: Probability statistic; CI: confidence interval; LL: lower limit; UL: upper limit; BCa: bias-corrected and accelerated.

*Bootstrapping specifications: (a) sampling method: simple; (b) number of samples: 993; (c) CI level: 95%; and (d) CI type: BCa.

**Bootstrapping specifications: (a) sampling method: simple; (b) number of samples: 990; (c) CI level: 95%; and (d) CI type: percentile.

***Bootstrapping specifications: (a) sampling method: simple; (b) number of samples: 991; (c) CI level: 95%; and (d) CI type: BCa.

^aB remained unchanged after bootstrapping.

^bBootstrapping specifications: (a) sampling method: simple; (b) number of samples: 1000; (c) CI level: 95%; and (d) CI type: BCa.

^cR = 0.445; R² = 0.198; adjusted R² = 0.168; p = .016.

^dR = 0.394; R² = 0.155; adjusted R² = 0.124; p = .035.

^eR = 0.063; R² = 0.004; adjusted R² = −0.051; p = .791.

^fR = 0.445; R² = 0.198; adjusted R² = 0.168; p = .016.

^gR = 0.047; R² = 0.002; adjusted R² = −0.035; p = .807.

^hR = 0.239; R² = 0.057; adjusted R² = 0.005; p = .310.

ⁱR = 0.365; R² = 0.133; adjusted R² = 0.101; p = .052.

^jR = 0.464; R² = 0.216; adjusted R² = 0.172; p = .039.

^kR = 0.552; R² = 0.305; adjusted R² = 0.189; p = .059.

^lR = 0.307; R² = 0.094; adjusted R² = 0.044; p = .188.

^mR = 0.598; R² = 0.357; adjusted R² = 0.250; p = .026.

ⁿR = 0.567; R² = 0.321; adjusted R² = 0.241; p = .037.

^oR = 0.447; R² = 0.200; adjusted R² = 0.155; p = .048.

^pR = 0.490; R² = 0.240; adjusted R² = 0.098; p = .210.

^qR = 0.452; R² = 0.204; adjusted R² = 0.158; p = .052.

^rR = 0.483; R² = 0.233; adjusted R² = 0.138; p = .119.

^sR = 0.455; R² = 0.207; adjusted R² = 0.114; p = .139.

^tR = 0.421; R² = 0.177; adjusted R² = 0.080; p = .191.

^u R = 0.343; R² = 0.118; adjusted R² = 0.069; p = .139.

^vR = 0.413; R² = 0.171; adjusted R² = 0.073; p = .204.

^wR = 0.240; R² = 0.058; adjusted R² = 0.005; p = .308.

^xR = 0.023; R² = 0.001; adjusted R² = −0.055; p = .924.

Physical wellbeing

The model comprised one independent variable: child-reported passive ST on a weekend, that explained 16.8% of the KIDSCREEN-52 physical wellbeing domain’s variance (adjusted R ² = 0.168, F (1, 27) = 6.671, p = .016, BCa 95% CI [19.475, 22.671]).

Autonomy

The model comprised one independent variable: child-reported passive ST on a weekend, that explained 16.8% of the KIDSCREEN-52 autonomy domain’s variance (adjusted R ² = 0.168, F (1, 27) = 6.654, p = .016, BCa 95% CI [18.183, 22.470]).

Financial resources

The model comprised one independent variable: parent/caregiver-reported interactive ST on a weekday, that explained 17.2% of the KIDSCREEN-52 financial resources domain’s variance (adjusted R ² = 0.172, F (1, 18) = 4.945, p = .039, BCa 95% CI [9.839, 14.160]).

School environment

The regression model comprised four independent variables: child-reported interactive ST on a weekday, interactive ST on a weekend, passive ST on a weekday and passive ST on a weekend. The model uniquely explained 25% of the KIDSCREEN-52 school environment score’s variance (adjusted R ² = 0.250, F (4, 24) = 3.336, p = .026, BCa 95% CI [24.956, 31.575]), based on 1000 bootstrap samples. It should be noted, however, that none of the independent variables made a unique individual statistically significant contribution to the overall model variance. For parent/caregiver-reported ST scores, the model comprised two independent variables: parent/caregiver-reported interactive ST on a weekday and interactive ST on a weekend. The model accounted for 24.1% of the KIDSCREEN-52 school environment score’s variance (adjusted R ² = 0.241, F (2, 17) = 4.019, p = .037, BCa 95% CI [24.800, 28.070]), based on 1000 bootstrap samples. The independent variables were not statistically significant.

Social acceptance

The model comprised one independent variable: parent/caregiver-reported passive ST on a weekend, that accounted for 15.5% of the KIDSCREEN-52 social acceptance score’s variance (adjusted R ² = 0.155, F (1, 18) = 4.495, p = .048, BCa 95% CI [9.790, 13.419]).

Participation

No statistically significant models were found reporting any type of ST as predictive of participation.

Discussion

Our study investigated the associations between different types of ST with HRQoL and participation in children. We found that all ST types were significantly correlated with different aspects of children’s HRQoL and participation. The main trends we noticed were (i) passive ST was a common predictor of physical wellbeing, autonomy and social acceptance and (ii) educational and social ST positively correlated with children’s home, school and community participation.

Passive ST as a common predictor of physical wellbeing, autonomy and social acceptance

Passive ST on a weekend was the highest reported ST for children and their parents in our study. Primary school-aged children typically engage in more passive ST, such as television-viewing, compared to adolescents (Brown and Bobkowski, 2011; Sanders et al., 2019). On weekends, passive ST may increase as children have more time to watch weekly TV series, movies and cartoons which is otherwise regulated during the school week by parents/caregivers and teachers. Our results found that passive ST on the weekend was a common predictor of multiple KIDSCREEN-52 HRQoL domains.

Firstly, higher passive ST on the weekend predicted lower scores in the KIDSCREEN-52 physical wellbeing domain. The KIDSCREEN-52 physical wellbeing scale rates children’s fitness, energy and physical activity level (Ravens-Sieberer et al., 2008). Our findings were similar to current research trends. For example, television watching is frequently reported to correlate with obesity and lower physical wellbeing in children (Stiglic and Viner, 2019), while screen time behaviours such as television-viewing have been associated with higher sweet and snack intake (Cartanyà-Hueso et al., 2021). Furthermore, passive ST such as television encourages children to be sedentary for extended periods and is related to reduced physical activity (McVeigh et al., 2016; Zhang et al., 2017). Our study contributes to the existing research literature highlighting the potential relationship between passive ST and lower physical wellbeing.

We also found that passive ST on the weekend was negatively correlated with and predictive of the KIDSCREEN-52 autonomy domain. The KIDSCREEN-52 autonomy scale measures children’s free time and opportunities to engage in desired activities (Ravens-Sieberer et al., 2008). Our findings differ from studies investigating similarly aged samples. Sebire and Jago’s (2013) study investigating a sample of 9–10 years old (n = 874) did not find any correlations between child autonomy and ST; however, ‘child agency’, or self-direction and confidence in children, was associated with higher television watching. Higher child agency may indicate reduced parental structures and rules around screen time, leading to increased television-viewing time (Sebire and Jago, 2013). Our divergent findings may be due to differences between child agency and autonomy (Sebire and Jago, 2013), two constructs that were not distinguished in our investigation of autonomy using the KIDSCREEN-52.

The KIDSCREEN-52 autonomy scale not only measures children’s choice in recreational activities but also asks children to rate their opportunity to shape personal free time (Ravens-Sieberer et al., 2008). Financial, personal and environmental factors may moderate opportunities. When children do not have the opportunity to pursue particular hobbies or meet with friends, they may gravitate to passive ST activities due to their accessibility and convenience. De Decker et al.’s (2012) qualitative study found that pre-school children frequently watch television out of boredom. In addition, parental television-viewing habits at home may also influence children’s behaviour (De Decker et al., 2012). Growing up with easy access to passive media and parental modelling may be influencing higher ST’s correlation with lower autonomy.

Promoting autonomy in children can help foster reduced assistance when participating in everyday occupations (Rosenberg et al., 2019). Occupational therapists should encourage children to pursue meaningful extracurricular activities to promote autonomy and lower passive ST behaviour. This may positively contribute to children’s occupational performance and participation (Law et al., 1996).

A notable finding was that ST on the weekend was positively correlated with and predictive of the KIDSCREEN-52 social acceptance domain. Social acceptance was measured by children’s rating of how often other children make fun of or bully them at school (Ravens-Sieberer et al., 2008). Our findings contribute to the growing understanding of television watching and social connection with peers. There is little research investigating ST and social acceptance in primary school-aged children; however, Brown and Bobkowski (2011) suggested that ST – and subsequently, online social media – provided adolescents with more socially connective opportunities. For instance, watching television shows can help adolescents connect with other fans (Brown and Bobkowski, 2011). As such, keeping up to date with popular shows may support children to build positive relationships with peers at school since it provides a point of commonality and connection. Similarly, passive media such as television and music drives and supports identity exploration in adolescents (Brown and Bobkowski, 2011), which may build confidence and support social acceptance among peers.

As health professionals, occupational therapists should continue to promote moderation in television watching among children to support healthy physical wellbeing and autonomy. However, we should also understand the potential benefits of passive ST to make informed contributions to our paediatric clients and support them to lead meaningful lives.

Impact of screen time on home, school and community participation

Participating in a wide range of activities may indicate successful occupational performance and participation in children and contribute positively to skill development (Rosenberg et al., 2019). The PEM-CY measures participation in various home activities including play, hobbies, chores, homework and family interactions (Coster et al., 2011). Our study found that higher educational ST was significantly correlated with higher frequency participation in home activities. There is scant research surrounding educational-based ST and its effect on participation. Hietajärvi et al. (2019) investigated ‘knowledge-oriented’ digital participation, which they defined as learning, discussing or creating new information about hobbies and things of the child’s interest. Their study determined that younger children participated in knowledge-oriented activities to share, follow and update others about their hobbies (Hietajärvi et al., 2019). In other words, knowledge-oriented ST is related to interest-driven behaviour in children (Hietajärvi et al., 2019). Educational ST may expose children to more information about hobbies and skills, promoting engagement in play and hobbies within the home environment. This is a potential explanation for the positive correlation we found in our study. Our study findings imply that educational ST does not adversely affect wellbeing and may even increase participation in home activities.

Weekday social ST was positively and significantly correlated with higher frequency participation in home activities in our study. On the weekend, social ST was positively correlated with higher frequency participation in community activities. As measured by the PEM-CY, this could include neighbourhood outings, events, physical activities and getting together with other children (Coster et al., 2011). Fairlie and Kalil (2017) also reported positive relationships between ST and social participation in their randomized control trial. In their study, the treatment group (n = 559, grade 6–10) received computers while the control group did not own any home computers (Fairlie and Kalil, 2017). They found the treatment group engaged in more social participating behaviours, including social networking and email usage (Fairlie and Kalil, 2017). Furthermore, the treatment group communicated with 1.57 more friends over 0.72 more hours in real life compared to the group without a computer (Fairlie and Kalil, 2017). Their study provided strong evidence that social ST can positively impact real-world social participation among children. As children spend more time with friends, this may indicate more participation in activities within the home and the community. Furthermore, social ST may facilitate easier organization of group activities by allowing children to communicate with their friends outside the school environment. Our results support Fairlie and Kalil’s (2017) study within an Australian context, with our findings implying that social ST can promote social development in younger children, which may correlate with increased participation in home and community activities.

Finally, we found that passive ST was significantly and positively correlated with increased participation in school-related activities. School activities measured on the PEM-CY include classroom and extracurricular activities, excursions, clubs and playing with peers (Coster et al., 2011). Parents have noted that television watching can help with language development, learning about nature and health and developing children’s imagination and curiosity (De Decker et al., 2012). These educational skills derived from passive ST at home may translate to the school environment and increased language skills may support participation in classroom activities. Developing strong imagination and creativity skills in children may also support play with peers, while additional exposure to documentaries or cartoons may promote interest-driven participation in school clubs (Hietajärvi et al., 2019). However, we did not observe correlations between educational ST and school participation, despite studies indicating a positive relationship between educational ST and better academic outcomes (Sanders et al., 2019). This is especially interesting given the study was conducted during the coronavirus pandemic, as educational ST played a pivotal role in supporting children’s continued participation in various school-related activities.

Implications for occupational therapy

It is evident that ST plays a significant role in influencing children’s HRQoL and participation. As occupational therapists have a unique role in fostering wellbeing through successful participation in meaningful activities, understanding the benefits and negatives of ST in school-aged children will help them to make informed decisions regarding children’s occupations and time-use as they grow up in the digital age. Furthermore, occupational therapists can guide parents to make informed decisions on the extent of children’s exposure to ST during their formative years that may impact positively children’s health and participation in meaningful occupations. Therapists could consider interventions based on children’s time-use, balancing time between children’s physical activity and ST, and promoting activity diversity amongst children’s daily occupational performance. Likewise, occupational therapists could collaborate with educators and other health providers in promoting the health, developmental and learning benefits of increased physical activity and moderating ST. It is recommended that these types of programs and interventions be formally evaluated to determine their effectiveness and efficacy.

Limitations

Our study is the first in Australia to examine the extent of the relationship between different types of ST, HRQoL and participation among typically developing school-aged children. However, this study does have acknowledged limitations. Firstly, the scope of our study is limited due to the participant recruitment method and small sample size. Most participants were recruited from the same metropolitan primary school, thereby limiting our sample’s sociodemographic and geographic variance. Although bootstrapping was utilized, the original sample size of 29 limits the generalizability of the study’s results to a larger population. Secondly, the study utilized three self-report measures, so social desirability bias is a further potential limitation. Thirdly, the difference in the number of child and parent/caregiver respondents (e.g. n = 29 versus n = 20) is an acknowledged limitation but given that participants took part in the study voluntarily, the wishes of the non-respondents had to be respected. Finally, given the study’s cross-sectional design, HRQoL and participation may play a moderating role on children’s ST, rather than the other way around.

Future research

Future research is warranted to investigate the relationship between ST, HRQoL and participation in a larger, nationally representative sample of Australian children. Our study looked at a wide range of participation in many settings. Future studies could investigate associations between ST and participation in specific childhood occupations, such as play, in more depth. Finally, we examined the impact of four different types of ST in children aged 8–14 years. As children get older, it is expected that commitments to social and educational ST will increase (Sanders et al., 2019). As such, this study may be re-conducted with an older age group of children to see how the impact of ST may change over time.

Conclusion

The study findings demonstrate that interactive and passive ST can negatively affect and are predictive of school-aged children’s HRQoL. However, educational, social and passive ST can also promote participation. These findings contribute to the growing literature on ST, HRQoL and participation in Australian settings. Occupational therapists should understand both the benefits and drawbacks of integrating ST into children’s lives to make informed decisions about their occupational participation and time-use as they grow up in the technological age.

Key findings

• • Passive ST was predictive of lower physical wellbeing, autonomy and wellbeing at school and higher social acceptance.

• • Educational and social ST positively correlate with increased participation in home, school and community activities.

What the study has added

• • Screen time types are differentially associated with quality of life and participation. Occupational therapists are encouraged to understand how different screen time types impact children’s activity engagement and wellbeing.

ORCID iDs

Ted Brown https://orcid.org/0000-0001-9403-5877

Mong-Lin Yu https://orcid.org/0000-0003-1662-5008