Results
Descriptive Data
The data collected included information relating to the specific ways in which young people reported engaging in SBMU. Here, the young people were asked to say whether they did or did not engage in each of 20 different screen based activities. As shown in Table 1, the most popular activity was watching TV (94%) and listening to music or watching videos (92%). The least reported activities included Twitter (8%), MMORGs (21%) and online shopping (21%).
While the main analyses below related to the presence or absence of problematic SBMU, we also present the means and standard deviations for the different forms of SBMU (see Table 2). Given the degree to which these variables were skewed, we did not conduct ANOVAs to look for effects of sex and grade, instead leaving investigation of these effects until our analyses below.
The durations to which children and adolescents reported SBMU during a typical weekday, in the light of AAP (and Australian Department of Health) recommendations is presented. This was investigated first for their general estimate of daily SBMU, and subsequently for SBMU pertaining to each group of screen based activities.
Overall Screen use: Within AAP Recommendations (≤2 Hours) or in Excess (>2 Hours)
The number and percentage of participants engaging in SBMU for > 2 hours on a typical weekday, by Sex, Grade, and Screen Activity, is shown in Table 3. On a typical weekday, 62.7% of participants exceeded the < 2 hours recommendation. Approximately 45% of the youngest participants (i.e., Year 3, age 8 years) exceeded the < 2 hours recommendation, rising to 80% by Year 9 (15–16 years of age). The most popular screen of use was TV, with almost 90% reporting watching TV in the week prior to the survey. This was followed by laptop (59%), iPad/Tablet (58%) and mobile phone (57%) use. A three-way frequency analysis was performed to develop a hierarchical linear model of Sex, Grade (3, 5, 7, 9), and overall SBMU (<2 hours, > 2 hours). Backward elimination produced a model that included only the three-way interaction effect, χ (3, N = 2515) = 56.18, p < .001. At Grade 3, there was no relationship between Sex and SBMU. At Grade 5, the relative odds ratio (Field, 2009) indicated that girls were 1.42 times more likely than boys to be in the > 2 hours group. At Grade 7, this trend was accentuated, with girls 3.71 times more likely to be in the > 2 hours group. Finally, this trend developed further in Grade 9 with girls 4.95 times more likely to be in the > 2 hours group than boys. The increasing sex disparity was evident despite the overall trend for increasing SBMU across both boys and girls (see Table 1).
Specific Screen use Group Activities: Within AAP Recommendations (<2 Hours) or in Excess (>2 Hours)
Next, we examined the extent to which each form of SBMU exceeded the recommendation of < 2 hours. To achieve this, we conducted four further hierarchical linear models, each with Sex, Grade (3, 5, 7, 9), and specific form of SBMU (<2 hours, > 2 hours) included.
For Gaming, the final model had a likelihood ratio χ (6, N = 2349) = 4.35, p = .629, indicating a good fit between the observed and predicted frequencies. There was no interaction between Sex and Grade, but there was a main effect of Sex: boys were 1.75 times more likely than girls to exceed the < 2 hour recommendation for gaming. There was also a main effect of Grade: Grade 3 children were approximately 1.4 times more likely than Grades 5 and 7 (relative odds 1.38 and 1.42 respectively), and 1.93 times more likely than Grade 9, to exceed the < 2 hour recommendation for gaming. Grades 5 and 7 did not differ from each other, and both these grades were more likely than Grade 9 (relative odds 1.39 and 1.36 respectively) to exceed the < 2 hour recommendation (see Table 3).
For Social Networking, the best fitting model included only the three-way interaction effect, χ (3, N = 2357) = 25.85, p < .001. At Grade 3, boys were 2.17 times more likely than girls to exceed the < 2 hour recommendation using social networking. However, at Grade 5 girls were 1.75 times more likely to exceed the < 2 hour recommendation using social networking, and this increased at Grade 7 (2.40 times more likely than boys) and again at Grade 9 (2.76 times more likely than boys). So, as was the case for overall SBMU, social networking increased with age, but increased much more rapidly for girls. Specifically, Grade 9 girls were 15.67 times more likely to exceed the < 2 hour recommendation using social networking when compared to Grade 3 girls, while the equivalent statistic for boys was 2.61 (see Table 3).
For Web Use, the best fitting model again included only the three-way interaction effect, χ (3, N = 2395) = 16.80, p = .001. At Grade 3, there was no association between Sex and SBMU relating to the Web. However, by Grade 5, girls were 1.63 times more likely to exceed the < 2 hour recommendation using the Web, and this increased at Grade 7 (2.40 times more likely than boys) and then remained at that level at Grade 9 (2.39 times more likely than boys). These effects seem to be due to a dip in boys' Web use at Grade 5 before returning to the same level again by Grade 9, whereas girls' Web use stayed the same at Grade 5 but increased thereafter (see Table 3).
With respect to TV/DVD/Movies, the final model had a likelihood ratio χ (6, N = 2419) = 9.365, p = .154. There was no interaction between Sex and Grade, but there was a main effect of Sex: girls were 1.44 times more likely than were boys to exceed the < 2 hour recommendation using TV/DVD/Movies. There was also a main effect of Grade. Grades 3 and 5 did not differ in their SBMU for this purpose, but Grade 7 were more likely than both Grades 3 and 5 (relative odds 1.24 and 1.30 respectively) to exceed the < 2 hour recommendation using TV/DVD/Movies. Grade 9 were slightly more likely to use screens for watching TV/DVD/Movies than were Grade 7 (relative odds of 1.18), but were more noticeably different from Grades 3 and 5 (relative odds 1.46 and 1.53 respectively) (see Table 3).
Multiple Screen use: Within AAP Recommendations (<2 Hours) or in Excess (>2 Hours)
Finally, in recognition of the fact that children and young people multi-task in their SBMU, we investigated the degree to which young people reported exceeding SBMU recommendations on more than one activity (Social Networking, Gaming, Web Use, TV/DVD/Movies). Our data indicated that 4.3% exceeded the < 2 hour recommendation on all four separate screen activities, 10.0% on three, 16.6% on two, and 24.6% on one. Under half (44.5%) did not exceed the < 2 hour recommendations on any of the four screen activities. This is higher than the single-item global estimate of SBMU reported above, which indicated that 37.3% of young people did not exceed the < 2 hour recommendations. Using this measure (number of screen activities where the < 2 hours recommendation is exceeded) as a dependent variable, we conducted a two-way independent ANOVA, with Sex (male, female) and Grade (Grades 3, 5, 7 and 9) as the independent variables. This revealed a small interaction between Sex and Grade: F (3, 2589) = 8.32, p < .001, ηp = .01. Boys in Grade 3 reported exceeding the < 2 hours recommendation on a greater number of activities than girls in Grade 3. By Grade 5, girls' scores on this measure marginally exceeded those of boys and in Grades 7 and 9 girls' scores clearly exceeded boys' (see Table 4).