10th Biennial ACSPRI Social Science Methodology Conference in conjunction with the Australian Social Network Analysis Conference (ASNAC)

Background: Children's screen time has been shown to vary by family structure, but evidence remains cross-sectional, making it difficult to determine whether family structure drives these differences or whether pre-existing sociodemographic inequalities do. No study has examined how screen time changes following family transitions (from two-parent to single-parent and vice versa).
Research question: What is the effect of family structure and its transition on children's screen time? Does a transition in family structure cause a change in children's screen time?
Design, Setting, and Participants: This pre-registered study (OSF: https://doi.org/10.17605/OSF.IO/U3JB4) used data from the Longitudinal Study of Australian Children (LSAC), a nationally representative cohort study that has followed children since 2003 with biennial data collection. The analytic sample comprised children aged approximately 6–13 years, drawn from the Baby cohort (waves 4–7) and the Kindergarten cohort (waves 2–5). Missing data were handled using multiple imputation via predictive mean matching (mice package). Associations between family structure, number of siblings, and children's screen time were examined using multilevel mixed-effects models, with observations nested within children and children nested within cohorts, adjusted for sociodemographic covariates. Because children who experience household transitions may differ systematically from those in stable households, propensity score matching was used to match transitioned children with comparable peers in stable households, based on age, sex, socioeconomic position (SEIFA), Indigenous status, number of siblings, and home language. Using these matched samples, piecewise linear mixed-effects models were then estimated separately for single-parent-to-two-parent and two-parent-to-single-parent transitions, to assess whether screen time changed following family structure transitions.
Main Outcomes and Measures: Children's weekly screen time (total minutes per week) was assessed via parent report and summed across three domains: television viewing, video game use, and computer or other electronic media use.
Exposure Variable: Family structure was defined by the number of resident parental figures (biological, adoptive, or foster) responsible for caregiving, classified as two-parent (≥2) or single-parent (1). Family transitions were identified by comparing family-type classification across consecutive waves.
Results: A total of 9043 children were included (mean age, 6.86 years; 51% male; 85% in two-parent households). Living in a single-parent household was strongly associated with higher screen time before adjustment (β = 89.80 min/wk; 95% CI, 69.28–110.32) but was not significant after full adjustment for sociodemographic factors (β = 11.98 min/wk; 95% CI, −7.41 to 31.37), with maternal education accounting for most of the attenuation. Each additional sibling was independently associated with lower screen time (β = −22.71 min/wk; 95% CI, −30.18 to −15.25). Family transitions showed no significant immediate effect on screen time (two-to-single: β = 18.45 min/wk; 95% CI, −94.48 to 131.39; single-to-two: β = −49.47 min/wk; 95% CI, −247.58 to 148.64), with parallel trajectories before and after transition.
Conclusions and Relevance: The association between family structure and children's screen time is largely explained by sociodemographic factors. Transitions in family structure do not produce significant changes in screen time at the time of transition. Interventions to reduce screen time need not target family structure or its transitions; rather, they should address sociodemographic inequalities and account for the role of siblings.