Technology Behaviors, Parent Involvement, and Health Outcomes in Adolescents: A Latent Class Analysis

• In Event: Shaping Early Adolescents’ Digital Media Use: Diversity in Philosophies, Approaches, and Links with Psychosocial Outcomes

Abstract

Nearly all teens (95%) report smartphone ownership, and many (45%) describe that they are online "almost constantly" (Anderson, 2018). Technology use has been associated with benefits such as improved social support (Vannucci, 2019) as well as risks such as poor sleep (Demirici 2015). Key considerations in the relationship between technology use and health/wellness outcomes include technology behaviors (Kross, 2016), and parent involvement, including rule-setting (Chen, 2019). These factors may have varied patterns across individual adolescents. Therefore, the purpose of this study was to leverage Latent Class Analysis (LCA) to develop profiles of 1) technology behaviors, 2) parent involvement and rules, and 3) health and wellness outcomes among adolescents.

Early adolescents (aged 13-18) and one parent were recruited to a cross-sectional online survey using the Qualtrics platform and panels. Our 3,970 participants had a mean age of 15.0 years (SD=1.4). Among them, 46.2% were female, 67.8% were Caucasian, and 75.3% lived in a household with income above the poverty line. Technology behavior measures included device ownership (e.g., smartphone, VR, video games) and adolescents' perceived importance of technology behaviors (such as tagging photos, identity expression, and peer communication). Parent involvement measures included household technology rules and quality of parent-child relationship/communication. Health measures included physical/mental health (e.g., sleep, problematic internet use, depression, anxiety), social health (e.g., extracurricular activities, social connection) and wellness (e.g., empathy, loneliness). Using these variables, the LCA created a series of candidate models, each of which divided study participants into a discrete number of mutually exclusive classes that were then evaluated and compared. The Lo–Mendell–Rubin likelihood ratio test was used to identify the optimal number of classes.

The LCA identified two distinct classes: 1) Class 1: Participants were more likely to report personal technology ownership across all types of devices and to consistently rate their technology behaviors as very important. Class 1 participants reported either no parental technology rules, or strict rules based on media time (62.8% vs. 54.3%, p<.001). Participants had higher rates of concern and risk across all health and wellness measures. We labeled this class as “at-risk” participants. 2) Class 2: Participants were more likely to report family ownership of devices. Class 2 participants were also more likely to report parental rules about media content (66.1% vs. 27.2%, p<.001), and frequent communication with parents about technology (74.5% vs. 47.5%, p<.001). Class 2 participants had consistently lower risks across all health/wellness measures. We labeled this class as “not at risk.”

The LCA process identified two distinct profiles of adolescents’ technology behaviors and parent involvement, with starkly different health/wellness outcomes. Parental involvement via communication and content discussions with teens was associated with lower rates of health/wellness risks, while an absence of parental rules and/or parental rules focused on amount of media time were associated with health/wellness risks. Further study is warranted to identify optimal parent education strategies for preventing negative outcomes from technology use.

Authors

• Kole Binger, University of Wisconsin
  Presenting Author

• Qianqian Zhao, University of Wisconsin
  Non-Presenting Author

• Jens Eickhoff, University of Wisconsin
  Non-Presenting Author

• Megan A. Moreno, University of Wisconsin - Madison
  Non-Presenting Author