No document available.
Abstract :
[en] The increasingly popularity of social media among young people have led to concerns and public debate about potential negative effects of social media. Given this popularity, it is imperative that risks associated with social media use are investigated. This dissertation aims to increase the knowledge about two specific risks associated with the social media use of young people: problematic social media use and exposure to sexist social media content. An integrative and interdisciplinary theoretical framework is used, including the Differential Susceptibility to Media Effects Model and the Bio-Ecological Model (Bronfenbrenner, 1977; Valkenburg & Peter, 2013). This theoretical framework guides the understanding between media and non-media variables and the diversity of factors related to the development of young people and media effects. A deeper understanding of social media related risks is gained by using a variety of different research methodologies for quantitative data analysis, such as an online study with a between- subjects design and specification curve analysis. The dissertation consists of five chapters: an overview of the theoretical framework and important definitions and concepts in Chapter 1, three empirical studies in Chapters 2 to 4, and a discussion of the work overall in Chapter 5. Chapters 2 and 3 focus on problematic social media use (PSMU) defined as compulsive social media use with addiction-like symptoms. Chapter 4 focuses on another facet of social media use that can be considered problematic: sexist social media content. Overall, the present studies show that a diverse range of predictors and detrimental health behaviours are associated with PSMU in adolescents and that benevolent and hostile sexist social media content influences young people on an emotional and cognitive level. Review of the literature indicated a knowledge gap regarding predictors for problematic social media use (PSMU), which are at present insufficiently understood (Schivinski et al., 2020; Stockdale & Coyne, 2020). Accordingly, the first study (Chapter 2) aimed to identify predictors of PSMU in a nationally representative sample of adolescents in Luxembourg using data from the Health Behaviour in School-aged Children (HBSC) study. Based on previous research, the Differential Susceptibility to Media Effects Model (DSMM) and the Bio-Ecological Model (Bronfenbrenner, 1977; Valkenburg & Peter, 2013), various predictors of PSMU were selected related to sociodemographics, social support, well-being and media use. Hierarchical multiple regression was used to analyse the data. The predictors included in the final model explained 22.3% of the variance in PSMU. The results indicate that cyberbullying perpetration, stress, psychosomatic health complaints, a preference for online social interaction and intensity of electronic media communication are important factors to consider in the development of PSMU. The findings of this study were coherent with the DSMM and the Bio-Ecological Model and underline the diverse range of factors that are linked with the development of PSMU, and the complexity of this concept.
Literature review indicated that the relationship between PSMU and adolescent health behaviour has remained largely unexplored. Apart from this, there has been discussion in the scientific community about the analytical flexibility of researchers, and its impact on obtained empirical results. Accordingly, the second study (Chapter 3) aimed to get insight in the relationship between PSMU and adolescent health behaviours by means of an exploratory specification curve analysis (SCA) approach. This approach applies SCA in an exploratory manner and - in contrast to the original method proposal - explicitly includes nonequivalent specifications. Using the 2018 international data from the HBSC study (i.e., 188,175 adolescents aged 11 - 15 years from 43 countries), three SCAs were conducted, with different types of health behaviour (i.e., substance use, physical activity, and dietary intake) as respective outcomes. Investigated specifications included different operationalisations of the dependent variables, inclusion of potential confounders (i.e., age, relative family affluence), and country as well as gender subsets. The SCAs indicated that the data analytical choices have an impact on the obtained results. Most specifications resulted in a significant relationship between PSMU and the different forms of health behaviour (i.e., substance use: 97% significant positive, physical activity: 78% significant negative, dietary intake: 92% significant negative). Overall, the results suggest that PSMU is correlated with detrimental adolescent health behaviour. Breakdown of the variance components in the results indicated different patterns per health behaviour investigated. The combination of country subsets and outcome variable explained a considerable proportion of the variance in the results of the three SCAs. The finding that PSMU is associated with detrimental health behaviours in adolescence is of high relevance, as during adolescence the foundation for health in adulthood is laid.
Online sexism is one example of antisocial behaviour that is increasingly prevalent in online spheres such as social media (Drakett et al., 2018; Fox et al., 2015; Paciello et al., 2021). Responses to media vary as a result of differences between individuals, and differences in media content (Fikkers & Piotrowski, 2020; Valkenburg & Peter, 2013). There is at present limited knowledge about which media content leads to which type of responses (Valkenburg et al., 2022; Valkenburg & Peter, 2013). In the third study (Chapter 4), the emotional and cognitive responses to sexist social media content were investigated. In addition, the influence of selected individual-difference variables on the responses were investigated, such as the emotional connection to social media use and well-being. In an online survey based on a between-subjects design, 419 young people aged 18 to 35 years were shown social media content that was either non-sexist, benevolent sexist, or hostile sexist. Content had a significant effect on negative affect and on the appraisal of the posts as interesting, appealing, and offensive, but not on positive affect or cognitive effort. Rather than the time spent online, the emotional connection to social media use and integration of social media use into routines were identified as potential differential- susceptibility variables. Overall, the results are in line with the propositions of the DSMM (Valkenburg & Peter, 2013).
By investigating two specific risks associated with the social media use of young people, this dissertation found that a multitude of factors are related to PSMU, media responses and effects, which shows the underlying complexity of these concepts. PSMU is not the result of one or two factors but is associated with a diverse range of predictors. PSMU is consistently associated with multiple detrimental health behaviours in adolescents. Additionally, benevolent and hostile sexist social media content has an impact on the emotional and cognitive response states of young people. The insights gained from these studies increase the understanding of the relationships between media and non-media variables and could be used in efforts to prevent PSMU and its effects.