Online sexual harassment in adolescence: definition, measurement, and comparison with other forms of online peer victimization

Sexual harassment, with its multiple dimensions and definitions, is a complex behavior to study. It is important to fully understand the nature, characteristics, and consequences of this behavior in order to analyze its characteristics and manifestations in the online context during a vulnerable period such as adolescence. Many different terms have been used for the definition of harassment, victimization, violence, and abusive behavior on the internet: "digital", "internet", "cyber" or "online" are the most common (Henry, Flynn, & Powell, 2020). The wide variety of terms, definitions and measures used in the literature makes it difficult to compare data on the prevalence and incidence of sexual harassment on the internet. There is still no agreement among researchers on how to define and describe online sexual harassment, despite numerous studies on the topic. The main objective of the present dissertation was an in-depth understanding of online peer sexual harassment, the definition of the phenomenon from both a theoretical and measurement point of view, and the description of its prevalence in the Italian context, with particular attention to the identification of possible similarities and differences between online peer sexual harassment and other form of online peer victimization, like cyberbullying. Specifically, this dissertation consists of three studies: 1) a scoping review aimed at defining online sexual harassment in adolescence; 2) a validation and measurement invariance study of the Peer Sexual Cybervictimization Scale (SCV) - revised; 3) a final study investigating the associations between online sexual harassment among peers and cybervictimization, identifying possible profiles and risk factors. In the first study (Chapter 1), we present a scoping review aimed at defining and describing online sexual harassment through the following objectives: a) to identify labels used to refer to online sexual harassment in the adolescent population; b) to describe the definition of online sexual harassment, considering specific criteria that emerge from the studies; c) to describe different typologies of abuse in which online sexual harassment occurs. In order to meet the objectives of our study, a great deal of information was analyzed and coded, such as labels, the type of relationship between the aggressor and the victim, the use of abusive connotations, the time frame in which the behaviors occurred, typologies of online sexual harassment, associated behaviors, and the focus on victimization and/or perpetration. The scoping review search was conducted in November 2021, following the PRISMA guidelines (Moher et al., 2009; Page et al., 2021), across four databases (Scopus, Web of Science, PsycInfo and PubMed). The initial search yielded 20958 results: after removing duplicates, screening, and reading the full text, 65 papers were included in the review. The speed with which platforms and digital tools evolve, and the emergence of new ways to share personal information of all kinds, makes it difficult to summarize in a single theoretical definition all that online sexual harassment can be. Through this scoping review, it has been possible to identify some key characteristics of the phenomenon: online sexual harassment has an abusive connotation as it is perceived as unwanted by the victim, it can occur in three main typologies (verbal, visual, cybersex) and even a single episode is enough to experience victimization. In terms of relational behavior, online sexual harassment includes unwanted sexual requests and non-consensual exchanges. One of the findings of the first study of this dissertation is that online sexual harassment can have distinct characteristics and consequences depending on the relationship that exists between the victim and the perpetrator. The literature has focused more on the (sexual) harassment of minors perpetrated by an adult (i.e., grooming), but how this is expressed in a peer-to-peer context is poorly understood (Project DeShame, 2017). A recent report (involving 3257 boys aged 13-17 from Denmark, Hungary and the UK, Project De Shame, 2017) highlights that 6% reported having their nude or nearly nude image shared with other people without their permission in the past year, 9% have received sexual threats online from people their own age in the past year, 24% have received unwanted sexual messages or images in the past year, with girls being significantly more likely to experience this (30%) than boys (13%). The prevalence found in studies varies widely, from 1% to 59% (Reed et al., 2019). For this reason, in the second study (Chapter 2), we propose a revised measure of the Peer Sexual Cybervictimization Scale (SCV) (Sànchez-Jiménez et al., 2017), adding a non-consensual sharing (NCS) dimension to the ambiguous sexual cybervictimization (ASCV) and personal sexual cybervictimization (PSCV) dimensions (Franceschi et al., 2023). The aim of the study is to analyze the psychometric properties of the revised measure in an Italian and Spanish sample. This will be done using confirmatory factor analysis (CFA). The second aim is to examine differences in the level of victimization between countries. This is done by testing the invariance of the CFA model. A total of 2167 participants (Spain: N=781, Mage=15.00, SD= .88, 49% girls - Italy: N=1386, Mage=14.85, SD=.90, 50.9% girls). Confirmatory factor analysis of the tested second order factor model shows an excellent fit: χ2 (51) = 176.159, p<.0000, CFI = .944, RMSEA = .069, confidence interval [CI] 90% = .057; .081; SMRS = .038. In addition, a fully scalar-invariant model was obtained. The presence of behavior is slightly higher in the Spanish sample (48.1%) than in the Italian one (42.1%). This is in line with previous studies (Sànchez-Jiménez et al., 2017; Longobardi et al., 2021). After the validation of the questionnaire, it was possible to deepen the phenomenon, trying to understand how this form of online peer victimization relates to other forms of victimization such as bullying and cyberbullying. For this reason, in the third study (Chapter 3) we aim to explore the possible links between online sexual harassment among peers and cybervictimization, and to identify and compare individual and contextual risk factors for these two forms of online peer victimization, highlighting similarities and differences between the two constructs. The research is based on the fourth and fifth surveys of a longitudinal project of national interest (PRIN). The final sample (i.e., the project control sample) included a total of 697 participants (Mage=15.17; SD=0.68; 42.3% female). 36.7% of the participants reported being victims of online sexual harassment at least once. The data regarding cybervictimization are similar (37.4%). Being a victim of cybervictimization is positively associated with being a victim of online sexual harassment (𝝆𝝆s=.426**). Although the two online victimization behaviors are similar, they are explained by different predictors. Specifically, cybervictimization is predicted by the presence of problematic relationships with peers (β=.169*, SE=.06) and lack of school connectedness (β=-.189*, SE=.08), whereas online sexual harassment is predicted by the presence of emotional symptoms (β=.244**, SE=.07) and lack of social norms (β=-.257**, SE=.08). Cybervictimization and online sexual harassment are aggressive behaviors involving an aggressor and a victim, and therefore often become two sides of the same coin in the peer context. However, there are differences between the two behaviors, as cybervictimization is more easily explained by contextual factors related to the classroom climate and a group phenomenon, whereas online sexual harassment is more easily explained by individual factors such as the presence of emotional symptoms and the internalization of social norms. The main findings of the three previous studies are discussed in detail in the concluding chapter (Chapter 4), together with their contributions to the literature and relevant strengths and weaknesses. Recommendations for policy and educational practice are also made, together with implications for future research and intervention.