The research team employed a convergent mixed-methods design, a strategy that gathers numerical and narrative evidence in parallel and then integrates the two streams to produce a richer picture than either could alone. Quantitative data came from an online survey administered to 400 children aged 8 to 17, collected through the KoboToolbox data platform, which is widely used in field research across low-resource settings. In parallel, the researchers conducted telephone interviews with 50 children who served as key informants, allowing them to probe the texture of children’s experiences in their own words. The statistical analysis relied on descriptive statistics to map the prevalence of different encounter types, Pearson’s chi-square tests to examine associations between demographic characteristics and risk, and Fisher’s exact test where cell counts were too small for the chi-square approximation to be reliable. The qualitative interviews were analyzed thematically, following the widely cited approach developed by Virginia Braun and Victoria Clarke.

The demographic gradients in the data are among the study’s most consequential findings. Girls reported unsolicited encounters at a rate of 78.7 percent, substantially higher than the overall sample average, a pattern consistent with a large international literature documenting that online harassment and unwanted sexual contact disproportionately target female users. Age mattered just as much: among older children in the sample, prevalence climbed to 85.9 percent. This developmental gradient makes intuitive sense. Adolescents spend more time online, maintain larger networks of contacts, use direct messaging more intensively, and are more likely to engage with strangers, all of which expand the attack surface for unwanted contact. Developmental psychology also suggests that adolescents are more susceptible to social influence and peer dynamics, which can normalize risky sharing and make it harder to disengage from hostile interactions once they begin.

What exactly are children encountering? The most common form of unsolicited contact, reported by 65.5 percent of those who experienced such encounters, was uncomplimentary remarks, essentially verbal abuse, insults, and demeaning comments directed at the child. This finding aligns with the extensive cyberbullying literature, including meta-analyses by Robert Kowalski and colleagues showing that digital harassment is associated with depression, anxiety, and psychosomatic complaints among adolescents. Nearly half of affected children, 49.9 percent, reported exposure to violent content they had not sought out, echoing decades of research on the effects of media violence on young people. Most alarmingly, 44.8 percent reported sexual harassment, a figure that places Tema’s children within the range documented in meta-analytic work by Sheri Madigan and colleagues on unwanted online sexual exposure and solicitation among youth globally. That nearly one in two children who face unwanted online contact experience sexualized harassment underscores how urgently child protection frameworks in the region need updating.

The qualitative interviews add crucial interpretive depth to these numbers. Children described how hostile interactions unfold in comment threads, private messages, and group chats, and how content that is violent, sexual, or otherwise inappropriate surfaces in their feeds without any deliberate search on their part. The interviews also illuminate the social context in which these encounters occur: children often know the perpetrators, at least by online identity, and the boundary between peer conflict and genuine victimization can be blurry. This matters for intervention design, because responses that treat all unwanted contact as predation by strangers may miss the more common reality of peer-to-peer cruelty, which requires different educational and disciplinary strategies than those aimed at adult offenders.

How do children respond when these encounters happen? The study found that the most common reaction, reported by 34.6 percent of affected children, was to tell a parent or guardian, a finding the researchers treat as encouraging because disclosure is the first step toward adult intervention. Close behind, 29.0 percent said they blocked the offender, using the platform-level technical controls that most major social media services provide. Blocking is an effective immediate defense, but it is a purely individual remedy: it removes one perpetrator while leaving the underlying dynamics of the platform untouched, and it does nothing for children who lack the confidence or knowledge to use these tools. The distribution of responses suggests that a meaningful share of children neither disclose nor block, silently absorbing the harm, a pattern that previous research on adolescents and parental communication has identified as a persistent challenge for safeguarding.

The theoretical framing of the study draws on ecological systems theory, the framework developed by Urie Bronfenbrenner, which holds that child development is shaped by nested layers of environment, from the immediate family to broader cultural and policy contexts. Applied to online risk, this perspective insists that a child’s vulnerability is not a fixed individual trait but a product of interactions between the child’s digital literacy, the supervision practices of parents, the design choices of platforms, and the legal and institutional environment of the country. Ghana has begun building that institutional layer: the country passed a Cybersecurity Act in 2020 and, in 2024, adopted a National Child Online Protection Framework. The Tema findings provide exactly the kind of empirical baseline that such frameworks need in order to prioritize resources and measure progress over time.

The authors argue that three lines of action follow from their results. First, digital literacy programs should be expanded and embedded in schools, teaching children not only how to use platforms but how to recognize manipulation, manage privacy settings, and respond constructively to harassment. Recent work on evidence-based digital literacy curricula for youth suggests these programs can be effective when they are assessed and iterated rather than delivered as one-off awareness campaigns. Second, parental involvement needs strengthening, which in the Ghanaian context may require parent education initiatives, since many caregivers did not grow up with social media and may underestimate both its benefits and its risks. Third, the researchers call for effective and enforceable policies, noting that platform safety features designed primarily for Western markets may not account for the linguistic, cultural, and infrastructural realities of African users.

The study also makes a methodological contribution to a field in which children’s voices are often mediated entirely through adult reports. By surveying children directly and interviewing a subset as key informants, the researchers treated young people as credible witnesses to their own digital lives, an approach grounded in the ethical guidance for research with children articulated by Priscilla Alderson and Virginia Morrow. The study received ethical approval from the National Child Protection Committee of Ghana’s Ministry of Gender, Children and Social Protection, and informed consent was obtained from all respondents. The authors note that the data cannot be shared openly due to ethical restrictions protecting child participants, though reasonable requests can be directed to the corresponding author.

For a global audience, the Tema findings carry a message that extends well beyond Ghana. As internet access expands across Africa faster than child protection infrastructure, hundreds of millions of children are coming online in societies where safeguarding institutions are still maturing. The patterns documented here, with girls and adolescents bearing disproportionate risk and sexual harassment approaching the prevalence of more widely discussed harms like cyberbullying, suggest that the global conversation about child online safety must center the experiences of children in the Global South rather than extrapolating from European and North American surveys. Two-thirds of children encountering unwanted contact is not a marginal problem to be managed; it is a structural feature of how current platforms intersect with childhood, and the Tema study offers both the evidence and the policy roadmap to begin changing that.

Subject of Research: Children's experiences of unsolicited online encounters on social media in Tema, Ghana

Article Title: Social media and child safety in Tema, Ghana: Examining unsolicited online encounters

Article References: Kyei-Gyamfi, S., Kyei-Arthur, F., Kyei-Gyamfi, Z., Kotey, V. K., & Appiah, M. (2026). Social media and child safety in Tema, Ghana: Examining unsolicited online encounters. SN Social Sciences, 6(10), Article 517. https://doi.org/10.1007/s43545-026-01805-x

Image Credits: AI Generated

DOI: 10.1007/s43545-026-01805-x

Keywords: social media, child safety, Ghana, cyberbullying, online sexual harassment, digital literacy, parental mediation, child online protection, mixed methods, Tema, online risk, adolescents

The research team searched four major databases, Web of Science, Scopus, Dialnet, and PsycINFO, to identify studies that had quantitatively examined the relationship between sexting and sexual violence or online victimization. After screening the literature according to systematic review standards, the researchers included thirteen studies that analyzed the link between sexting and sexual violence, and seven studies that examined the relationship between sexting and online victimization. This two-track approach allowed the team to treat these outcomes separately, an important methodological decision, because sexual violence in offline contexts and victimization that occurs through digital channels, such as nonconsensual forwarding of intimate images or sextortion, are conceptually and legally distinct phenomena.

Meta-analysis is a statistical technique that combines the effect sizes reported in individual studies into a single pooled estimate, weighting each study by its precision. By aggregating correlations across independent samples, meta-analysts can detect patterns that would be invisible in any single investigation, which is often underpowered and subject to sampling error. The Spanish team followed established methodological guidance for synthesizing correlations, assessing the methodological quality of the primary studies with validated appraisal tools, and examining heterogeneity, the degree to which effect sizes vary across studies, using standard measures of inconsistency. They also applied funnel-plot-based techniques, including the trim-and-fill method, to probe for publication bias, the well-known tendency of journals to publish statistically significant findings more readily than null results.

The central finding is that sexting is significantly associated with both sexual violence and online victimization, but the moderators of these two associations differ in revealing ways. The relationship between sexting and sexual violence was stronger in samples that were more gender-heterogeneous, meaning groups that included a more balanced mix of male and female participants. By contrast, the association between sexting and online victimization was higher in younger groups. In other words, the demographic context in which sexting occurs appears to shape which kind of risk is most pronounced, a nuance that earlier narrative reviews, which simply catalogued studies without statistically pooling them, could not capture.

Why would gender composition moderate the link to sexual violence? One plausible interpretation, consistent with the broader literature on technology-facilitated sexual violence, is that in mixed-gender samples the dynamics of sexual aggression, including coercion to send images, pressure within dating relationships, and the nonconsensual distribution of intimate images, are more fully represented. Research cited within the meta-analysis has documented that pressured or coerced sexting is a recognized component of intimate partner aggression, and that adolescents who experience dating violence are more likely to report sexting under duress. Studies of young adults have similarly found that sexting coercion overlaps with other forms of sexual victimization, and that men arrested for domestic violence show elevated rates of sexting alongside sexual violence and alcohol use. In gender-heterogeneous samples, these perpetration pathways may be more visible, inflating the pooled association.

The age pattern for online victimization points to a different mechanism. Younger adolescents, who are still developing the judgment, emotional regulation, and digital literacy needed to navigate intimate online exchanges, may be especially vulnerable to having shared images misused. The literature reviewed by the team includes longitudinal work showing reciprocal relationships between sexting, online sexual solicitations, and cyberbullying among minors, as well as studies linking sexting to online grooming risk and to image-based abuse such as sextortion and nonconsensual forwarding, sometimes called revenge porn. Younger users may also be less aware of the permanence of digital content and less equipped to respond when a private image is weaponized against them, making the pathway from sending a message to experiencing victimization shorter and more direct.

The meta-analysis builds on a substantial body of prior quantitative work. Earlier systematic reviews and meta-analyses established that a considerable proportion of adolescents and young adults engage in sexting, with prevalence estimates rising over time as smartphone ownership has become nearly universal among teenagers. Those same reviews flagged the darker side of the practice: coerced sexting, nonconsensual dissemination, and the psychological distress that follows. Individual studies have connected sexting to sexual coercion among female adolescents, to dating violence among Italian and Spanish youth, and to sexual assault risk among college women, particularly when combined with problematic alcohol use. What the new analysis adds is a formal statistical synthesis that separates offline sexual violence from online victimization and identifies distinct moderators for each, resolving apparent inconsistencies across the primary literature.

The authors emphasize that their findings should not be read as a claim that sexting itself is inherently dangerous. For many adults and couples, consensual sexting is a normal expression of sexual intimacy, and research on adults’ sexual well-being has found no uniform harm from online sexual activities. The risk emerges at the intersection of sexting with coercion, nonconsensual sharing, and unequal relationship power. This distinction matters for prevention: blanket abstinence messages that demonize the behavior are unlikely to help, and may discourage young people from seeking support when images are misused. Instead, the results suggest that interventions should target the conditions under which sexting becomes a vector for violence, including pressure from partners, lack of consent education, and norms that shame victims, particularly girls, whose images are treated as more damaging when circulated.

The identification of vulnerable groups carries practical weight for educators, clinicians, and policymakers. Because the association with online victimization is strongest in younger groups, school-based digital safety programs may need to begin earlier than they typically do, before sexting behavior becomes established, and should cover concrete skills such as recognizing coercive requests, understanding the legal and social consequences of forwarding images, and knowing how to report image-based abuse. Because the association with sexual violence is stronger in gender-heterogeneous groups, prevention efforts within mixed settings, such as comprehensive sexuality education and dating violence programs, should explicitly address technology-facilitated coercion rather than treating online and offline violence as separate silos. The findings also align with a growing legal and scholarly consensus that image-based sexual abuse, including sextortion, constitutes a serious form of sexual violence in its own right.

Like all meta-analyses, the study has limitations that the authors acknowledge. The included studies relied predominantly on cross-sectional, self-reported data, which cannot establish the direction of causation: it remains possible that sexting increases exposure to violence, that victims of violence are later pressured into sexting, or that shared underlying factors, such as risky online behavior, low self-control, or exposure to hostile sexual norms, drive both. The number of studies available for each outcome, particularly the seven studies of online victimization, is modest, and heterogeneity across samples was expected given differences in how sexting and victimization were defined and measured. The authors call for future research to incorporate additional variables, most notably the role of sexual orientation in these associations, and to use longitudinal designs that can disentangle temporal ordering. The study was approved by the Universidad Europea of Madrid Institutional Review Board, and the authors declared no conflicts of interest and received no external funding. As sexting continues to normalize across generations, this synthesis provides a rigorous evidence base for distinguishing a common intimate behavior from the coercive contexts that turn it into a pathway to harm.

Subject of Research: Meta-analytic examination of the association between sexting and sexual violence and online victimization

Article Title: Association Between Sexting and Different Types of Sexual Violence: A Meta-Analysis

Article References: Association Between Sexting and Different Types of Sexual Violence: A Meta-Analysis. (n.d.). https://doi.org/10.1007/s10508-026-03526-x

Image Credits: AI Generated

DOI: 10.1007/s10508-026-03526-x

Keywords: sexting, sexual violence, online victimization, meta-analysis, image-based abuse, sextortion, adolescents, dating violence, coerced sexting, digital safety, Archives of Sexual Behavior, technology-facilitated violence