Media Framing of Sexual Abuse Involving School-Age Girls in Rural Ethiopia: A Qualitative Content Analysis (2020–2025)

Mieraf Mesfin Girma  

December 4, 2025 

Mieraf Mesfin Girma, Gibson School Systems, mirafmesfin9@gmail.com 

Thank you for the guidance of Linsay Flowers mentor from Ottawa University in the development of this research. 

Media Framing of Sexual Abuse Involving School-Age Girls in Rural  Ethiopia: A Qualitative Content Analysis (2020–2025) 

Abstract - This qualitative content analysis examines how Ethiopian news media framed sexual abuse cases  involving school-age girls (aged 5–18) between 2020 and 2025, with particular attention to rural incidents.  Grounded in framing theory, 27 purposively sampled articles from government-affiliated and independent  

outlets were analysed using a hybrid deductive-inductive approach. Four dominant frames emerged:  institutional response (59 %), victim-blaming (44 %), perpetrator-excusing (37 %), and advocacy-oriented  (26 %). Rural cases (only 22 % of the sample) were severely underrepresented and characterised by extreme  brevity, absence of survivor voices, episodic framing, and euphemistic language, whereas urban cases received  greater narrative depth and occasional systemic critique. Government-aligned media emphasised state  efficacy; independent outlets highlighted structural failures. The findings reveal geographic erasure and  linguistic minimisation that perpetuate stigma and silence rural victims, underscoring the urgent need for  geographically equitable, survivor-centred journalism. 

Index Terms - Gender-based violence, Geographic erasure, Media framing, Rural Ethiopia, School-age girls, Sexual  violence 

Introduction 

Sexual abuse of school-age girls (aged 5–18) remains one of Ethiopia’s most pervasive yet least  publicly recognised human rights violations. In rural regions, which are home to more than 80 %  of the population, girls routinely face sexual violence while walking long distances to school,  collecting water from distant streams, or performing household chores in isolated fields and  forests. Poverty forces many families to prioritise boys’ education, leaving girls to travel alone on  unmarked paths that cut through bushland for five, ten, or even fifteen kilometres each way. Early  marriage practices, inadequate school infrastructure with no fences or lighting, chronic  understaffing of rural police posts, and deeply entrenched cultural norms that equate family honour  with female silence create an environment in which abuse is both commonplace and systematically  concealed. Recent empirical studies estimate that between 31 % and 34 % of female students have  experienced some form of sexual violence during their schooling, and researchers consistently  emphasise that rural prevalence is almost certainly higher because of chronic under-reporting, fear  of ostracism, and intense community pressure to resolve cases privately through elder mediation  rather than formal justice systems (Hailu et al., 2024; Mekonnen et al., 2024).

Despite a series of progressive legislative and policy reforms—the 2005 Criminal Code  amendments that explicitly criminalised marital rape and child sexual abuse, the 2018 National  Costed Roadmap to End Child Marriage and Female Genital Mutilation, and the 2023 National  Gender-Based Violence Policy Framework—enforcement in remote areas remains conspicuously  weak. Courts are often hundreds of kilometres away, public transport is scarce or non-existent,  police outposts are under-resourced and understaffed, and local elders frequently mediate disputes  in ways that prioritise reconciliation, compensation, or marriage between victim and perpetrator  over prosecution and victim support. In this institutional vacuum, news media emerge as one of  the very few mechanisms capable of transforming private, localised suffering into a matter of  national public concern and political accountability. 

The five-year period under examination, from 2020 to 2025, was exceptionally turbulent and  therefore analytically rich. It encompassed the Tigray civil war and its devastating aftermath, the  rapid proliferation of digital news platforms and social-media penetration even into rural areas,  recurring climate-induced food crises and drought, soaring inflation that further strained household  survival strategies, and shifting gender-policy priorities under the Prosperity Party government.  These overlapping shocks and transformations created a unique historical conjuncture: a moment  when national and international attention to gender-based violence arguably reached its highest  level in decades, yet routine, peacetime sexual violence against rural schoolgirls continued largely  unseen and unremarked in mainstream media narratives. 

This study examines how Ethiopian news media framed cases of sexual abuse involving school age girls during these five critical years, with particular and deliberate attention to rural incidents  that have historically been marginalised in both research and public discourse. 

The paper is structured as follows: 

 A literature review synthesises existing knowledge on prevalence, media representation, and  persistent research gaps. 

 A dedicated section presents the personal and intellectual rationale for the study, states the central  research question and specific objectives, and articulates its scientific and social significance. 

 The methodology describes the qualitative content analysis design, purposive sampling strategy,  and analytical procedures. 

 The results section details the dominant frames identified, the stark rural–urban disparities,  ownership differences, and pervasive linguistic patterns. 

 The discussion interprets these findings in relation to framing theory, journalistic ethics, and  policy implications while acknowledging limitations. 

 Finally, the conclusion summarises the core insights and proposes concrete, actionable pathways  forward for more equitable and survivor-centred media practice in Ethiopia.

Literature Review 

Research on sexual violence against school-age girls (aged 5–18) in Ethiopia can be organised into  four broad, interconnected themes that collectively underscore both the severity of the crisis and  the persistent limitations of existing scholarship. 

The first and most developed theme consists of prevalence and risk-factor studies, which  consistently document alarmingly high rates of sexual violence, particularly in rural and peri-urban  settings. Hailu et al. (2024), in a cross-sectional survey of 412 high-school girls in Soddo Town  and surrounding rural kebeles in Southern Ethiopia, found that 34.2 % had experienced at least  one form of sexual violence, ranging from unwanted touching to completed rape. Key risk factors  identified included entrenched patriarchal attitudes, inadequate parental or family discussion of  sexual and reproductive health, long and unsafe walking distances to school (often 5–15 km), and  school environments marked by poor lighting, absence of perimeter fencing, and minimal adult  supervision after classes. Similarly, Mekonnen et al. (2024) reported a lifetime prevalence of 31.8  % among 384 secondary-school students in the predominantly rural Arba Minch Zuria woreda,  emphasising poverty-driven early marriage, cultural tolerance of bride abduction, and the  normalisation of teacher–student sexual relationships as “love affairs.” These findings are  consistent with broader Sub-Saharan African patterns: Abdullahi et al.’s (2019) systematic review  and meta-analysis of 28 studies estimated a pooled prevalence of childhood sexual violence  between 18 % and 37 %, with schools and school routes repeatedly identified as primary sites of  abuse. Although these quantitative studies powerfully illustrate the systemic and structural nature  of the problem, they almost never examine how public narratives—especially those constructed  and circulated by news media—influence stigma, formal reporting behaviour, help-seeking  patterns, or girls’ continued access to education. 

A second theme examines media representation of gender-based violence more broadly and reveals  enduring patterns of sensationalism, episodic framing, and structural avoidance. Ejigu’s (2008)  pioneering analysis of 120 Ethiopian newspaper articles published between 2005 and 2007 remains  the most comprehensive local study: it concluded that girls were overwhelmingly portrayed as  passive, at best, passive victims requiring rescue and, at worst, complicit through alleged “risky  behaviour,” while sexual violence itself was framed as isolated individual tragedy rather than a  predictable outcome of systemic failure. This pattern persists internationally. Bello et al. (2020),  using machine-learning techniques on more than 10,000 global news articles, demonstrated that  criminal proceedings, graphic injury descriptions, and courtroom drama dominate coverage,  whereas poverty, unsafe school infrastructure, long commutes, and survivor recovery are  systematically marginalised.  

In Ethiopia’s rapidly expanding digital sphere, Nigatu and Raji (2024) showed that Amharic language YouTube recommendation algorithms frequently redirect users searching for educational  or religious content toward sexualised videos, thereby amplifying spectacle and shock value rather  than prevention or contextual understanding.

Collectively, these studies suggest that both traditional and digital media often reinforce  stereotypes and hinder prevention efforts, yet none specifically investigates downstream effects on  young, rural audiences or differentiates between government-aligned and independent outlets. 

A third theme highlights the dramatic visibility gap between conflict-related sexual violence and  routine, peacetime abuse. Systematic weaponisation of rape during the Tigray war (2020–2022)  received sustained international attention, detailed UN investigations, and calls for war-crimes  prosecution (The Guardian Staff, 2025). By contrast, everyday sexual violence in schools, homes,  and villages across non-conflict regions attracts only sporadic, decontextualised mentions that  rarely progress beyond the initial police report (Obamo, 2025). This disparity constructs conflict 

related atrocities as exceptional horrors worthy of global outrage while normalising peacetime  abuse as inevitable background noise, even though the latter affects orders of magnitude more girls  on a daily basis. 

The fourth and final theme is the near-total absence of longitudinal, experimental, or causal  research linking media framing to measurable behavioural or educational outcomes. Cross sectional prevalence studies (Hailu et al., 2024; Mekonnen et al., 2024), content analyses (Ejigu,  2008; Bello et al., 2020), and algorithmic audits (Nigatu & Raji, 2024) provide rich description  but cannot determine whether sensationalised or victim-blaming coverage increases fear-driven  school absenteeism, whether survivor-centred narratives encourage formal reporting, or how  stigmatising language affects mental health trajectories among rural girls. The field therefore  remains largely descriptive, retrospective, and—crucially—urban-biased in both data collection  and analytical focus. 

Taken together, these four bodies of work leave several critical gaps that the present study directly  addresses. First, rural school-age girls have never been centred as the primary subjects of media framing research in Ethiopia; existing studies either aggregate all age groups or prioritise urban  and conflict-related cases. Second, media ownership—government-affiliated versus  independent—is almost never differentiated, despite its obvious implications for frame selection  and source access. Third, radio, which remains the dominant information source for more than 80  % of rural households (UNESCO, 2023), is routinely excluded from content analyses due to the  absence of transcripts. Fourth, linguistic mechanisms of responsibility deflection—euphemism,  passive voice, moralistic appeals to “honour” or “tradition—remain under-theorised as active  framing devices rather than mere stylistic choices.  

By deliberately prioritising rural incidents, systematically comparing government-aligned and  independent outlets, treating geography and linguistic deflection as explicit framing moderators,  and grounding the analysis in framing theory, this study seeks to fill these interconnected omissions  and provide a more nuanced understanding of how media contribute to the continued  marginalisation of Ethiopia’s most vulnerable girls.

Research Rationale, Question, Objectives, and Significance 

My encounter with this topic did not begin in academic journals but in August 2023, when the  brutal rape and murder of seven-year-old Heaven Awot in Bahir Dar violently shattered the fragile  sense of safety many Ethiopian families still cling to. A single mother and nurse, Abekyelesh  Adeba, had lived in the same rented house for two years. One evening, she watched her daughter  step out to use the outdoor toilet—and Heaven never returned. The landlord, Getnet Baye,  strangled her, forced sand and pebbles into her mouth, mutilated her genitalia, and slashed her legs  with blades to stage an accident. He then attempted to silence the family with a bribe. For nearly  a year, the case remained buried. It surfaced only when Abekyelesh, driven by grief and  desperation, began posting raw videos and photographs on social media around the first  anniversary of her daughter’s death. Within weeks, an online petition demanding justice gathered  over 240,000 signatures, fueled by public outrage at the perpetrator’s 25-year sentence—an  outcome widely perceived as yet another miscarriage of justice. 

As I followed the case and spoke to residents in Bahir Dar, I uncovered at least four similarly  horrific child murders in the same town—cases that never reached the police, the courts, or the  media because the victims’ families lacked smartphones, internet access, or the social capital to  make their stories go viral. A search of digital archives of major Ethiopian news outlets revealed  that Heaven’s case had appeared only fleetingly, and sensationally, because social media had  forced it into existence. Thousands of other cases left no trace at all. 

In an interview with Eyoha Media, Abekyelesh clutched a photograph of Heaven in her little dress  and said, “If my videos had been ignored like all the others, it would be as if she had never lived.” 

That sentence crystallized a profound and painful disjuncture in contemporary Ethiopia: the stark  divide between lived suffering and mediated visibility. It convinced me that news media—and, at  times, even social media—do not merely report social problems; they actively determine whose  pain is granted public existence, whose grief is deemed worthy of collective recognition, and  whose demand for justice is permitted to matter. 

This study is therefore guided by one central research question: In what ways did Ethiopian news  media represent cases of sexual abuse involving school-age girls in rural areas between 2020  and 2025?  

Four specific objectives flow directly from this question.  

The first is to identify and describe the dominant frames employed in such coverage. The second  is to compare framing practices—including narrative depth, source selection, emotional tone,  linguistic choices, and structural contextualisation—across rural versus urban incidents and  between government-affiliated and independent outlets.

The third is to analyse the rhetorical and linguistic devices (euphemism, passive voice, victim blaming tropes, moralistic appeals to “honour” or “tradition”) that minimise or deflect perpetrator  responsibility and amplify or displace stigma onto victims and families. The fourth is to formulate  concrete, evidence-based recommendations for more geographically equitable and survivor centred journalistic practice. 

Scientifically, the research addresses several long-standing gaps simultaneously. No previous  study has centred rural school-age girls as the primary subjects of media-framing analysis in  Ethiopia; existing work either aggregates age groups or privileges urban and conflict-related cases.  Media ownership—government-affiliated versus independent—has rarely been systematically  differentiated despite its obvious implications for frame selection and critical distance. 

Geography and linguistic deflection have never been treated as explicit framing moderators rather  than incidental variables. By deliberately prioritising rural incidents, comparing ownership types,  and examining how the same frame can function as amplification in urban reports yet erasure in  rural ones, this study extends framing theory itself and offers the first comprehensive account of  urban bias in contemporary Ethiopian gender-based violence discourse. 

Socially, the stakes could not be higher. When rural girls’ experiences are flattened into brief,  voiceless “tragic incidents” devoid of context or agency, public empathy, donor funding, NGO  programming, and policy attention remain stubbornly urban-centred. Perpetrators in remote  woredas continue to enjoy near-total impunity, and legislative reforms—the 2005 Criminal Code  amendments, the 2018 National Costed Roadmap to End Child Marriage and FGM, and the 2023  Gender-Based Violence Policy Framework—remain largely unimplemented in areas where the  problem is rarely framed as a public crisis requiring structural solutions. This geographic injustice  sustains a cruel two-tier system of recognition: urban survivors are more likely to be constructed  as credible victims deserving support and protection, whereas rural girls are rendered invisible or  implicitly blameworthy through suggestions that they “should have been more careful” or that  their families failed in supervision.  

By exposing the mechanisms through which brevity becomes erasure, official sources crowd out  survivor voices, and euphemism dilutes brutality, this study seeks to inform targeted interventions:  mandatory trauma-informed journalism modules in university curricula and in-service training,  investment in community radio stations broadcasting in local languages, the drafting of national  ethical guidelines for GBV reporting that explicitly mandate geographic equity and linguistic  responsibility, and rural media-literacy programmes that empower families and girls to recognise  and challenge stigmatising frames. Only when rural schoolgirls are transformed from silenced  statistics into rights-bearing citizens whose stories demand accountability and systemic change can Ethiopia begin to claim meaningful progress toward gender justice. Making the invisible visible is  therefore not only an academic contribution but a moral and political imperative.

Methodology 

This study adopts qualitative content analysis (QCA) as its core methodological approach to  investigate the framing of sexual abuse cases involving school-age girls (aged 5–18) in Ethiopian  news media between 2020 and 2025. Qualitative content analysis is especially appropriate for  framing studies because it moves beyond surface-level description or simple frequency counts and  instead privileges interpretive depth, contextual sensitivity, and the systematic deconstruction of  meaning-making processes embedded in media texts. 

It allows the researcher to examine not only what is explicitly stated but also how particular  linguistic choices, narrative structures, source hierarchies, emotional registers, rhetorical  strategies, thematic emphases, and deliberate omissions work together to construct particular  versions of reality.  

In the context of sexual violence against rural schoolgirls, such an approach is indispensable  because the harm is often reproduced not through outright denial but through subtle discursive  mechanisms: euphemisation, agent deletion, temporal displacement, geographic erasure, and the  systematic privileging of official voices over those of survivors and their families. 

The theoretical scaffolding of the analysis rests primarily on Robert Entman’s (1993) widely cited  definition of framing, which posits that to frame is to select some aspects of a perceived reality  and make them more salient in a communicating text in such a way as to promote a particular  problem definition, causal interpretation, moral evaluation, and/or treatment recommendation.  Entman’s framework is complemented by insights from Gamson and Modigliani (1989) on media  packages and interpretive repertoires, as well as by feminist adaptations of framing theory (e.g.,  Lamb, 1999; Meyers, 2004; Benedict, 1992) that highlight how gendered power relations are  routinely reproduced through victim-blaming scripts, myths of stranger danger, and the erasure of  structural and everyday forms of violence.  

By integrating these perspectives, the study traces how Ethiopian media frames simultaneously  reflect and reinforce a deeply stratified social order in which urban, elite, and conflict-related  suffering is granted visibility and moral urgency, while the routine, peacetime violations  experienced by rural girls remain largely outside the sphere of public recognition and political  accountability. 

Data were collected from newspaper articles and online news reports published in Amharic and  English by major Ethiopian outlets during the six-year period from 1 January 2020 to 31 December  2025. The outlets were deliberately chosen to represent the country’s polarised media landscape:  government-affiliated platforms (Ethiopian News Agency – ENA, Fana Broadcasting Corporate,  Walta Information Centre, Ethiopian Press Agency products) and prominent independent or  privately owned outlets (Addis Standard, The Reporter Ethiopia, Capital Ethiopia, Addis Fortune,  Ethiopia Insight, and Addis Zeybe). 

This binary distinction between state-aligned and independent media is analytically crucial  because ownership structures in Ethiopia profoundly shape editorial priorities, source access,  linguistic register, and the degree of critical distance journalists can maintain from official  narratives. 

Sampling followed a purposive, criterion-based strategy designed to maximise theoretical  relevance, narrative richness, and variation along three key dimensions: geographic location of the  incident (rural vs. urban/peri-urban), media ownership type, and the presence of identifiable  framing devices.  

Articles were included only if they met all of the following criteria: (a) they reported a specific  incident or tightly linked cluster of incidents of sexual abuse or attempted sexual abuse; (b) the  victim or victims were female, aged 5–18, and either enrolled in school or travelling to or from  school at the time of the incident; (c) the incident occurred within Ethiopia’s national territory  (excluding diaspora or refugee-camp cases); and (d) the article contained sufficient descriptive and  contextual detail, typically a minimum of 150 words of substantive narrative, to permit meaningful  framing and linguistic analysis. Articles were excluded if they offered only passing references,  presented aggregated statistics without case-specific narration, focused exclusively on court  proceedings or sentencing without describing the original incident, or functioned primarily as  opinion pieces or editorials rather than news reports. 

The search process combined multiple strategies to overcome the well-documented  incompleteness and frequent deletion of Ethiopian digital archives. Keywords and phrases in both  Amharic (e.g., ደፋር, መድፈር, ወሲብ ጥቃት, ገፈፋ/ተገፋ, ተማሪ ሴት ልጅ, ትምህርት ቤት መንገድ, አፈና, ደፈራ) and  English (rape, sexual violence, abduction, defilement, schoolgirl, student, minor, assault) were  used across the internal search functions of each outlet’s website. 

Where archives were missing, paywalled, or had been removed, site-specific Google searches,  Boolean operators, date-range restrictions, the Internet Archive’s Wayback Machine, and the  Ethiopian Media Authority’s public repository were systematically consulted. Snowballing from  citations in human-rights reports, activist posts on social media, and references in academic studies  helped surface additional cases that had received only fleeting mainstream coverage. 

The final sample consists of 27 articles. Despite sustained and deliberate efforts to locate and  prioritise rural incidents, including targeted searches in regional editions of government media and  the use of Amharic keywords more common in rural reporting, only six articles (22%)  unambiguously described incidents that took place in clearly rural settings (woredas outside major  towns, involving victims travelling long distances on foot, collecting water or firewood, or  attending under-resourced primary schools). The remaining 21 articles (78%) focused on incidents  in Addis Ababa, regional capitals, or peri-urban zones. This stark rural–urban imbalance is not a  product of sampling error but an empirical reflection of actual coverage patterns and constitutes  one of the study’s central findings.

Analysis was conducted iteratively using Microsoft Excel as the primary organisational and coding  tool. A master spreadsheet was created with one row per article and separate columns for article  metadata (title, outlet, date, language, ownership type, location of incident, victim age, etc.) and  for emerging analytical categories. In the first phase, each article was read multiple times and  subjected to line-by-line open coding; codes and illustrative quotations were entered directly into  dedicated columns, allowing rapid visual comparison across cases. Recurrent frames (e.g.,  “isolated tragic incident,” “breach of family honour,” “failure of parental supervision,” “law enforcement success”), source patterns, linguistic devices (euphemisms, passive constructions,  agent deletion), and significant silences were documented systematically. 

In the second phase, axial coding was performed by creating additional worksheets that grouped  and compared codes across rural/urban location and ownership type. Conditional formatting,  filters, and pivot tables facilitated the identification of patterns and divergences. In the third phase,  the most salient frames were interpreted in relation to broader socio-cultural norms, journalistic  routines, the political economy of Ethiopian media, and the rural–urban visibility gap. Thick  description and verbatim excerpts (translated where necessary) are provided in the results section  to enable readers to assess the interpretations offered. 

Methodological rigor was enhanced through prolonged engagement with the texts over more than  fourteen months, peer debriefing sessions with two senior Ethiopian gender scholars experienced  in both media analysis and rural fieldwork, meticulous maintenance of an audit trail documenting  all coding decisions and revisions, and reflexive memos on researcher positionality. The use of  Excel, while less automated than dedicated qualitative software, offered complete transparency,  ease of version control, and the flexibility required for the relatively modest sample size while still  permitting systematic, replicable analysis. 

This methodological framework thus delivers a transparent, theoretically grounded approach that  deliberately centres the rural, routine, peacetime cases that have historically been marginalised in  both Ethiopian journalism and academic scholarship. By illuminating the discursive mechanisms  through which rural schoolgirls’ suffering is rendered invisible or trivial, the study lays an  empirical and ethical foundation for advocating more geographically equitable and survivor 

centred media practice.

Results 

The analysis of the 27 articles revealed a media ecosystem characterised by profound geographic,  ownership, and linguistic inequities that systematically disadvantaged rural school-age victims of  sexual abuse. 

Despite a purposive sampling strategy explicitly designed to maximise the inclusion of rural  incidents, only six articles (22 %) focused on abuse occurring outside urban or mixed settings,  while twenty-one (78 %) concerned urban or peri-urban cases. 

This stark underrepresentation emerged as a finding in itself, confirming the systemic urban bias  the study sought to interrogate. Rural reports were dramatically shorter, averaging just 180 words  compared with 420 words for urban reports — a difference that severely restricted narrative depth,  emotional resonance, and the possibility of structural contextualisation from the outset.  Government-affiliated outlets (Ethiopian News Agency, Fana Broadcasting Corporate, Walta)  contributed fourteen articles (52 %), while independent outlets (Addis Standard, The Reporter,  Capital) produced thirteen (48 %). Temporal distribution showed seventeen articles (63 %)  published after the 2023 National Gender-Based Violence Policy Framework, yet rural  representation exhibited no discernible improvement.