Systematic analysis of child protection and early learning outcomes

1. Introduction

1.1 Background

Early childhood education and the first years of primary schooling establish the foundations for literacy, numeracy, language, self-regulation and participation in later education. These foundations do not develop through classroom instruction alone. A child who is hungry, frequently absent, frightened, neglected, ill or exposed to violence may have fewer opportunities to concentrate, practise skills and build secure relationships with adults and peers. Child protection is therefore relevant to learning because it concerns the conditions that enable children to enter school, remain engaged and benefit from instruction. In this article, child protection refers broadly to the prevention of and response to abuse, neglect, exploitation, harmful practices and unsafe environments, together with actions that strengthen care, health, nutrition and access to services.

Research consistently associates childhood adversity with weaker educational engagement and attainment. Children exposed to adverse experiences have lower school engagement, while resilience and supportive systems can reduce some of this disadvantage (Bethell et al., 2014). Early adversity is also associated with poorer kindergarten outcomes (Jimenez et al., 2016), and maltreatment can affect language, emotion regulation and school readiness, which subsequently influence achievement (Panlilio et al., 2018). A global systematic review found that violence in childhood is linked with absence, dropout and lower academic achievement (Fry et al., 2018). Population-linked data similarly show that children reported to protection services are more likely to perform below average in Grade 3 and Grade 5 reading and numeracy, even after adjustment for other adversities (Laurens et al., 2020). These findings suggest that educational disadvantage may emerge through several connected pathways rather than through a single direct effect.

The pathways are especially important in resource-constrained settings. Poverty may reduce households' ability to provide food, uniforms, stationery, transport, and health care. It may also increase children’s exposure to neglect, hazardous work, transactional sex, early marriage and school withdrawal. Harmful norms can normalise violence or discourage reporting, while weak services can leave families without timely support. Kenyan and regional evidence demonstrates that physical abuse and neglect have psychological consequences across cultural settings (Mbagaya et al., 2013), and research in South Africa identifies poverty, caregiver vulnerability and limited support as factors associated with abuse victimisation (Meinck et al., 2015). Effective prevention consequently requires more than individual awareness. Evidence from low- and middle-income countries points towards combined strategies involving safe environments, parent and caregiver support, education, economic strengthening and responsive services (Pundir et al., 2020).

Schools and communities can also act as protective systems. The Good School Toolkit in Uganda reduced physical violence from school staff, illustrating that whole-school change can improve safety (Devries et al., 2015). Yet protection depends on whether adults recognise harm, accept responsibility and trust reporting mechanisms. Qualitative research has shown that adults may value child protection but encounter corruption, fear, unclear roles and practical barriers when attempting to act (Turner et al., 2022). These constraints are relevant to early-grade learning because a child’s absence, withdrawal, poor health or reduced participation may be noticed first by caregivers, faith leaders, community health volunteers or teachers. Their interpretations influence whether the child receives support or is blamed, ignored or removed from school.

1.2 Research Claim

Child protection strengthens the foundations of early learning indirectly by supporting school attendance, physical wellbeing, emotional security and consistent adult care, whereas failures in protection weaken those foundations. The claim is examined at the level of community-perceived mechanisms and proposed responses. It is not treated as a causal claim that the discussion guide or any specific protection intervention directly improved literacy, numeracy or other measured learning outcomes. Therefore, the following three questions guided this article:

1. Which protective practices did participants associate with children’s education and wellbeing?
2. Which risks were perceived to undermine those protective conditions and children’s school participation?
3. Which community responses were proposed to strengthen child protection, and, by implication, the conditions required for learning?

2. Methods

2.1 Research Design

The study used a qualitative descriptive design and secondary thematic analysis of focus group discussion notes.

2.2 Research Setting

The available records placed the discussions in Bamba in Ganze Sub-County, Kilifi County, in Kenya’s coastal region. More detailed records identified Mazunde village for the persons living with disabilities group and Goshi 1 for the local community leaders’ group. The documented dates for several groups were 22 and 24 September 2024.

2.3 Population and Sample

The study population comprised 47 adult community stakeholders involved in, affected by or knowledgeable about child wellbeing activities. The groups were not documented as consisting exclusively of parents of pre-primary or lower-primary learners. Accordingly, the article describes community perspectives and does not claim a parent-only sample.

Purposive stakeholder grouping was used to capture distinct social perspectives. Five focus group discussions were analysed. The faith-based organisation and faith-leader group included nine participants; the persons living with disabilities group included five; the men’s group included nine; the women’s group included eleven; and the local community leaders’ group included thirteen. The combined listed sample was 47 participants.

2.4 Data Collection Instrument

The semi-structured guide addressed participation in child wellbeing activities, a community vision for child wellbeing and education. Faith-based participants were also asked about Channels of Hope workshops and the social norms discussed. Handwritten responses were transcribed into a structured digital document. The available material consisted of concise responses and selected verbatim statements rather than full audio transcripts. Consequently, the analysis could identify recurring content but could not examine pauses, interactional dynamics, tone or the full sequence of discussion.

2.5 Data Analysis Procedures

The analysis followed the principles of thematic analysis described by Braun and Clarke (2006). First, the researcher read the complete transcription repeatedly and separated responses by stakeholder group. Second, meaning units were coded using short descriptive labels such as school attendance, learning materials, feeding, poverty, drought, parental responsibility, unsafe ceremonies, early marriage, reporting fear and civic education. Third, related codes were clustered into candidate themes. Fourth, themes were reviewed against all five groups and refined to distinguish protective supports, risks and proposed responses. Fifth, each theme was named and linked to the study claim through an explicit pathway to attendance, wellbeing, emotional safety or adult support. Finally, representative quotations were selected without altering their substantive wording.

2.6 Interpretation of Group-Level Percentages

The percentage column in the findings tables reports group-level thematic presence. A theme mentioned in all five focus groups is shown at 100%; four groups at 80%; three at 60%; two at 40%; and one at 20%. These values are not participant prevalence estimates and should not be interpreted as inferential statistics. Group-level calculation was selected because the notes did not record responses from every participant to every question and sometimes reported collective answers, such as “all in unison”. Participant identities were replaced with stakeholder codes: FBO for faith-based organisation or faith leader, PLWD for persons living with disabilities, MEN for the men’s group, WOM for the women’s group and LCL for local community leaders, followed by the recorded respondent number.

2.7 Ethical Considerations

The source transcription included participant names, but the manuscript omitted them to reduce identifiability. The supplied records did not include an ethics approval number, consent forms, or a full data management protocol. These details must be obtained from the original baseline-survey records and inserted before submission to a journal that requires formal ethical documentation.

3. Results

3.1 Overview of the Findings

Although the stakeholder groups differed in their social roles, their accounts converged on the understanding of child wellbeing. Education was not discussed as an isolated school activity. It related to food, health, household income, parental conduct, community norms, physical safety and access to reporting institutions. Three broad findings were developed: protective conditions that support learning, risks that weaken protection and school participation, and community actions perceived as capable of strengthening those conditions.

3.2. Protective Conditions Supporting Education and Development

3.3.1 Education and School Participation

All five groups referred to education, school participation or knowledge as part of child wellbeing. Faith-based participants described providing books, stationery, uniforms, fees and feeding support. Men identified school attendance, health, and nutrition as ways to improve children’s lives. Women mentioned school construction, toilets, healthy behaviour and balanced diets. Local leaders defined their child wellbeing vision through “education for all children” and proper nutrition. These responses show that stakeholders understood educational participation as dependent on both material access and supportive community action.

3.2.2 Nutrition and Health

Nutrition and health were especially prominent. Participants referred to school feeding, treatment of malnutrition, vaccination, medical facilities, balanced diets, breastfeeding and developmental milestone checks. These practices were not described using formal learning indicators, but they were understood as necessary for children to grow, attend school and function well.

3.2.3. Parental Care and Emotional Openness

The persons living with disabilities group emphasised parental love, cooperation and education on child wellbeing. Parental love and open communication were framed as forms of protection. A participant explained that children should be loved so that they feel free to share problems. Such disclosure can be important when harm is occurring at home, school or in the community.

3.2.4 Community and Faith-Based Support

Faith-based and community participants described congregational awareness, child-rights messages, school-fee support, assistance to orphans and vulnerable children, and community action. These accounts indicate that protection and learning support were viewed as shared responsibilities rather than duties belonging only to schools or households.

3.3 Risks Undermining Protection and Learning

3.3.1. Poverty and Resource Scarcity

Material deprivation was the most consistent risk across groups. Participants used the terms poverty, low economic conditions, unemployment, and insufficient resources. They connected these conditions with inability to provide basic needs, poor nutrition, limited follow-up and pressure on children.

3.3.2 Hunger, Drought and Inadequate Water or Nutrition

Participants identified hunger, drought, famine, lack of clean water and inadequate diets as threats to child wellbeing. These conditions were perceived to reduce children’s physical readiness for school and families’ ability to sustain regular attendance.

3.3.3 Parental Disengagement and Limited Knowledge

Parental conduct and low knowledge were repeatedly identified. Participants referred to irresponsible parents, parents failing to fulfill their roles, misunderstandings between parents, poor parental education, and resistance rooted in cultural beliefs. These observations indicate that learning barriers were perceived as both structural and relational. Poverty constrained care, but participants did not regard poverty as the only explanation; they also expected adults to provide love, guidance, basic needs and responsible supervision.

3.3.4 Harmful Norms, Unsafe Places and Sexual Risks

Women and local leaders identified risks to girls that included rape, pregnancy, early marriage and sex linked to lack of funds. Participants also named unsafe social spaces, including disco matangas, wedding-night ceremonies, shopping centres, recreational areas, peer groups and online environments. One local leader identified schools themselves as potentially unsafe, demonstrating that protection concerns were not confined to households.

3.4.5 Barriers to Reporting and Enforcement

Reporting was supported in principle but weakened in practice. Persons living with disabilities said abuse should be reported to police, while women mentioned chiefs, community leaders and police. However, participants identified fear of harm, misinterpretation, community judgement, ignorance and corruption as barriers. The report of a child beaten by a teacher illustrates that educational settings can be sites of abuse and that protection responses may require action beyond the classroom. These barriers may prolong exposure, discourage disclosure and contribute to absence, distress or withdrawal.

3.4 Community Responses and Their Perceived Relevance to Learning

Participants recommended more schools, improved transport, toilets, medical services, screening, water access and other infrastructure. These proposals show that respondents perceived protection as a combined social, economic and service responsibility. Their suggestions can reasonably be interpreted as mechanisms to improve learning conditions because they address attendance, health, safety, and caregiver capacity. The source data, however, did not contain a direct question asking whether the focus group guide itself would improve learning outcomes. No authentic percentage can therefore be reported for the tool's endorsement.

4. Discussion

4.1 Integration of Child Protection and Early Learning

The findings support the central claim regarding perceived mechanisms. Participants consistently connected children’s education with protection-related conditions: adequate food, access to health care, school materials, parental responsibility, emotional openness, safety from sexual and physical harm, and functioning community support. This understanding aligns with evidence that adversity affects school engagement and achievement through attendance, behaviour, emotion regulation, language and developmental readiness (Bethell et al., 2014; Jimenez et al., 2016; Panlilio et al., 2018). The qualitative results do not demonstrate that a specific protection intervention raised test scores. They show that community stakeholders recognised the conditions through which protection may enable young children to benefit from teaching.

The first implication is that child protection and early learning should not be planned as separate sectors. When FBO participants described books, stationery, uniforms, fees and feeding, they combined educational access with welfare support. When women and men discussed nutrition, health facilities and toilets, they described physical conditions that influence attendance and readiness to learn. Longitudinal and record-linkage studies show that maltreatment and protection-system contact are associated with later cognitive and educational disadvantage (Laurens et al., 2020; Strathearn et al., 2020). The local accounts help explain how this association may operate in everyday settings: a child without food, treatment, transport or basic school supplies has reduced opportunity to attend regularly and engage confidently.

4.2 Poverty, Drought and Material Deprivation

The second implication concerns poverty and drought. Every group raised material constraints, making deprivation the most widespread theme. Poverty should not be treated as equivalent to abuse, but it can intensify protection risks and limit a caregiver’s capacity to meet children’s needs. The suggested responses therefore included agriculture, irrigation, access to water, and employment and livelihood strengthening. This supports an ecological approach in which economic interventions complement parenting, school and reporting interventions. The evidence and gap map by Pundir et al. (2020) similarly identifies income and economic strengthening, caregiver support, education, safe environments and response services as connected intervention areas. A narrow awareness campaign would be insufficient where families understand child wellbeing but lack food, water, or access to services.

4.3 Gendered and Place-Based Protection Risks

The third implication is that gendered, and place-based risks require explicit attention. Women and leaders described rape, pregnancy, early marriage, transactional sex, unsafe ceremonies and online adult content. These risks can disrupt education through fear, stigma, caregiving responsibilities, pregnancy and school dropout. The fact that participants identified ceremonies, shopping centres, recreational spaces, schools and online environments means that a child-protection tool should map risk by time and place rather than referring only to the home. It should also avoid presenting girls as the only vulnerable group. Women acknowledged that both boys and girls may leave school to support the family, and effective safeguarding must remain attentive to different forms of harm affecting all children.

4.4 Reporting Systems and Institutional Trust

The fourth implication concerns reporting and institutional trust. Participants expressed willingness to report abuse, but fear, corruption, judgement and uncertainty could obstruct action. Similar tensions have been documented where adults conceptualise themselves as protectors but face weak or compromised systems (Turner et al., 2022). For early childhood and lower-primary settings, a useful guide should therefore do more than ask whether participants would report. It should identify locally verified contacts, explain what information is required, clarify confidentiality, specify immediate steps when a child is in danger and provide a private route for sensitive disclosure. Schools should also monitor sudden absences, withdrawals, injuries, hunger, and declining participation as possible indicators that require supportive enquiry rather than punishment.

4.5 Contribution of Faith and Community Structures

The fifth implication is the potential contribution of faith and community structures. Faith-based participants reported taking training messages to congregations and challenging norms such as not taking children to school. They also provided material and feeding support. These networks may extend child-protection communication beyond schools, particularly where formal services are distant. Nevertheless, faith-based support should complement, not replace, trained statutory and professional services. Prayer, charity and moral guidance cannot substitute for medical care, formal reporting, psychosocial support or legal action when abuse occurs.

4.6 Implications for the Protection-and-Learning Guide

The findings also clarify how the discussion guide should be described. Respondents did not directly evaluate the guide or state that its use had improved children’s learning. What they did provide were community-generated explanations and actions that could inform the development of a locally adapted protection-and-learning tool. Its value would lie in structuring dialogue, identifying risks, mapping resources and agreeing responsibilities. Evidence from school-based intervention research shows that organised whole-school approaches can reduce violence (Devries et al., 2015), while reviews of maltreatment and education emphasise the importance of family, school and welfare systems (Romano et al., 2015). A future version of the guide should therefore be piloted prospectively and linked to measurable outcomes such as attendance, retention, classroom participation, early literacy and numeracy.

5. Conclusion

The findings support the central claim, but only at the level of qualitative, community-perceived pathways. Across five stakeholder groups, participants consistently associated children’s education with school attendance, learning materials, nutrition, health, parental love, responsible supervision, safety and access to community support. They also identified poverty, drought, hunger, harmful norms, unsafe spaces, sexual exploitation, early marriage, weak parenting and unreliable reporting as threats capable of disrupting children’s wellbeing and participation in school. The evidence therefore supports the proposition that child protection strengthens the conditions required for learning, while protection failures weaken those conditions.

The findings do not support a causal conclusion that child protection interventions or the focus group guide improved learning outcomes. No literacy, numeracy, attendance frequency, or classroom engagement measures were collected, and participants were not directly asked to assess whether the guide itself improved learning. The conclusion must therefore remain limited to perceived mechanisms, local priorities and plausible pathways.

The justified implication is that the guide elicited relevant community knowledge and identified priorities for integrated child protection and early learning programming. A simplified protection-and-learning tool may be developed from the identified themes to structure community dialogue, map risks and resources, clarify responsibilities and strengthen referral planning. Its educational effect should, however, be evaluated prospectively using qualitative feedback alongside independent child-level indicators such as attendance, retention, classroom participation, early literacy and numeracy.

6. Limitations

Focus groups are vulnerable to social desirability bias, conformity, and dominance by more vocal members. Sensitive issues such as sexual abuse, corruption, harmful parenting and violence by teachers may have been underreported. Group discussion is not suitable for the disclosure of confidential case information and cannot replace professional risk assessment or case management.

The guide did not directly assess its own usability, acceptability or effect on learning. Any statement that respondents believed “the tool” improved learning would exceed the supplied evidence. Future piloting should include explicit tool evaluation questions, private follow-up options, referral instructions, and observation of whether agreed actions are implemented.