Education - Safe Learning Model Research

SRGBV prevention and response mechanisms are embedded in all Concern education programmes and we are committed to the development of models of best practice and lessons learned from SRGBV programmes.

The Safe Learning Model adopts a holistic approach to education that brings together transformative gender interventions within an education programme. The model, being piloted and evaluated in Sierra Leone from 2017-2021, is based on the assumption that children’s educational progress will be enhanced when they live in communities that are underpinned by support for gender equality and children’s wellbeing. This three level approach to learning seeks to counter the issues of SRGBV and gender stereotypes, which can keep children from learning or cause them to drop out of school.

Three levels of intervention:

• Comprehensive Literacy Interventions including competency based Continuous Teacher Professional Development
• School Level SRGBV Prevention & Response including transformative approaches with teachers and school management and age appropriate Social and Emotional Learning with children
• Community Level SRGBV Prevention & Response including family level transformative gender approaches and community dialogue

In addition to the interventions, the Safe Learning Model includes a research component being conducted in partnership with University College Dublin (UCD) School of Education. Aimed at testing whether the different levels of intervention are effective and how they interact, the study is divided into two phases: a pilot phase in the first year (2017-2018) and the main trial from 2018 to 2021. The rigorous mixed methods randomized controlled trial combines quantitative measurements with thick qualitative descriptions of children’s everyday lives to create a comprehensive picture of the multi-faceted intervention. The longitudinal study will engage with 100 schools randomly allocated to 4 levels of intervention and follow approximately 3,000 children in their first three years of primary education in Sierra Leone.

During the period between September 2017 and May 2018, Concern Worldwide and the University College Dublin (UCD) School of Education conducted a pilot of the Safe Learning Model; an integrated holistic approach to education programming, in 10 schools (five intervention and five control) in the Tonkolili district of Sierra Leone.

The pilot phase allowed the implementation team to refine methods of delivering the Safe Learning Model as an integrated programme and enabled the UCD research team to explore the feasibility of evaluating the intervention. During the pilot, the research team were able to gain insight and understanding of children’s everyday lives in Sierra Leone and develop research tools and protocols that could be used during a full-scale Safe Learning Model.

Findings:

• Children reported higher levels of feeling good and doing well when they had enough water, felt healthier, and did more chores.
• Analysis of data collected at endline indicated that intervention group performed better overall than the control group on the EGRA.
• Caregiver literacy was associated with EGRA performance, with children who reported that their caregiver could read or write scoring significantly higher.
• Performing chores at home more often was associated with higher performance on the EGRA suggesting potential inter-linkages with feelings of competency and self-efficacy and reading performance.
• Statistically significant, positive correlation between literacy and wellbeing scores at endline.
• Children observed positive attitudes towards boys education more often than positive attitudes towards girls education.

The main trial of Concern’s Safe Learning Model intervention is being evaluated by University College Dublin (UCD) School of Education and involves tracking the same children (and their communities clustered into four groups of tiered level of intervention) over three academic years.

In September 2018, the baseline was conducted with over 3,000 children as they entered their first year of primary education in the 100 schools taking part in the randomised control trial. The mixed methods approach includes the administration of structured questionnaires and assessing children’s literacy levels using an Early Grade Reading Assessment (EGRA) with all children each year of the intervention, along with interviews of their teachers and head teachers (100). The study also includes an in-depth exploration of four case study communities (drawn from the four arms of the trial) to include Class 1 children, their teachers, headteachers and a selected subsample of 16 children and their extended families (parents and elders / grandparents).

Findings:

• Most Class 1 children engage in domestic work with evidence of gender labour division.
• There are differences in expectations for schooling for girls and boys, with lower expectations for girls.
• Girls have lower expectations for their own ‘brilliance’.
• About 70% of respondents (boys and girls) are whipped or caned regularly by parents and teachers.
• In spite of the levels of corporal punishment, an overwhelming majority of teachers identified schools as spaces where both boys and girls are ‘very safe’ or ‘somewhat safe’ (e.g. free from bullying, intimidation, physical violence, etc. in school or on the way to school).

In May 2019, post-test one was conducted with the approximately 3,000 children taking part in the randomised control trial as they finished their first year of primary education in the 100 schools in Tonkolili Sierra Leone. The quantitative and qualitative data collected provides further insight into the children’s lived experiences both in and outside of school. As the study remains ongoing, and in order to maintain research integrity, results directly related to the effectiveness of each level of the model will not be released until the conclusion of the study and as such, it is not yet possible to make assumptions related to the effectiveness of the various levels of intervention. However, over the first year of intervention, evidence emerges of small but statistically significant changes in literacy and wellbeing outcomes among the study population.

Findings

• Higher increases in average child wellbeing in intervention compared to control schools, with increases significantly higher among girls than boys.
• Significant decrease in EGRA zero scores (baseline of 45% decreased to 26% after one year) with no significant gender differences, but significantly higher improvements in intervention schools for some tasks.
• Significant, substantial increase in proportion of children reporting domestic chores, or work outside the home, but this did not reduce play time.
• Girls tend to do domestic chores more frequently than boys, but both boys and girls experienced a similar increase in this work.
• 56% of the children reported being whipped or caned by their teachers and about 59% by their parents.
• A higher proportion of boys were whipped or caned by teachers and girls by their families, but gender differences were not significant.
• Fewer incidences of direct violence (physical and psychological violence) reported between baseline and post-test 1. Lower average level of direct violence in intervention compared to control, but differences were not significant.
• Children’s perception of positive attitudes toward gender equality among teachers increased, but among family decreased from baseline to post-test 1.
• In schools that teachers considered as very safe for pupils, children perceived a higher level of positive gender equality attitudes from teachers in comparison to schools not considered as very safe.

Within the overall Safe Learning Model research framework, a sub-study was conducted on the impacts of COVID-19 on schools, children and their learning. COVID-19 has had, and continues to have significant impacts on all sectors of social, economic and political life in almost all countries of the world, with schools in Sierra Leone officially closed on 31st of March 2020.

This sub-study provides information on the implementation of school closures in Tonkolili district and on the immediate and perceived impact on children, teachers and school principals. The sub-study utilised a mixed method approach including quantitative structured interviews with head teachers from 77 of the 100 schools participating in the main Safe Learning Model study and qualitative semi-structured in-depth interviews with four head teachers of schools that are part of the in-depth case studies for the wider Safe Learning Model study. The data collected for this sub-study has the potential to map the impact of the school closures in the context of the Safe Learning Model with a view to informing subsequent analysis of both the ‘moment in time’ impact and the impact over the duration of the implementation of the Safe Learning Model intervention.

Findings:

• Schools were ‘a little prepared’ or ‘not prepared’ for closures.
• Perception of little communication or Government support.
• Closures have had a negative impact on learning outcomes.
• Schools receiving support for continued learning more likely to stay in contact with pupils and engage in small learning groups.
• Schools with higher results prior to closures more likely to engage students in continued learning, widening gap in learning.
• Most children not utilising the radio programme, but are more likely to use it when engaged in learning groups with teachers.
• Children in hard to reach areas excluded from learning activities.
• Few pupils are engaged in learning with their family.
• Boys and girls are engaging in increased work, with girls disproportionately affected and exposed to protection risks.

In November 2020, and May 2021 post-tests two and three of the three-year longitudinal study with approximately 3,000 children, their teachers and caregivers from 100 primary schools in Tonkolili, Sierra Leone were conducted. The randomised control trial conducted over a three year period utilised a mixed methods design that included; 376 semi-structured interviews and group discussions along with participatory approaches such as photo voice, 42 hours of classroom observational data, analysis of school records, annual Early Grade Reading Assessments (EGRA) and the Child and Adolescent Personal and Social Assessment of Wellbeing (CAPSAW). The study examined wellbeing, gender equality, experiences of school related gender based violence and literacy development of a cohort of pupils and their everyday experiences in school and at home including examining the bidirectional relationship between literacy, wellbeing and safety and ultimately how this may be affected by the programme.

Findings

• The Safe Learning Model has positive impacts on basic Literacy skills and wellbeing, especially   for girls, despite significant challenges of endemic poverty, gender stereotypes, structural problems of the educational system and the impact of school closures during the COVID-19 pandemic.
• There is a clear link between Literacy progress and Wellbeing, particularly for girls.
• Overall, improvements in Literacy are present for both boys and girls over time.
• Literacy skills of children in the Safe Learning Model, in particular girls, generally improve more quickly than children not receiving the intervention.
• On average, boys perform better than girls across subtasks, but girls in the Safe Learning Model progress faster, closing the gender gap.
• There was no differences in teaching practices of qualified and unqualified teachers.
• Teachers in the Safe Learning Model identify letter sounds as important and spend significant time teaching this, which is reflected in children’s improved basic Literacy skills at an earlier stage.
• Teachers, principals and caregivers have high aspirations for the education of girls and believe it is necessary, but have lower expectations for their success in school with the majority believing boys are more brilliant than girls.
• Children’s psychological wellbeing, specifically ‘feeling good’ declined during COVID-19, however girls receiving the Literacy component of the Safe Learning Model show significant increases in ‘feeling good’ at the end of the intervention.
• Children that feel safer with a teacher are more likely to report having been whipped or caned by a teacher.

Coming soon:

• Safe Learning Model Literacy, Learning and Teaching Brief 2023 
• Safe Learning Model Wellbeing and Violence Brief 2023 
• Safe Learning Model Implementation Lessons Learned Brief 2023 

This research is being undertaken by University College Dublin- School of Education on the Safe Learning Model being implemented by Concern Worldwide in partnership with the Sierra Leone Ministry of Basic and Senior Secondary Education (MBSSE). The Safe Learning Model intervention and research is funded by a grant from the Irish Government, however the content within this publication is entirely the responsibility of the authors and does not necessarily represent or reflect Irish Aid Policy.