Enhancing resilience among disadvantaged children through universal social and emotional learning

• Main Author: Pert, Kirsty
• Format: Dissertation
• Language: English
• Published: University of Manchester 2018

Socio-economic disadvantage has been empirically established as being a risk factor that contributes to poorer outcomes, including children's mental health and/or academic achievement (Bradley & Corwyn, 2002; Department for Eduction, 2013; Green, McGinnity, Meltzer, Ford, & Goodman, 2005; Hetzner, Johnson, & Brooks-Gunn, 2010). Given the longer term consequences of these negative outcomes, exploring ways to buffer the negative effects of socio-economic disadvantage is an important area for education research (Collishaw, Maughan, Goodman, & Pickles, 2004). Universal social and emotional learning (SEL) interventions, such as the Promoting Alternative THinking Strategies (PATHS) curriculum (Greenberg, Kusche, Cook, & Quamma, 1995), which aim to develop key skills through explicit teaching, yield great promise as an effective means through which to build resilience in children exposed to risk (Domitrovich, Cortes, & Greenberg, 2007; Durlak, Weissberg, Dymnicki, Taylor, & Schellinger, 2011; Greenberg et al., 1995; Humphrey, 2013). The literature base suggests that interventions are not always implemented as fully intended by programme developers, with consequences for the success of expected outcomes (Lendrum & Humphrey, 2012). However, despite its importance, the role of implementation variability in the achievement of outcomes is a neglected area of research, with many studies failing to include implementation data in their analysis (Durlak et al., 2011). The aim of the current study was to investigate the differential gains, in mental health and academic outcomes, after two years of exposure to PATHS, for children eligible for Free School Meals (FSM). A further aim was to examine the association between implementation variability (dosage, fidelity and quality) on outcomes for children eligible for FSM. A mixed methods design was used, with the qualitative strand providing complementary and explanatory data to the quantitative strand. The data was from the PATHS to Success cluster-randomised controlled trial, involving n=45 schools and N=5218 children (Humphrey et al., 2015). The mental health outcome was measured by the teacher-reported Strengths and Difficulties Questionnaire (SDQ), collected at baseline and after two years of implementation of PATHS. Academic attainment data was retrieved from the National Curriculum Test data for all pupils in Year 6 at the end of the trial. Multilevel Modelling (MLM) (Paterson & Goldstein, 1991) analyses were utilised in order to determine whether there were differential gains for children eligible for FSM, as well as exploratory analysis on the association between implementation variability and intervention outcomes for this group of children. Additionally, thematic analysis of 24 teacher interviews was conducted to provide supplementary data regarding perspectives of the implementation of PATHS. Results indicated that, while there was an initial difference in mental health and academic outcomes, for children eligible for FSM compared with their non-eligible peers at baseline, overall there were no significant positive gains for children eligible for FSM after undertaking PATHS. With regard to implementation variability, exploratory analysis found that there was not a significant association between dosage and mental health outcomes, but high dosage was associated with an increased mathematics scores for children eligible for FSM. High and moderate quality lessons predicted higher externalising symptoms, while moderate fidelity was associated with higher internalising symptoms, for children eligible for FSM. Neither quality nor fidelity predicted significant differences in academic scores. The qualitative findings revealed reasons why programme implementation varied. Additionally, teachers' views on the impact of PATHS overall provided depth to conclusions drawn from the quantitative data. The implications of these findings are discussed, along with directions for future research.