Self-explanation may be more effective than presenting pupils with an explanation

Researchers at Simon Fraser University in Canada conducted a meta-analysis on research that investigated learning outcomes for pupils who received self-explanation prompts while studying or solving problems. Self-explanation is a process by which pupils use prior knowledge to make inferences in order to fill in missing information or monitor understanding.

Their study, published in Educational Psychological Review, examined 69 independent effect sizes from 64 studies (5,917 participants). Studies had to include a treatment condition in which learners were directed or prompted to self-explain during a learning task, with a comparison treatment where learners were directed not to self-explain. The measure was a cognitive outcome such as problem solving or comprehension. Learning activities were mostly of short duration (less than an hour) and carried out mostly with undergraduate students.

The analysis found an overall weighted mean effect size of +0.55 on learning outcomes for pupils who were prompted to self-explain compared to those who were not. However, most of the studies were very brief and artificial, so the outcomes cannot be assumed to apply to actual classroom practice. Moderating variables were also examined in order to investigate how learning outcomes varied under a range of conditions, but were found to have no significant difference on effect sizes. The study concludes that having pupils come up with an explanation themselves is often more effective than presenting them with an explanation.

Source: Inducing self-explanation: a meta-analysis (September 2018), Educational Psychological Review, Volume 30, Issue 3

Parental engagement programme has mixed impacts in early education

An evaluation of the Education Endowment Foundation’s trial of Families and Schools Together (FAST), delivered by Save the Children, did not appear to make a difference to children’s achievement, but was found to be an effective mechanism for engaging parents in their children’s early education. FAST was also shown to have a positive impact on children’s social and behavioural outcomes across the whole year group and not just for the children who participated in the programme.

FAST is a parental engagement programme that aims to support parenting and enhance links between families, schools and the community. Parents and their children attend eight weekly two-and-a-half-hour group sessions delivered after school by accredited FAST trainers.

The school-level randomised trial measured the impact of FAST for the whole year group on Key Stage 1 (KS1) reading and arithmetic achievement, and children’s behavioural and pro-social outcomes (measured using the Strengths and Difficulties Questionnaire). One hundred and fifty eight schools took part in the trial, with a total of 7,027 pupils across the Year 1 cohort in these schools, and 632 pupils taking part in the eight-week programme.

The evaluation found no evidence that FAST had an effect on KS1 reading and arithmetic outcomes for the whole year group (effect size = +0.01). There was also no evidence that FAST had an impact on KS1 outcomes for the children whose families took part in the eight-week programme. However, FAST showed some promise on non-academic outcomes, with positive outcomes for the whole year group. Immediately after the eight-week programme, Year 1 pupils in the intervention schools had a higher average pro-social score and a lower average total difficulties score than pupils in comparison schools. However, these effects diminished by the end of Year 2.


Source: Families and Schools Together (FAST) evaluation report and executive summary (November 2018), Education Endowment Foundation

Planning ahead for summer

Heather L Schwartz and colleagues from the RAND Corporation have released a final report on a six-year study of the National Summer Learning Project, an initiative from The Wallace Foundation that was implemented in 2011 in five urban school districts in the US. The summer programmes in these districts were district-led, voluntary summer learning programmes that featured both academic teaching and enrichment opportunities to improve outcomes for low-income pupils.

The overall study combined a randomised controlled trial with correlational analysis and implementation research to examine whether voluntary, district-run summer learning programmes can improve academic, behavioural, and social and emotional outcomes for low-income, urban children in both the short and long terms. The study followed approximately 5,600 pupils from third to seventh grade (Years 4 to 8). Data included surveys, observations and test data.

Findings showed that pupils who received a minimum of 25 hours of mathematics teaching in a summer performed better on the subsequent state maths test, and those receiving 34 hours of English lessons performed better on the subsequent state English language assessment.

These outcomes need to be viewed with caution, however, as pupils who actually attended summer school, as opposed to those who signed up but did not attend, are likely to be more highly motivated and better achieving, introducing possible bias.

Based on their research, the authors offer several recommendations for planning for summer learning, including:

  • Commit in the autumn to a summer programme, and start active planning by January with a programme director who has at least half of his or her time devoted to the job.
  • Prior to the start of the summer programme, professional development for summer teachers should include specific guidance on use of the summer curricula, minimising loss of teaching time, and on checking for pupil understanding.
  • Operate the programme for five to six weeks with three to four hours of academic lessons per day.

A more detailed and comprehensive list of recommendations can be found in the report.

Source: Getting to work on summer learning. Recommended practices for success, 2nd edition (2018), RAND Corporation

The effects of cooperative learning in middle school on reducing bullying

While many studies show positive effects of cooperative learning on pupil achievement, a recent study examined the effects of cooperative learning on reducing bullying in middle school.

A total of 15 rural schools (n=1,460 seventh graders) in the Pacific Northwest were matched based on size and free-lunch percentage, and then seventh graders (Year 8) were randomly assigned to either receive a cooperative learning programme (n=792) or to continue business as usual (n=668). The cooperative learning programme used techniques by Johnson & Johnson, incorporating peer tutoring, collaborative reading, and methods where classmates rely on each other to learn new information while being held individually accountable for what they have learned. The theory behind this study was that in cooperative groups, bullies would not be reinforced by their peers to continue bullying, and socially isolated pupils would have opportunities to interact with others more and make new friends. All participating teachers received a copy of Cooperation in the Classroom and received three training days in person, and check-ins by video conference during the course of the 2016–17 school year. Pre-tests and post-tests (online surveys completed by pupils) evaluated pupils’ bullying and victimisation, stress levels, emotional problems, relatedness and engagement.

After five-and-a-half months of the cooperative learning programme, results showed significant reductions in bullying (effect size = +0.37), victimisation (+0.69), and stress levels (>+0.99) for pupils who had been shown to be marginalised at pre-test, and reduced emotional problems (+0.30) and greater relatedness (+0.43) for all pupils, regardless of their feelings of victimisation/isolation at pre-test.

Source: Cooperative learning in middle school: A means to improve peer relations and reduce victimization, bullying, and related outcomes (November 2018), Journal of Educational Psychology, 110(8)

The impact of web-based coaching on mathematics instruction

A new working paper, published by Brown University, reports on an online coaching programme that aimed to support maths teachers.

Mathematical Quality of Instruction (MQI) is an observational instrument that helps structure teachers’ and coaches’ reflections about maths teaching. There are four “dimensions” of the instrument – richness of the mathematics, Common Core-aligned pupil practices, working with pupils and mathematics, and teacher errors. In several studies, teachers’ scores on MQI have predicted pupils’ academic achievement gains.

In MQI coaching, teachers selected an element of their practice to work on and filmed one of their lessons. A coach then selected a couple of extracts and chose a comparison stock film clip. The teacher watched the clips and then the two discussed them and developed a plan for improvement.

In the current trial, 142 upper elementary and middle school teachers (Years 4-9) from 51 schools in a mid-western state in the US were assigned to MQI coaching or a control condition. Teachers assigned to MQI coaching took part in a two-day summer school, followed by a bi-weekly coaching cycle for the following academic year.

At the end of that year, the coached teachers showed substantial improvements in their scores on the MQI instrument. They also had improved scores in pupil perceptions of classroom practices. However, there was no measurable impact on state achievement tests (the Smarter Balance Assessment Consortium (SBAC)). In the follow-up year, there were still large impacts on the four MQI dimensions, but none were found on pupils’ assessment of teachers or pupil achievement.

Source: Developing ambitious mathematics instruction through web-based coaching: A randomized field trial. (November 2018). Brown University Working Paper

IPEELL writing intervention finds scale-up challenging

The Education Endowment Foundation has published the results of a randomised controlled trial of IPEELL

The IPEELL intervention is a writing process model in which pupils are encouraged to plan, draft, edit, and revise their writing. IPEELL stands for Introduction, Point, Explain, Ending, Links, and Language. The strategy provides a clear structure to assist writers and can be used for most genres of writing, including narrative writing. In addition to the writing process, the IPEELL intervention also involves ‘memorable experiences’ for pupils designed to act as a stimulus for their writing.

The trial tested the impact of one year of IPEELL for children in Year 6 and the impact of two years of IPEELL for children who started it in Year 5 and continued in Year 6. In total, 84 schools and 2,682 children in the north of England participated in the one-year trial and 83 schools and 2,762 children participated in the two-year trial. Writing outcomes were measured using Key Stage 2 (KS2) writing outcomes for the one-year trial and a bespoke writing test based on historic KS2 writing tests for the two-year trial.

The results showed that pupils who used IPEELL for two years made more progress in writing (effect size = +0.11) than pupils who did not. However, they made less progress in reading, spelling and mathematics than pupils in the control group (ES = -0.17—0.30). Pupils who used IPEELL for one year made less progress in writing, reading, spelling and maths than comparison pupils.

A previous trial of the approach had shown large positive results, but there were important differences between the two trials. In this latest trial, the model used teacher trainers who had never seen IPEELL delivered in the classroom. It also measured the average impact across all pupils, while the first looked only at pupils with low prior attainment. In this latest trial, pupils with low prior attainment who used IPEELL for two years made more progress in writing (effect size = +0.26) than pupils who did not – a larger effect size than the figure for all pupils.

Source: Calderdale Excellence Partnership: IPEELL evaluation report and executive summary (November 2018), Education Endowment Foundation