A major new meta-analysis of academic studies on ways to improve the school results of low socio-economic status (SES) students has identified several interventions that substantially improve achievement. They include small group tutoring, feedback and progress monitoring and co-operative learning in the classroom. Several other interventions also have smaller positive effects on achievement.
The analysis covered 101 randomized controlled trials and quasi-experimental studies of intervention programs in elementary and middle schools in the United States aimed at low SES students. Outcomes were measured by standardized tests in reading and mathematics. The meta-analysis is published in the April issue of the academic journal Review of Educational Research. It concluded:
…. the review has clearly shown that there are interventions capable of substantially improving educational achievement of students from families with low SES. Tutoring, cooperative learning, and feedback and progress monitoring seem to be especially promising components of such interventions. In sum, the results of this review provide motivation to increase efforts both to implement interventions for low SES students, and to design studies that may answer the pending questions of what types of interventions are most cost-effective. [p. 276]
The analysis showed that tutoring, feedback and progress monitoring, small group instruction and cooperative learning have comparatively large and robust effects in improving reading and mathematics for low SES students. Tutoring and feedback and progress monitoring had the largest effects. In each case, the magnitude of the effect represented a substantial reduction in the achievement gap between high and low SES students.
Tutoring programs studied provided students with additional pedagogical support from an instructor, either one-to-one or in a small group (five students or fewer). Tutors could be volunteers, paid non-teachers, or professional teachers. The interventions included highly structured programs implemented over a limited time period, typically 12 to 20 weeks.
Feedback and progress monitoring included interventions that added a specific feedback or progress monitoring component, where teachers or students received detailed information about the students’ development. The objective was often to customize instruction to the individual student’s needs.
Cooperative learning, or peer-assisted learning, referred to interventions where students work together in pairs or small groups in a systematic and structured manner. Examples included students acting as pedagogical instructors for each other, as when more able students help less able students.
Small group instruction involved students placed in groups smaller than regular class sizes but larger than for tutoring programs.
Coaching/mentoring of school personnel, curriculum content changes, and increased resources also had positive and statistically significant effects on the achievement of low SES students. Coaching and mentoring involved providing teachers with a coach or mentor, and most often was connected to the implementation of a specific reading or mathematics program. Change in curriculum content mostly involved supplementary material to the broader curriculum that was directed at improving achievement. The meta-analysis included only two studies of increasing resources: one increased resources by increasing the supply of teachers, and the other gave stipends to students to attend higher performing private schools. Computer-assisted instruction programs were found to have a relatively large positive effect but it was not statistically significant.
While the effects of small-group instruction, coaching/mentoring of school personnel, content changes, and increased resources were positive and statistically significant, they were less robust in the sensitivity analysis conducted in the meta-analysis.
The study also found that several interventions directed at improving the results of low SES students had small positive effects that were not statistically significant. These included incentive programs involving monetary rewards to students or teachers, after-school programs, summer programs, coaching and mentoring of students, psychological or behavioural interventions, and personnel development.
The review of studies did not consider changes to the entire school system, such as changes to the grade system, the national/regional curriculum, and the introduction or expansion of school choice and private schools. It also excluded interventions where a group best practices are implemented in low-performing schools, whole-school reform strategy concepts and studies of different types of schools, such as charter schools. While some of these interventions, such as whole school reform, have proven successful in improving the educational achievement of low SES students, the review excluded these interventions because their complexity makes it difficult to identify the intervention components and distinguish their separate effects.
It also noted that effect sizes should not be the sole basis for choosing among interventions. It says that the cost-effectiveness of interventions should also be considered in choosing which to implement.
The study also reviewed explanations of the negative correlation between low SES and educational achievement. It found strong evidence that significant differences in cognitive development and school readiness between high and low SES students are present already prior to school starting age and that this has more to do with a range of factors affecting the development of young children than differences in innate abilities.
It found that recent research evidence from the United States indicates that hereditary factors are not a significant factor in the achievement of low SES students. For example, one study shows that there are no significant differences between children in high and low SES families on a test of infant mental ability at the age of 10 months. However, but by age 2 children in high SES families scored significantly higher than children in low SES families, indicating environmental factors are restricting low SES children from reaching their full cognitive potential. Another study found no differences significant differences on the same test among Hispanic, Asian, Black, and White infants aged 8 to 12 months, but large differences between Black and White children, which typically differ in SES, by age 3. Such evidence refutes claims by some that low SES and Black children have less innate ability compared to high SES and White children.
The study reports that a number of environment factors cause low SES students to enter school with fewer of the cognitive and social skills. Low SES children have worse health on a very broad range of measures, including foetal conditions, physical health at birth, incidence of chronic conditions, and mental health problems—problems which later influenced educational and labor market outcomes. Family environments are also different between high and low SES children in other aspects thought to affect educational achievement, from early on and onwards. High SES families are more likely to provide a rich language and literacy environment, have different parenting practices, and direct additional resources to early childhood education, health care, nutrition, and enriching spare-time activities.
This review of a large number of studies shows that it is possible for schools to substantially improve the educational achievement of low SES students by adopting several proven intervention programs. However, it was not possible to fully explain why some interventions worked better than others and the study emphasised that the impact of an intervention depends on the local context. The results of the review should be seen as a source of knowledge for policy-makers and teachers, but it does not provide a complete blueprint for how to increase the performance of low SES students.
Trevor Cobbold