Impact of shared book reading on children’s language development

A meta-analysis conducted by Claire Noble and colleagues explores the impact of shared reading interventions (where an adult reads with a child) on children’s language skills, and whether they are equally effective across a range of different outcome variables, for children from different socioeconomic backgrounds, and across a range of study designs.

The analysis included 54 studies conducted between 1989 and 2017. These studies included 316 effect sizes and 5,569 participants. Nine of the studies reported follow-up effects. Children in the studies were typically age 7 years or younger.

Their findings suggest that, while there is an effect of shared reading on language development, the effect size is smaller than suggested in previous meta-analyses (+0.23). They also found that the effect size is moderated by the type of control groups, and when compared to active control groups, is closer to zero (+0.04). In addition, the meta-analysis indicates only modest differences between types of language outcome, no effect for socioeconomic background, and a near-zero effect at follow-up.

However, given the low dosage of many of the studies included in the meta-analysis, the authors caution against the conclusion that shared reading interventions have no real effect on children’s language development.

Source: The impact of shared book reading on children’s language skills: A meta-analysis (October 2018), PsyArXiv

Examining research on the Big Lift preschool initiative

Celia Gomez and colleagues from the RAND Corporation have released a new research brief that examines Big Lift, a preschool to third-grade initiative designed to boost literacy skills and ensure that children are reading proficiently by third grade (Year 4). The initiative has been implemented in seven US school districts in San Mateo County, California, that have below-average third-grade reading levels. According to the brief, Big Lift seeks to improve third-grade reading through a set of four co-ordinated and integrated “pillars”: High-Quality Preschool, Summer Learning, School Attendance and Family Engagement.

The researchers have examined outcomes for two cohorts of pupils: Cohort 1 includes pupils in four districts who receive Big Lift services, and Cohort 2 an additional three districts. Data sources include early childhood cognitive assessments, kindergarten (Year 1) and first-grade (Year 2) entry forms completed by parents, and the San Mateo County Office of Education’s countywide data system.

The current research brief is part of a multiphase evaluation of Big Lift, and reports on findings after two years of implementation. Key findings are as follows:

  • Big Lift preschool children in the 2017–2018 kindergarten class were better prepared for kindergarten than demographically similar peers who did not attend preschool — but they were less prepared than similar peers who attended non–Big Lift preschool programmes.
  • Children who attended two years of Big Lift preschool were more kindergarten-ready than similar peers who attended only one year.
  • In the 2016–2017 kindergarten class, Big Lift preschool children had reading levels at the end of kindergarten and the start of first grade that were on par with similar peers who attended other preschool programmes and higher than similar peers who attended no preschool at all.

Source: The Big Lift preschool, two years in: What have we learned so far? (2018), RAND Corporation Research Briefs RB-10047-SVCF

Do pupils benefit from longer school days?

A study published in Economics of Education Review looks at the evidence from the extended school day (ESD) programme in Florida to determine whether pupils benefit from longer school days.

In 2012, Florida introduced the ESD programme, increasing the length of the school day by an hour in the lowest-performing elementary (primary) schools in order to provide additional reading lessons. The lessons had to be based on research, adapted for pupil ability, and include phonemic awareness, phonics, fluency, vocabulary and comprehension. Schools were selected using school-level reading accountability measures. For this study, David Figlio and colleagues looked at reading scores for all pupils in Florida between grades 3 and 10 (Years 4 and 11) using school administrative data from 2005–06 and 2012–13, and employed a regression discontinuity design to estimate the effect of lengthening the school day, looking at the different performance of schools either side of the cut-off point.

Results indicated that the additional one hour of reading lessons had a positive effect on pupils’ reading achievement. ESD schools showed an improvement of +0.05 standard deviations on reading test scores in the first year. The annual cost of the ESD programme was $300,000-$400,000 per school, or $800 per pupil.

Source: Do students benefit from longer school days? Regression discontinuity evidence from Florida’s additional hour of literacy instruction (December 2018), Economics of Education Review, Volume 67

Small class size has at best a small effect on academic achievement

Reducing class size is often suggested as a way of improving pupil performance. However evidence from a new Campbell systematic review suggests that reducing class size has at best only a very small effect.

The review summarises findings from relevant studies that measured the effects of class size on academic achievement. A total of 127 studies were analysed, including 45 studies that used data from the US Student Teacher Achievement Ratio (STAR) programme that reduced class sizes substantially in kindergarten to grade 3 (Years 1 to Year 4). However only ten studies, including four of the STAR programme, could be included in the meta-analysis.

Their analysis focused on effects on maths and reading and found a small positive effect of reducing class size on pupils’ reading achievement and a negative, but statistically insignificant, effect on maths. For reading, the weighted average effect size was +0.11, and the weighted average effect size for maths was -0.03.

For the four studies using data from the STAR programme, the researchers found a positive effect of smaller class sizes for both reading and maths. However, the average effect sizes were still very small and do not change the overall finding.

Source: Small class sizes for improving student achievement in primary and secondary schools: a systematic review (October 2018), Campbell Systematic Reviews 2018:10

The advantages of print vs. digital reading: A meta-analysis

A recent meta-analysis showed that paper-based reading yields better outcomes in reading comprehension than digital reading. In an article appearing in Educational Research Review, Pablo Delgato and colleagues from Spain and Israel analysed 54 studies from 2000–2017 comparing the reading comprehension outcomes of comparable paper and digital texts. They examined if one medium has an advantage over the other for reading outcomes, and what factors contribute to any differences found.

Results showed that paper text has an advantage over digital text (effect size=+0.21). Influencing factors favouring paper text include reading under time limitations, text type (informational or informational plus narrative), and publication year—later publications showed increased advantages for paper reading than earlier publications.

While the authors do not advocate getting rid of digital texts given their convenience, cost advantages and pervasiveness, they reflect that these study findings should be considered when pupils are required to perform digitally-related tasks under time constraints.

Source: Don’t throw away your printed books: A meta-analysis on the effects of reading media on reading comprehension (November 29018), Educational Research Review, Volume 25

One-to-one technology and pupil outcomes

An evaluation published in Educational Evaluation and Policy Analysis evaluates the impact of the Digital Conversion Initiative on pupil outcomes for one US school district in North Carolina.

The initiative provided laptop computers to every pupil from the fourth grade (Year 5) upwards, while also providing teachers with training on how to best use the technology in their lesson plans.

Marie Hull and Katherine Duch used administrative school data from 2005 to 2013 to determine the programme’s impact on maths and reading achievement for pupils in grades 4 to 8 (Years 5 to 9), as well as the impact of the programme on pupil behaviour. They compared the district’s data from before and after implementation, as well as data from neighbouring school districts without one-to-one programmes to determine the short- and medium-term effects.

Their results suggest there is potential for one-to-one laptop programmes to help improve pupil outcomes. They found that:

  • Maths scores for pupils improved by 0.11 standard deviations in the short term and 0.13 standard deviations in the medium term.
  • No significant change in reading scores in the short term, and mixed evidence of improvement in the medium term.
  • Time spent on homework stayed constant.
  • Pupils spent more of their homework time using a computer.

Source: One-to-one technology and student outcomes: Evidence from Mooresville’s Digital Conversion Initiative (September 2018), Educational Evaluation and Policy Analysis