Category: Language Development

Categories
Educational Administration and Leadership K-12 Education Language Development

Learning to read: The dis-advantage of digital practices for younger learners

As digital devices replace physical books in classrooms and homes, many parents, teachers, and school administrators wonder whether this digital reading practice has the same benefits for developing readers as traditional books. Altamura, Vargas, and Salmeron tackled this question in a recently published study in the Review of Educational Research, specifically studying the effect of “leisure” reading on a digital device on reading comprehension. They reviewed 26 studies (representing 469,564 individuals) published between 2000 and 2022.

The authors found an effect size (correlation) of r = 0.06. This shows a link between greater leisure digital reading activities and greater reading comprehension. However, this effect size pales in comparison to prior effect sizes from reviews of print-reading leisure on reading comprehension, which averages closer to r = 0.35 for grades 1-12. The authors also identified substantive variation in this effect by age group: the relationship was negative for students in primary and middle school grades but positive for high school and college age, implying that digital reading in younger students has a negative impact on their reading comprehension.

Of note for interpreting this effect size, the digital reading activities included not just reading of virtual texts but other online “reading” activities including email, browsing the internet, chatting online, etc. However, the authors’ investigation comparing more text-based (“linear”) activities to more social-communicative activities did not reveal a significant difference in the relationship to reading comprehension.

These findings mirror conclusions found in other research studies of print vs digital reading that have been summarized in BEIB and may be accessed here.

 

Source: Altamura, L., Vargas, C., & Salmerón, L. (2025). Do new forms of reading pay off? A meta-analysis on the relationship between leisure digital reading habits and text comprehension. Review of Educational Research, 95(1), 53–88. https://doi.org/10.3102/00346543231216463… Read the rest

Categories
K-12 Education Language Development Maths and Science Learning

How early do STEM and verbal abilities stereotypes start among children?

In a recent meta-analysis on children’s gender stereotypes about STEM and verbal abilities, data from 98 studies across 33 nations involving more than 145,000 children were integrated. The findings reveal that verbal stereotypes favoring girls’ abilities (b = 0.19) are stronger than aggregated STEM stereotypes (b = 0.09). A closer look shows that at age 6, stereotypes reflected in-group bias—boys favoring boys, and girls favoring girls. These gender differences declined with age, and by age 16, the stereotypes tended to align with traditional gender roles: boys were favored in STEM ability, and girls in verbal ability.

Girls showed stronger pro-female verbal stereotypes than boys, and while verbal stereotypes increasingly favored girls with age, STEM stereotypes remained more stable. Girls’ STEM beliefs shifted from pro-female to pro-male around ages 10–12, while boys’ verbal beliefs flipped to pro-female around ages 8–10. By age 8, verbal stereotypes already significantly favored girls. STEM stereotypes varied by domain, with stronger male-favoring beliefs in computer science, engineering, and physics (b = 0.25), and weaker or negligible biases in general math (b = 0.06) and science (b = 0.09). Background also mattered: children identified as Black held weaker STEM stereotypes than those identified as White, especially among girls. Black children’s views were largely gender-neutral, while White children slightly favored boys.

These results suggest that stereotypes emerge early, and may shape children’s interests and academic choices. Addressing these beliefs calls for cross-disciplinary collaboration and greater research focus on understudied areas like computing, engineering, and verbal ability development. Interventions could aim to challenge or broaden stereotypes and mitigate their impact, guided by a clear understanding of when such beliefs typically take root.

 

Source (Open Access): Miller, D. I., Lauer, J. E., Tanenbaum, C., & Burr, L. (2024). The development of children’s gender stereotypes about STEM and verbal abilities: A preregistered meta-analytic review of 98 studies. Psychological bulletin, 150(12), 1363–1396. https://doi.org/10.1037/bul0000456… Read the rest

Categories
Higher Education Language Development

Will GenAI Chatbot Enhance Self-Regulated Learning and Reading Engagement for EFL Students?

A quasi-experiment by Pan and colleagues explored whether a generated AI chatbot, Reade, could enhance self-regulated learning (SRL) in reading strategy and reading engagement among English as a foreign language (EFL) students. The study involved 61 first-year English major students (over 80% female, mean age = 18.5) at a university in eastern China. Two classes were randomly assigned to a treatment (n=31) or a control group (n=30).

Using the research team-developed ReadMate platform, powered by ChatGPT 3.5 Turbo, students accessed Reade, which had two main functions: (1) recommending reading materials based on student proficiency and preferences, and (2) supporting self-regulated reading strategies through a Plan-Enact-Reflect approach. At the “Enact” phase, the treatment group students had access to full functions in ReadMate, including GenAI-empowered SRL interaction for further guidance on SRL strategies, while the control group used ReadMate without this feature. The 12-week study required participants to complete reading tasks via ReadMate as part of their coursework. Questionnaires reported self-regulated reading strategy use and reading engagement before and three weeks after the intervention.

After controlling for baseline scores, the results revealed that the treatment group significantly outperformed the control group in SRL strategy use across metacognitive, cognitive, motivational, and behavioral dimensions. Platform logs indicated that the treatment group accessed Planning Strategies and Self-Checklist panels more frequently. Additionally, the treatment group exhibited higher reading engagement across cognitive, behavioral, and emotional dimensions, as well as greater time and effort invested in reading tasks on the platform.

Although the study highlights the potential of GenAI chatbots like Reade in supporting SRL and reading engagement, its reliance on a small sample and self-reported data underscores the need for more rigorous research to explore AI’s role in enhancing EFL students’ reading outcomes.

 

Source: Pan, M., Lai, C., & Guo, K. (2025). Effects of GenAI-empowered interactive support on university EFL students’ self-regulated strategy use and engagement in reading. The Internet and Higher Education, 65, 100991. https://doi.org/10.1016/j.iheduc.2024.100991… Read the rest

Categories
Language Development Primary School Education Programme Evaluation Secondary School Education

Optimizing reading comprehension intervention : A Bayesian network meta-analysis

Peng and colleagues conducted a Bayesian network meta-analysis (BNMA) to evaluate the effectiveness of various reading comprehension strategies for students with reading difficulties (RD) in grades 3 to 12. They included 52 experimental/quasi-experimental studies which examined strategies including main idea (or summarizing), inference, text structure, retell, prediction, self-monitoring, and graphic organizers, along with their combinations. The researchers explored potential moderators like background knowledge instruction, age, text type, RD status, assessment type, intervention dosage, and study quality.

Unlike traditional meta-analysis, network meta-analysis can compare multiple interventions simultaneously and rank their effectiveness, even when some interventions haven’t been directly compared in individual studies, based on both direct and indirect evidence. The study found that the most effective strategy combination was the main idea-text structure-retell (MTR) combination (ES =+1.72, SUCRA = 0.89). Other effective strategies included the main idea-text structure-self-monitoring-graphic organizers (MTSG) combination (ES =+1.13, SUCRA = 0.76) and the main idea strategy alone (ES =+1.07, SUCRA=0.77). Background knowledge (vocabulary / content knowledge) was the only significant moderator that enhanced the effectiveness of these strategies. Without it, no strategies or combinations showed significant efficacy. The number of strategies in a combination did not influence the reading comprehension outcome, suggesting that teaching too many strategies could increase cognitive load, making it difficult for struggling readers to apply them effectively.

These insights can help educators design more effective reading interventions by focusing on optimal strategy combinations and incorporating background knowledge instruction to enhance comprehension for struggling readers.

 

Note: SUCRA = the surface under the cumulative ranking curve scores, which indicates how likely an intervention strategy would be evaluated as the most efficacious (the worst 0% to the best 100%).

 

Source (Open Access): Peng, P., Wang, W., Filderman, M. J., Zhang, W., & Lin, L. (2024). The Active Ingredient in Reading Comprehension Strategy Intervention for Struggling Readers: A Bayesian Network Meta-analysis. Review of Educational Research, 94(2), 228–267. https://doi.org/10.3102/00346543231171345… Read the rest

Categories
Kindergarten Language Development Programme Evaluation

Enhancing children’s reading outcomes: The interplay of topic background knowledge and parent-child interaction

Reading comprehension often depends on prior topic knowledge, as evident in everyday experiences like understanding technical manuals. Dong and colleagues studied how topic background knowledge (TBK) and parent-child storybook reading styles affect kindergarteners’ Chinese language skills, reading interest, and reading anxiety. TBK allows children to relate their experiences to the new information presented in the book. The study compared the effects of high-level TBK (H) and low-level TBK (L) across two interaction types: dialogic reading (DR), where adults and children discuss the story, and typical reading (TR).

The research randomly allocated 346 families (mean age of children: around 5.5 years) from the southern cities in China into four groups: H-DR, H-TR, L-DR, and L-TR They engaged in 25-minute reading sessions twice a week over eight weeks using the same eight storybooks. Children in high-level TBK groups (H-DR and H-TR) received pre-learning activities relevant to the reading material, while low-level groups (L-DR and L-TR) received activities using unrelated materials. Additionally, DR-group parents participated in workshops to enhance their interactive reading techniques, and hints of DR were provided along with storybooks.

The results, after controlling for background characteristics, revealed that:

  • DR groups: High-TBK children outperformed low-TBK children in all outcomes.
  • TR groups: High-TBK children showed better receptive vocabulary and listening comprehension.
  • Among high-TBK groups, DR outperformed TR across all measures.
  • In low-TBK groups, DR showed advantages only in vocabulary and listening comprehension.

This study highlights the benefits of prior knowledge of a book’s content combined with high-level interactive reading on children’s oral language skills and affective outcomes during parent-child reading sessions.

 

Source: Dong, Y., Chow, B. W.-Y., Xia, G., Mo, J., & Dong, H. (2025). Contribution of topic background knowledge to language learning outcomes through parent–child dialogic reading. Reading Research Quarterly, 60(1), e608. https://doi.org/10.1002/rrq.608… Read the rest

Categories
Educational Administration and Leadership Language Development Primary School Education Programme Evaluation

A meta-analysis of effective Spanish reading interventions

A meta-analysis conducted by our Best Evidence in Brief (BEiB) colleagues José Luis Arco-Tirado, Amanda Neitzel, and the late Bob Slavin, examined the effects of Spanish reading interventions for students in grades K–6. The review, published in Review of Educational Research, included studies focused on class- or school-based interventions aimed at improving reading skills among students whose first language was Spanish. The analysis included only high-quality studies that met rigorous criteria, such as the use of experimental or quasi-experimental designs and quantitative standardized measures of reading achievement.

The final best-evidence synthesis and meta-analysis was based on 11 studies with 51 effect sizes. Overall, Spanish reading interventions showed a significant average effect size of +0.49 on reading performance. While results varied considerably, the researchers explored grade level and outcome type as potential moderators of the effect. However, due to the small number of studies, no significant differences were found between grades K–2 and 3–6 or across the five categories of outcome type.

The authors concluded, on one hand, that some effective interventions for improving Spanish reading comprehension are available to help readers in K–6 education. On the other hand, more rigorous evaluation studies have to be implemented to expand the pool of proven reading practices available for teachers and educators in Spanish-speaking countries.

 

Source (Open Access): Arco-Tirado, J. L., Fernández-Martín, F. D., Hervás-Torres, M., Jiménez-Fernández, G., Calet, N., Defior, S., Neitzel, A. J., & Slavin, R. E. (2024). A Best-Evidence Synthesis and Meta-analysis on Effective Reading Programs in Spanish. Review of Educational Research, 00346543241297668. https://doi.org/10.3102/00346543241297668… Read the rest