More outdoor play during preschool years linked to better mental health later in childhood

 

Children who spend more time playing outdoors between the ages of two and four may be less likely to develop emotional and behavioural difficulties later in childhood.

That’s according to new research led by the University of Exeter, published in the Journal of Child Psychology and Psychiatry. Previous studies have shown a link between outdoor play and children’s mental health, but this is the first study to explore how outdoor play in the early years relates to children’s mental health over time.

Most children have low levels of mental health difficulties that stay low across childhood, but some increasingly experience difficulties with their mental health and others have difficulties from an early age. These new findings suggest the more often children play outdoors as preschoolers, the more likely it is that their mental health problems will be low through to middle childhood (aged eight).

Researchers analysed data from 4,151 children from the Growing Up in Scotland cohort dataset and looked at symptoms of mental health when children were aged four, five, six, and eight years old. This included externalising symptoms - which are problem behaviours such as aggression, impulsivity and hyperactivity - and internalising symptoms such as anxiety and depression. 

The research found those who played outdoors more frequently at ages two, three and four were more likely to remain in a low-symptom, good mental health group through to middle childhood. Specifically, the results showed that for each additional day that a child plays outdoors in a typical week during the preschool years, the odds of that child having a healthy profile of mental health symptoms through to age eight increases by between six and 14 per cent.

Professor Helen Dodd from the University of Exeter led the study and said: “Our findings suggest that providing young children with more opportunities to play outside could be a simple, low-cost way to support better mental health and should be considered within public health, education and planning policies. This includes providing adequate funding for the provision and maintenance of playgrounds and protection for the range of spaces that children and families use for play, which include informal spaces close to home, parks and other green spaces. These public spaces are especially important for people without access to a garden.”

To isolate the effect of outdoor play, the researchers controlled for a range of other related variables including child sex, ethnicity, highest education level within household, number of physical conditions that the child experiences, working status of parents, and whether the family had access to a park within ten minutes of home and/or access to a garden.

Marguerite Hunter Blair OBE, chair of the UK Children’s Play Policy Forum, welcomed the study and said: “These findings clearly demonstrate the importance of play-based early interventions that can have a long-lasting positive impact on preschool children’s mental health. This evidence shows that our young children will benefit significantly from more play opportunities and better spaces to play. To support this, governments and local authorities must build outdoor play into key policies and work with communities to create and improve these essential play spaces."

The paper titled ‘Early Outdoor Play Predicts Trajectories of Child Mental Health in a Population-Based Cohort’ is published in the Journal of Child Psychology and Psychiatry.

The study, in collaboration with the University of Glasgow, University College London, and Complutense University of Madrid, Spain, was supported by funding from UK Research and Innovation (UKRI) and National Institute for Health and Care Research (NIHR).

ENDS

---

Journal

Journal of Child Psychology and Psychiatry

DOI

10.1111/jcpp.70175 

The Effect of Height on Adolescents' Body Image Perceptions

 This paper estimates the causal effect of height on adolescents’ body image, encompassing self-perceptions of weight, the accuracy of those perceptions, and weight-management aspirations. Using data from the National Longitudinal Survey of Youth 1997 (NLSY97), the authors isolate the effects of height on body image outcomes, finding that taller adolescents are more likely to perceive themselves as overweight, even holding BMI constant, and are less likely to underestimate their weight category. However, the direction of misclassification diverges by gender: taller boys are more likely to correctly assess their weight category, while taller girls are more likely to overestimate it. These perception shifts translate into behavioral responses for girls but not for boys. 

A one-inch increase in height raises the probability that a girl reports wanting to lose weight by 3.0 percentage points (6.2 percent), with effects concentrated among girls in the normal weight range. 

Taken together, these findings establish height as a salient determinant of body image in adolescence, operating independently of body mass.

Low-cost, remote parental engagement interventions via text messages etc have little effect

 The authors of this study conduct a meta-analysis of 82 randomized controlled trials across more than 20 countries to estimate the effects of low-cost, remote parental engagement interventions delivered through text messages, phone calls, and apps. They estimate a joint likelihood function that incorporates both written studies and unwritten studies identified through trial registries, funder records, research labs, evidence clearinghouses, and other sources. By also recording sample sizes for unwritten studies, the model estimates the distribution of standard errors, identifies write-up probabilities conditional on significance, and characterizes the file drawer by estimating effect distributions for written and unwritten studies. 

Bias-corrected effects are 0.05 SD for test scores, 0.07 SD for grades, 0.05 SD for attendance, and 0.03 SD for enrollment. In the best-identified domain, test scores, statistically insignificant results are still written up at high rates. 

The larger studies tend to estimate smaller latent effects, which could indicate that true effects are correlated with study precision, violating a common meta-analysis assumption. In smaller-sample domains, our approach helps identify selection probabilities by anchoring the absolute write-up rates. 

Any single study estimate is unlikely to dissuade adoption because parent interventions have high marginal value of public funds. Instead, future research is most valuable when it can explain heterogeneity across settings.

Parental cooperation with the kindergarten is the most important way to support preschoolers' academic skills

 

Research into the academic skills of five-year-old children shows that parents' beliefs and cooperation with their kindergarten are more important than the abundance of parental activities at home in supporting the academic skills of five-year-old children.

 

The study, conducted by Anne-Mai Meesak, Doctorate in Educational Sciences, involved more than 500 five-year-old children and 300 parents. The study assessed children's cognitive processes (i.e. attention and perception, working memory, thinking) and learning, language and mathematical skills individually on a tablet using an e-assessment tool for child development (LAHE). At the same time, parents answered a questionnaire about their beliefs, expectations, cooperation with their kindergarten and activities at home.

To verify the suitability of the assessment tool for assessing children's early skills, the children's results were also compared with teachers' assessments of the same areas. The teachers' assessments and the children's actual results were found to be largely similar. The exception was study skills, which may be more difficult for teachers to assess, as they were based on children's own assessments of their own interest, perceived ability and confidence levels.

The results showed that although children's early cognitive, learning, language and mathematical skills are interlinked, they can also be distinguished and assessed separately. Children's academic skills were significantly predicted by their own cognitive processes and learning skills. Although girls performed slightly better in language skills and Russian-speaking children scored slightly higher in study skills, the skills of Estonian five-year-olds were generally quite similar.

 

Parental beliefs and behaviour

The study’s most telling finding concerns parents' beliefs about their children. Parents who perceived their children as having cognitive difficulties engaged in the least amount of activities at home that supported their children’s skills. This means that the children who needed the most support received the least at home. At the same time, the amount of home activities did not depend on the parents’ educational level.

The study also revealed an interesting connection. Although higher parental expectations regarding the skills children need for school were associated with more frequent home activities, neither parental expectations nor home activities were linked to children's actual performance in language and mathematics. Instead, the parents’ active participation in kindergarten activities proved to be important. This supported both the children's academic skills and encouraged parents to contribute more to shared activities with their child at home.

 

What do the results show?

The results suggest that early childhood education in Estonia plays an important role in supporting children's early skills. Most five-year-olds in Estonia attend kindergarten, where learning and educational activities are play-based and follow the national curriculum, which allows children with different home backgrounds to achieve similar academic results. This ensures the equality and quality of early childhood education in Estonia.

From the societal aspect, the research highlights two findings that help ensure a child’s holistic development. Firstly, it is important to support parents in recognising their children's strengths and weaknesses, so that they can better support their children's early learning. Secondly, parents should be actively involved in kindergarten activities. Therefore, the research confirms that a child’s development is best supported by strong cooperation between home and kindergarten.

---

DOI

10.60518/etera/143 

Method of Research

Preparing future math teachers to teach data science

 When Eric Weber, professor and chair of mathematics at Iowa State University, talks about data science with future math teachers, he doesn’t begin with code, algorithms or buzzwords.

Instead, he asks them to imagine the scientific method — form a hypothesis, collect data, conduct experiments — running in reverse. 

“In data science, you don’t start with a hypothesis or prediction,” Weber said. “You start with the data that already exists — maybe numbers someone collected years ago, or information gathered for a totally different purpose — and you work backward. You look for patterns, connections or surprises in the data, and those clues help you figure out what questions you should even be asking. So, instead of testing a hypothesis, you’re discovering one.”

This definition is the basis for curriculum Weber and colleagues at Iowa State and the University of Northern Iowa (UNI) have designed to help prepare future math teachers to teach data science in high-school classrooms. Their work reflects a growing national consensus that data science literacy should be part of secondary education.

“Multiple professional societies in mathematics, statistics and mathematics education have released statements in support of teaching data science in high schools,” Weber said. “But while high schools are being encouraged to add data-science courses, the teachers expected to teach them often receive little to no preparation.” 

In a new paper published by Scatterplot, the MAA Journal of Data Science, Weber and his co-authors argue that future math teachers are the educators best positioned to take on this role — but only if their training programs give them the tools to do it.

“Our goal is to help close that gap with the curriculum we’ve created,” Weber said. 

Weber’s co-authors are Heather Gallivan, associate professor of mathematics education at UNI; Lydia Butters, a former math education student at UNI who now teaches at Cedar Falls (Iowa) High School; and Stephen Nathan Mercil, a former mathematics doctoral student at Iowa State who is now an instructor at the University of St. Thomas, Minnesota.

Teaching data science by starting with what teachers already know

The curriculum, which is a five-week, self-contained module delivered within coursework taken by pre-service math teachers at Iowa State and UNI, focuses on the relationship between math and data science.

“We want to show pre-service math teachers that data science isn’t a separate universe from the math they already study,” Weber said. “It’s built on it.”

Many data-science ideas, including modeling, optimization and visualization, grow directly out of algebra, geometry and calculus, so instead of focusing on coding or software, the curriculum module uses familiar mathematical structures to introduce new concepts, Weber said.

A regression line becomes a model. 

A classification problem becomes a geometry puzzle. 

An optimization routine becomes a function‑minimizing exercise.

Weber said this strategy helps pre-service teachers get past the intimidation factor.

“If we can break down the initial barrier of, ‘I don’t know what data science is,’ then their ability to make that transition becomes pretty quick,” he said.

A project shaped by timing and a growing need

The idea for this project began in 2019, when Weber and Mercil first piloted the curriculum at Iowa State. The first full run happened in spring 2020, just as the pandemic forced classes online, Weber said.

The project expanded after Weber teamed up with Gallivan, whose background in statistics helped merge the two universities’ approaches. Funding from the Iowa Space Grant Consortium allowed the team to refine the lessons and offer the curriculum at both campuses starting in 2023.

“The module has been taught every spring at Iowa State and UNI since then, and each year, we add improvements based on student feedback and classroom experience,” said Weber, who is also a member of a committee assembled by the Iowa Department of Education to help write data science learning standards for the state.

To help future teachers see how data science works in practice, the curriculum uses a mix of synthetic and real‑world datasets.

One set simulates animal‑tracking data — timestamps, locations and headings — to give students a chance to explore visualization, dimensionality reduction and prediction. Another uses housing data collected by local high‑school students, allowing pre‑service teachers to practice multiple regression and think about how they might guide their own students through similar projects.

These examples, Weber and team said, help teachers understand how data‑science questions emerge from the data itself — and not from a prewritten hypothesis.

Preparing teachers for an AI-driven world

Weber said a broader goal of the project is to prepare teachers for classrooms where artificial intelligence and automated decision‑making are already part of students’ daily lives, and to help future teachers understand the relationship between AI and data science (“they’re closely related,” Weber said, “but they aren’t the same thing.”).

“Data science is the bigger field,” Weber said. “It’s about using math, statistics and computer tools to make sense of data and find patterns.”

Artificial intelligence, he explained, is about creating systems that can do tasks that usually require human thinking. AI systems learn from data, so they depend heavily on the work data science does.

The link between data science and AI comes from machine learning, a part of AI that learns patterns directly from data.

“Machine learning uses the same math and statistics that data science uses,” Weber said. “Simply put, data science helps us understand what the data is saying, and AI uses that understanding to make decisions or take action.”

The U.S. Bureau of Labor Statistics projects data science jobs will grow 34 percent between 2024 and 2034, a rate that is significantly faster than the average for all occupations.

“Artificial intelligence is powerful, but we'll still need data scientists — humans in the loop,” Weber said. “AI systems don’t ‘think’ the way humans do; they learn patterns from large amounts of data and make predictions based on probability. Without someone who understands how that data was collected, what it represents and where it might be misleading, the results can be wrong or even harmful. Data scientists can interpret and contextualize the output of those systems.”

Early results show promise

The researchers’ curriculum has now run for four consecutive spring semesters at Iowa State and UNI, Weber said, adding that one former student is already teaching data science at a high school. 

Additionally, a pre- and post-assessment administered during the first implementation showed measurable gains in students’ understanding of data science concepts, suggesting the approach is helping future teachers build both confidence and competence.

Weber said these early signs reinforce the need for continued investment in teacher preparation.

“We hope to obtain additional funding that will help us expand our work and support teachers who are already working in the field with in-service programming and classes that could earn teaching licensure renewal credits,” Weber said.

Read the paper:  Weber, E., Gallivan, H., Butters, L., & Nathan Mercil, S. “Leveraging Mathematical Knowledge to Prepare Future Math Teachers to Teach Data Science,” Scatterplot, 3(1). Published online April 2026.

Relationship of social media content & mental health from the perspective of Latino youth

 A new book titled SocialsVoice shines a light on the relationship between social media content and mental health from the perspective of Latino youth—a group that engages with social media across multiple languages and cultural perspectives. Through concrete examples, the book presents a complex portrait of their experiences online, including both the mental health risks posed by certain content and the presence of supportive, anti-stigmatizing voices.

The book project was led by Melissa DuPont-Reyes, assistant professor of epidemiology and sociomedical sciences at Columbia University Mailman School of Public Health, working in partnership with research collaborators and participants. The book draws on findings from a research study focused on 41 participating Latino youths ages 13 to 24, and 28 of their parents, all recruited from community-based organizations across the United States. Unlike previous research that relied on surveys or app data, the book is based on participatory research. Youth didn’t just answer survey questions; they also shared social media clips they encountered and analyzed how those clips helped or harmed their mental health. (Download a copy of the book here.)

Through dozens of richly illustrated examples and reflections throughout the book, youth identify rampant stigmatizing content, including posts claiming mental illness isn’t “real,” minimizing depression, reinforcing stereotypes, and promoting toxic masculinity. At the same time, they point to evidence of a powerful, youth-led anti-stigma movement, featuring mental health education, symptom management, suicide awareness, and self-care strategies. The book also shows how Latino youth use social media to discuss stigma and social issues occurring in real life, like racism, immigration, vaccine hesitancy, school shootings, poverty, sexual assault, and LGBTQIA+ support.

“We hope that this book helps elevate youth voices to inform policies, practices, and programs concerning social media. Too often, youth voices are misunderstood or ignored altogether. The SocialsVoice project also exemplifies how participatory research approaches are a powerful, community-generated response to concerns about the safety and utility of social media,” says DuPont-Reyes.

Example of a Negative Social Media Post

A 22-year-old female study participant reacts to a video clip featuring a man speaking to the camera about depression who says, “That’s some made-up sh*it.”:
“In this clip, he states that depression isn’t real. It is self-preservation, but only being able to self-preserve and survive day after day instead of living makes a depressed person.”

Example of a Positive Social Media Post

A 17-year-old male participant reacts to a TikTok video of a woman speaking about how she gives herself time to process and act on her feelings:

“This clip can educate others on how important it is to allow ourselves time to deal with our problems and not just set them to the side or ignore them.”

“Mindfulness Behind the Screen”

The book also highlights how young people are learning to set boundaries, curate positive content, and use “mindfulness behind the screen.” In the words of one 16-year-old female participant: “Social media is both good and bad, because you could be randomly using it and a bad video pops up, and then it makes your mental health worse, and then it keeps happening. However, you could also use social media to look for better videos and be like ‘Oh, okay, it’s actually not as toxic as initially shown’ because then you’re actively making changes to your algorithm and making sure that it’s better for your mental health.”

About the SocialsVoice Study

SocialsVoice began with youth participants defining what they considered to be positive and negative mental health content. Then, the youth were randomly assigned to groups of either the positive or negative mental health–themed content and invited to share social media clips depicting their assigned theme. Throughout seven video-chat sessions, the youth discussed their thematic social media clips in their groups. The study concluded with youth co-creating their own videos about their research findings that their peers and parents would find relevant and useful. Youth and parent participants were invited to watch the co-created videos together during a virtual film screening event. Links to the videos are available in the book.

Performance of students with autism in interaction with the NAO robot

 This paper investigates NAO robot’s role in supporting autistic students in real classroom settings. Most prior research focuses on one-on-one intervention, lacking group-based evidence. The study uses a within-subject design with six autism spectrum disorder (ASD) students, comparing robot-assisted and regular classrooms across attention, communication, interaction, and emotion. 

Statistical analyses reveal significant enhancements in attention duration, communication frequency, interaction scores, and positive emotions, with reduced negative behaviors. NAO’s compact design and predictable interaction reduce intimidation and boost sustained engagement. 

Findings validate the feasibility of robot-integrated teaching for ASD, providing a basis for long-term curriculum design. 

Limitations include small sample size; future research will adopt larger samples and longitudinal tracking.

The work entitled “Classroom Performance of Students with Autism in Interaction with the NAO Robot” was published on Frontiers of Digital Education (published on April 25, 2026).

---

Journal

Frontiers of Digital Education

DOI

10.1007/s44366-026-0083-1