Abstract
Background
School- and curriculum-based outdoor programs may promote emotional and physical wellbeing in children. This systematic review aims to identify the mental, physical, and emotional outcomes of a variety of school- and curriculum-based outdoor-education programs.
Methods
We used online databases to identify English peer-reviewed journal articles that reported any outcomes of outdoor education programs that fell within our inclusion/exclusion criteria. We excluded studies if they were duplicate studies, the papers did not pass the title and abstract review, participants were not between the ages of 3 to 19 years, the studies were not a school- or curriculum-based program, the studies failed to report at least 1 health outcome, or the papers involved a homeschool program or college and university programs.
Results
We included 20 papers in our analysis. Eleven studies described cognitive outcomes; 9 studies described social, affective, and psychological outcomes; and 10 studies described physical activity outcomes.
Conclusions
School- and curriculum-based outdoor programs may have a beneficial role in the wellbeing and quality of life of children. These results support further research into the application of school- and curriculum-based outdoor-education programs to improve intellectual, social, physical, psychological, and health-related outcomes.
Introduction
Over the past decade, research regarding the general positive effects of nature and green spaces on various health dimensions for children has increased substantially.1-3 Generally, the benefits of nature on the emotional and physical well-being of children are clear.1-3 They have been observed in multiple settings ranging from green spaces in neighborhoods and urban cities to outdoor learning programs, forest schools, and school gardens. Specifically, studies looking at school- or curriculum-based outdoor-learning programs have reported numerous positive effects including improved concentration, prosocial behavior, increased student engagement, improved psychological well-being, and self-determination.1-3
Given the benefits of nature on children’s emotional and physical well-being, promoting and implementing time in nature can help to combat chronic illnesses children face. It Around 30% of school-aged North American children suffer from chronic conditions such as diabetes and obesity.4,5 In the last few decades, there has been an increase in the prevalence of children’s health concerns such as asthma and attention deficit hyperactivity disorder, and concerns associated with poor behavioral, social, economic, and environmental determinants of health. These same determinants underpin the cardiovascular, pulmonary, carcinogenic, and mental health conditions prevalent in adulthood, the seeds of which are planted in early life.6
Time spent in nature and green spaces, specifically on school grounds, seems to encourage children to interact with others in a way that promotes social skills and connectedness.6 Outdoor learning, in particular, has been shown to benefit communication, social interaction, and teamwork among peers.7 Natural environments have also been shown to have immense psychological benefits, helping mitigate childhood anxiety and stress and improving attention, engagement, focus, and cognitive functioning.6,7
Existing reviews of outdoor education provide great insight into its benefits. But there appear to be only a few reviews that focus solely on incorporating outdoor education into the school curriculum. The purpose of this systematic review is to collate the current literature on the health outcomes of outdoor learning in school-based education programs, by screening the literature on a global scale. The design of this study was adopted from the systematic review by Becker (2017), which explored the effects of regular outdoor-education classes on students’ learning, social, and health dimensions but with length restrictions.8
This review aims to identify the impact of outdoor education with school- or curriculum-based activities without restrictions on the length of the program.8 We also seek to clarify the type of outdoor activities that have been studied, the duration of programs, and the physical, cognitive, social, affective, psychological, and behavioral outcomes among children participating in curriculum-based outdoor education.
Methods
Search Strategy
On December 8, 2020, we conducted an internet-based literature search using PubMed, GreenFILE, APA PsycArticles, Education Source, ERIC, Web of Science, JSTOR, and APA PsycInfo electronic databases for English-language, peer-reviewed journal articles. We created keyword categories and used them in each database search using a combination of words from each of the categories. We combined terms using Boolean operators such as, “AND,” “OR,” etc. The search terms used within the databases spanned 4 categories: “population,” “intervention,” “setting,” and “outcome.” The following categories with their relevant search terms were used in different combinations to find pertinent peer-reviewed journal articles.
The first category was “population,” which included the search terms: children, youth, child, or student. The second category looked at the type of “intervention” the study used and included the search terms: outdoor education, outdoor learning, learning outside the classroom, experiential learning, expeditionary learning, forest school, nature school, environmental education, place-based education, out-of-classroom, outdoor teaching, green spaces, curriculum, outdoor curriculum, or nature curriculum. The third category included the “setting” of the studies, which included: school, schooling, classroom, high school, high school green space, restorative environments, nature learning, outside learning, elementary green space, or school green space. The final category included in the search was “outcomes,” which included: health, pediatric health, children’s mental health, physical health, emotional health, mental health, and/or wellness.
Eligibility Criteria
Our search identified studies that measured students in association with exposure to a formal school- and curriculum-based outdoor education program of any length involving children and adolescents (aged 3–19 years). We included all types of study designs (eg, control-group design, quasi-experimental design, and case studies) that reported on at least 1 student-oriented health outcome. There were no restrictions on publication periods or location of study. We excluded studies that included students in homeschool programs and college and university students.
Selection Process
We identified 519 articles using the identified search terms; 456 were unique. Using a group Mendeley account, 4 independent reviewers (DB, RO, ZD, and MW) screened titles and abstracts for relevance according to the a priori criteria. There were 151 articles that appeared to meet eligibility criteria, and these underwent full-text assessment; we distributed articles in such a way as to ensure each paper was being reviewed by a new reviewer. Full-text review resulted in 20 eligible articles. The process by which we obtained and screened articles for eligibility is presented in the PRISMA flow diagram below (Figure 1).
Figure 1: PRISMA 2020 Flow Diagram. Legend: The PRISMA diagram details the search-and-selection process applied during our systematic literature search and critical review. Adapted from: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi: 10.1136/bmj.n71.
Critical Appraisal Assessments
We used 2 checklists for assessment of methodological quality and the risk of bias: Child Care and Early Education Research Center (CCEERC) Quantitative Research Assessment Tool (Table 1) and the Joanna Briggs Institute (JBI) Checklist for Qualitative Research (Table 2). Reviewers (ZD and MW) independently assessed the risk of bias in the 20 included studies. For validity and replicability purposes, since this review was modeled after the Becker (2017) paper, all the modifications made to the 2 checklists were also adapted for this review.8 This included taking away question 6 from the JBI and adding the ethics question from JBI to the CCEERC.
Quantitative studies (see Table 3) were rated on a scale from 0 to 1+ using the (CCEERC) Quantitative Research Assessment Tool modified by Becker et al. 2017.8 The mean (M) score of all items was calculated, and papers were rated as having low (M<0.30), moderate (0.30≤M≤0.60), or high (M>0.60) methodological quality.
Qualitative studies were assessed using the JBI Checklist (see Table 4). Using the adapted criteria, we rated the qualitative studies based on 9 questions (see Table 2), and in order to answer the questions and provide a numerical rating, the values were based on: Yes=1, No=–1, and Unclear=0. Mixed-methods studies were assessed using both assessment tools.
The overall assessment is presented in Tables 3 and 4. Any disagreement between reviewers concerning methodological quality was resolved through discussion. Studies were not excluded on the grounds of low methodological quality.
Results
Descriptive Characteristics of Studies
Table 5 shows the main descriptive characteristics of the 20 included studies. Distribution of the studies consisted of 30% from Denmark, 20% from the United Kingdom, 20% from the United States, 10% from Germany, and 5% each from Sweden, Finland, Scotland, and Turkey. Sample sizes varied considerably across the studies, ranging from 10 to 671 participants; 13 of the 20 studies had sample sizes over 50. In terms of the age range, 65% of eligible studies looked at child participants above the age of 9 years, while the remaining 35% of eligible studies included children from ages 4-8 years of age. Specific age ranges for each eligible study are reported in Table 5.
The nature of the intervention, including the location, also differed between the studies, as reported in Table 5. Interventions included forest environments, school gardens, parks, field trips, nonschool nature-based environments, outdoor-learning courses, and a camp setting. Eight of the 20 studies had an intervention length longer than 8 months.
From the selected papers in Table 5, 50% were quantitative studies, 30% were qualitative studies, and 20% were mixed study designs. The methodologies also varied amongst the 20 studies; they included randomized controlled trials, nonrandomized cohort-control studies, noncontrolled pseudo-experimental studies, and case studies, which are reported in Table 5.
Studies that reported outcomes pertaining to intellectual and academic motivation totaled 55%, while the other 45% reported outcomes related to psychological well-being and affective and social development; 50% of studies also reported outcomes related to physical activity. Outcomes for each study are presented in Table 6.
Methodological Quality Assessment
The overall methodological quality of the quantitative studies (see Table 3) was moderate to high, with 46% of the studies falling within the moderate range and 40% falling within the high range, according to the CCEERC scoring tool. Only 2 of the studies were rated as being of low methodological quality. The reasons for the low ratings were: failing to mention response and attrition rate of the studied population, missing data were not specified, sample sizes were small, statistical technique and/or rationale for selected statistical method(s) was not properly explained, and insufficient information was provided with respect to ethical criteria.
The methodological quality for the qualitative studies (see Table 4), as rated via JBI, were predominantly high: 78% of the studies were classified as high methodological quality, and the remaining 22% were rated as moderate quality. The moderate ratings resulted from 2 questions that were not adequately answered throughout 22% of the research papers. These questions were whether the researchers had an influence on the participants and whether the voices and opinions of the participants were adequately represented in the studies.
Outcomes
Intellectual Development and Academic Motivation
In terms of intellectual development and academic motivation, 50% of eligible studies reported beneficial outcomes. The most commonly reported outcomes included increased motivation, autonomy, ambition, and competence.9-13 Outdoor-education programs were associated with improvements in mathematics, vocabulary, English, reading, science, technology, and engineering.14,15 Improvements in writing skills, linguistic skills, and psycho-motor skills were also reported.16,17 Outdoor-education programs were shown to increase children’s environmental knowledge and sensory experience by fostering the children’s connection to nature.13-19 Increased imaginative and exploratory play were also reported.19
Psychological Well-Being, Affective Development, and Social Development
Of the eligible studies, 40% reported on psychological well-being, affective development, and social development outcomes (Table 2). Two studies showed that health-related quality of life (HRQoL) improved after interventions utilized education outside the classroom (EOtC).15,20 Outdoor education was associated with an increase in prosocial behavior, a reduction in hyperactivity-inattention, and decreased mental health struggles and peer problems.10,11,21 Better teamwork among peers was reported, with students developing greater respect and better communication and interaction skills.13,22 Teamwork improved after outdoor education, and students experienced a greater sense of belonging.14 Jørring et al found that students’ social patterns were rearranged, promoting a more inclusive and collaborative environment for peers after students participated in EOtC.9 Students who participated in outside education gained improved self-esteem and confidence.18,22 EOtC allowed students to develop skills other than academics, such as running or learning how to light a campfire; this appeared to be particularly beneficial for academically lower-achieving students.9
Physical Activity
In terms of exercise, 50% of the eligible studies reported outcomes of outdoor-learning programs on children’s physical activity levels. The most commonly reported outcomes included decreased sedentary behavior, increased daily steps, and increased general physical activity.22-27 Improvements in physical motor skills, fine motor coordination, balance, and muscle strength were also reported.13,18
Discussion
This paper reviews the evidence for outcomes associated with school- and curriculum-based outdoor-education programs for children. Results from the included studies suggest overall positive effects in the domains of intellectual/cognitive development, social skills, psychological well-being, and physical activity. These results support the findings from previous reviews.
Our findings suggest that children who participated in outside education gained improved self-esteem and confidence compared to students who participated only in indoor learning.18,22 This is consistent with the review by Rickinson et al that investigated the impacts of school fieldwork and outdoor visits, outdoor adventure education, and school grounds/community projects on children’s health.28 These authors found that outdoor adventure education can positively impact students’ attitudes, beliefs, and self-perceptions, including increased feelings of independence, confidence, self-esteem, locus of control, self-efficacy, personal effectiveness, and coping strategies.28 The emergence of self-esteem and confidence is important in children’s development as they are able to acquire better behavior and social relationships amongst peers and in their society.29,30
This analysis found that peers were more helpful,22 had improved teamwork skills,13,14 and had a greater sense of community13 when nature-based environments were incorporated into an academic setting. The review by Wray et al also investigated the impact of outdoor interventions occurring in a school or park on physical activity levels and social connectedness in children and youth.31 They found that naturalized environments were a strong determinant of social connection and that exposure to nature and outdoor settings had a strong influence on physical activity levels.10
The impact of nature-based education goes beyond improved social cohesion and self-esteem. It was also found to increase physical activity levels,13,18,19,22-28 improve both coordination13 and fine and gross motor skills,18 and decrease cortisol levels.28 Time spent in green spaces has been documented to have numerous health benefits, such as alleviating stress and anxiety, in both observational and interventional studies,33 which further highlights the impacts of outdoor exposure on a variety of different health outcomes.
Recommendations for Knowledge Translation
“Nature prescribing” is a key strategy to address health.34 It is an important tool to prevent chronic conditions, such as obesity and diabetes, that children may face when developing long-term sedentary habits.6 It is a proactive and sustainable approach to pediatric health and can directly benefit the healthcare system financially by relieving the burden the system faces from chronic illness. However, there are a number of barriers to implementing nature-based outdoor-education programs, including concern over the health and safety of children, teachers’ lack of confidence in teaching outdoors, and lack of time, resources, weather, and support.35 Healthcare providers can support these efforts by advocating for policy changes that reduce these barriers, including prioritizing funding to implement opportunities and infrastructure for EOtC. Even small changes can harness the benefits of outdoor play, such as selecting naturalized play environments over equipment-based playgrounds and paved areas.36
'Nature prescribing' is a key strategy to address health
Recommendations for Future Research
This review revealed that outcomes of outdoor-learning programs on pediatric health are varied and multifaceted. Motivation, cognitive and social development, psychological well-being, and physical activity are all mutually influenced by outdoor education. More rigorous assessment of specific programs may be necessary in order to fully understand their benefits, shortcomings, and outcomes. Future research should be designed with rigorous methodology and more in-depth descriptions of interventions.37
Additional research is necessary to study the effect of learning in an outdoor setting for children younger than 9 years of age (65% of studies looked at children older than this). Ages 2 to 7 years are an extremely important phase of child development during which the central nervous system is tremendously influenced by the external world through sensory experiences.38 Providing outdoor-learning programs at a younger age may have preventative benefits against childhood chronic illnesses and subsequent adult health conditions.
Strengths and Limitations
This review followed PRISMA design and search protocols. Inclusion and exclusion criteria resulted in a different selection of studies than had been assessed in previous reviews, which yielded a unique analysis. We used several online databases for the literature scan, allowing for a broad range of study designs and outcomes. All included studies were critically appraised by 2 independent reviewers. Study quality was assessed using validated critical appraisal tools, and most studies were of moderate- to high-quality methodology. This was an improvement over the review by Becker (2017), which yielded mostly low to moderate ratings.8 This could be due to the different selection of articles, older studies, and a smaller sample of research overall. In general, newer studies seem to have more rigorous methodological quality than older studies, and this may contribute to the discrepancies found in the ratings. Differences between the assessments could also be due to possible inconsistencies in how the reviewers interpreted individual checklist questions. Low- to moderate-quality studies should be interpreted with caution. Most studies indicate that results are not meant to be generalized due to the uniqueness of individual programs and their teaching methods, settings, participants, etc. As stated by Becker et al, these ratings can be used to assess limitations in this area of research, and results should be interpreted with respect to the study design and its limitations regarding generalizability, validity, and reliability.8
Only English-language, peer-reviewed studies were evaluated. No gray literature was used, and the references of included studies were not scanned for relevant articles. This may have reduced the number of studies included in this review. All eligible studies were included in the analysis regardless of methodological quality, possibly compromising validity, reliability, and generalizability of the findings. It is difficult to avoid biases in these types of studies due to difficulties implementing randomization or blinding protocols. In some study designs, teachers, educators, and researchers play a large role in the development of research questions and other aspects of the intervention, which may pose a risk of confirmation bias. A few reasons for low-to-moderate validity in these studies included short study duration, lack of baseline evaluation, incomplete data presented, failure to provide sufficient information on ethical criteria, and small sample sizes.
Most of the studies included in this review were conducted in Denmark, the United Kingdom, and the United States. It is difficult to generalize these findings to other nations. More research on the impact of outdoor-education programs for children in different contexts is important in order to ensure these benefits are universal.
Conclusion
This review demonstrates there are myriad benefits to children spending time learning in an outdoor setting. Nature-based educational programs seem to increase children’s intellectual development, academic motivation, psychological well-being, affective and social development, and physical activity levels, while decreasing sedentary time. Given that positive health in childhood is a determinant of adult health, it is plausible that a more widespread integration of outdoor learning may improve not only pediatric health, but adult and population health as well.
This review can encourage and promote children spending time in nature. Having knowledge about local forest schools and outdoor-education programs can help foster open dialogue around the mental, emotional, cognitive, and physical health benefits of learning in an outdoor environment. Additionally, primary care providers can recommend to their patients these educational programs that incorporate outdoor learning and maximize the opportunity to increase connectedness with nature. This review can serve as inspiration to influence widespread funding and implementation of learning strategies that support all children in strengthening their connection to nature and to one another. Not only does this seem to benefit their health during childhood, but also into adulthood.
Tables
Table 1 CCEERC Quantitative Research Assessment Tool
Question (Q) | Score = (+1) | Score = (0) | Score = (-1) |
|---|---|---|---|
1. Population. Does the population that was eligible to be selected for the study include the entire population of interest? Or, is the eligible population a selective subgroup of the population of interest? | Eligible population includes entire population of interest or a substantial portion of it | Population represents a limited, atypical, or selective subgroup of the population of interest | Population represents a limited, atypical, or selective subgroup of the population of interest |
2. Randomized Selection of Participants. Were study participants randomly selected for the study? | Random selection | Nonrandom selection | No description of the sample selection procedure |
3. Sample Size. How many participants were selected for the study? Does the sample include enough participants from key subgroups to accurately assess subgroup differences? |
Sample size larger than similar studies
| Sample size the same as similar studies | Sample size smaller than similar study or sample size not given |
| 4. Response and Attrition Rate. What proportion of the selected sample completed the study? In longitudinal studies, what proportion of sample members participated in follow-up studies? | High response or participation rate (over 65% response rate, over 90% participated in follow-up studies) | Moderate to low response rate (response rates of less than 65%) | No information on response rate or participation rate |
| 5. Main Variables or Concepts. Are each of the main variables or concepts of interest described fully? Can the main variables or concepts be matched to the variables in the tables? | Accurately described and can be matched | Vague definition or cannot be matched | No definition of main variables or concepts |
6. Operationalization of Concepts. Did the authors choose variables that make sense as good measures of the main concepts in the study? Have these variables been used in previous studies or are they an improvement over previous studies? | Key concepts are measured with variables that make sense. Or, variables have either been previously used in research or are improvements over previous measures. | Key concepts are measured with variables that do not make sense, and variables have not been used in previous research studies | Variable operationalization is not discussed |
| 7. Numeric Tables. Are the means and standard deviations/standard errors for all the numeric variables presented? | Means and standard deviations/standard errors presented | Means, but no standard deviations/standard errors presented | Neither means nor standard deviations/standard errors presented |
8. Missing Data. Are the number of cases with missing data specified? Is the statistical procedure(s) for handling missing data described? | Number of cases with missing data are specified and the strategy for handling missing data is described | Number of cases with missing data specified, but these cases are removed from the analysis | Missing data issues not discussed |
9. Appropriateness of Statistical Techniques. Does the study describe the statistical technique used? Does the study explain why the statistical technique was chosen? Does the study include caveats about the conclusions that are based on the statistical technique? | Statistical techniques, reasons for choosing technique, and caveats are fully explained. | Statistical technique is explained, but the reasons for choosing technique or the caveats are not included. | Statistical technique, reasons for choosing technique, and caveats are not explained. |
10. Omitted Variable Bias. Could the results of the study be due to alternative explanations that are not addressed in the study? | All important explanations are included in the analysis | Important explanations are omitted from the analysis | Variables and concepts included in the analysis are not described in sufficient detail to determine whether key alternative explanations have been omitted |
11. Analysis of Main Effect Variables. Are coefficients for the main effect variables in the statistical models presented? Are the standard errors of these coefficients presented? Are significance levels or the results of statistical tests presented? | Model coefficients and standard errors or hypothesis tests for the main effects variables are presented | Either model coefficients or hypothesis tests for the main effects variables are presented | Neither estimated coefficients or standard errors for the main effects variables are presented |
12. Is the research ethical according to current criteria or, for recent studies, and is there evidence of ethical approval by an appropriate body? ** | Yes | Unclear | No |
**Question taken from JBI Critical Appraisal Checklist for Qualitative Research. Adapted from: https://www.mdpi.com/1660-4601/14/5/485/s1
Table 2 JBI Critical Appraisal Checklist for Qualitative Research
| Question (Q) | Yes (+1) | No (-1) | Unclear (0) |
|---|---|---|---|
| 1. Is there congruity between the stated philosophical perspective and the research methodology? | |||
| 2. Is there congruity between the research methodology and the research question or objectives? | |||
| 3. Is there congruity between the research methodology and the methods used to collect data? | |||
| 4. Is there congruity between the research methodology and the representation and analysis of data? | |||
5. Is there congruity between the research methodology and the interpretation of results? | |||
| 6. Is the influence of the researcher on the research, and vice-versa, addressed? | |||
| 7. Are participants, and their voices, adequately represented? | |||
8. Is the research ethical according to current criteria or, for recent studies, and is there evidence of ethical approval by an appropriate body? | |||
| 9. Do the conclusions drawn in the research report flow from the analysis, or interpretation, of the data? |
Adapted from: Lockwood C, Munn Z, Porritt K. Qualitative research synthesis: methodological guidance for systematic reviewers utilizing meta-aggregation. Int J Evid Based Healthc. 2015;13(3):179–187.
Table 3 Methodological quality assessment for quantitative studies
| Source | Q1 | Q2 | Q3 | Q4 | Q5 | Q6 | Q7 | Q8 | Q9 | Q10 | Q11 | Q12 | Mean | SD |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Becker 2019 | 0 | 0 | -1 | -1 | 1 | 1 | 1 | -1 | 1 | 1 | 0 | 1 | 0.25 | 0.86 |
Bølling 2018 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 0.83 | 0.39 |
Bølling 2019 | 1 | 0 | 1 | 0 | 1 | 1 | 1 | -1 | 1 | 1 | 1 | 1 | 0.67 | 0.65 |
Dettweiler 2017 | 1 | 0 | 1 | -1 | 1 | 1 | 1 | -1 | 1 | 1 | 1 | 1 | 0.58 | 0.76 |
Ekenga 2019 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | 1 | 0.42 | 0.51 |
Finn 2018 | 0 | 0 | -1 | 1 | 1 | 1 | 1 | -1 | 0 | 1 | 1 | 1 | 0.42 | 0.79 |
Gustafsson 2012 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | -1 | 1 | 1 | 1 | 1 | 0.67 | 0.65 |
Johnstone 2019 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 0.83 | 0.39 |
Mygind 2007 | 1 | 0 | -1 | 0 | 1 | 1 | 0 | 1 | -1 | 1 | 0 | -1 | 0.17 | 0.83 |
Quibell 2017 | 1 | 0 | 1 | -1 | 1 | 1 | 1 | -1 | 1 | 1 | 1 | 1 | 0.58 | 0.79 |
Romar 2019 | 0 | 0 | -1 | 1 | 1 | 1 | 1 | -1 | 1 | 1 | 0 | 1 | 0.42 | 0.79 |
Schneller 2017 | 1 | 0 | 1 | 0 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 0.67 | 0.49 |
Sprague 2020 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | 1 | 0.67 | 0.45 |
Trapasso 2018 | 1 | 0 | 0 | 0 | 1 | 1 | 1 | -1 | 0 | 1 | 1 | 1 | 0.5 | 0.67 |
Yildirim 2017 | 1 | 1 | -1 | 1 | 1 | 1 | 1 | -1 | 1 | -1 | 1 | 0 | 0.42 | 0.9 |
SD: Standard deviation
Table 4 Methodological quality assessment for qualitative studies
| Source | Q1 | Q2 | Q3 | Q4 | Q5 | Q6 | Q7 | Q8 | Q9 | Mean | SD |
|---|---|---|---|---|---|---|---|---|---|---|---|
Dettweiler 2017 | 1 | 1 | 1 | 1 | 1 | -1 | -1 | 1 | 1 | 0.56 | 0.88 |
Dismore 2005 | 1 | 1 | 1 | 1 | 1 | -1 | 0 | 1 | 1 | 0.67 | 0.71 |
| Jørring 2020 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 0.86 | 0.33 |
| O’Brien 2007 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 0.78 | 0.44 |
| Romar 2019 | 1 | 1 | 1 | 1 | 1 | -1 | -1 | 1 | 1 | 0.56 | 0.88 |
| Sprague 2020 | 1 | 1 | 1 | 1 | 1 | -1 | 1 | 0 | 1 | 0.67 | 0.71 |
| Trapasso 2018 | 1 | 1 | 1 | 1 | 1 | -1 | 1 | 1 | 1 | 0.78 | 0.67 |
| Wistoft 2013 | 1 | 1 | 1 | 1 | 1 | -1 | 1 | 0 | 1 | 0.67 | 0.71 |
| Zamani 2016 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 |
SD: standard deviation
Table 5 Descriptive characteristics of studies
| Study | Sample Size | Age/Grades | Country | Study Design | Data Collection Method | Intervention Period | Intervention Length |
|---|---|---|---|---|---|---|---|
| Becker et al., 2019 | 48 | Gr. 5 & 6 | Germany | Prospective longitudinal quasi-experimental design study | Used accelerometers and objectively measured cortisol levels and cfDNA through saliva samples | School year: 2013 - 2014 | One year; one outdoor school day per week in a forest environment |
| Bølling et al., 2018 | 311IG, 88CG | Gr. 3 - 6 (age 9 - 13) | Denmark | Quasi-experimental pre-test and post-test intervention study | SRQ-A, Danish Occupational Social Class Measurement | School year: 2014 - 2015 | One school year; 9 months IG: EOtC minimum 5 hours per week across one or two weekly sessions in school nature-based environments CG: teaching as usual |
| Bølling et al., 2019 | 511IG, 120CG | Gr. 3 - 6 (age 9-13) | Denmark | Quasi-experimental pre-test and post-test intervention study | SDQ | N/A | IG: A curriculum-based 5-6 day residential outdoor learning course “research weeks” CG: teaching as usual |
| Dettweiler et al., 2017 | 281 | Female (mean age 12.48 years) Male (mean age 12.49 years) | Germany | Prospective longitudinal study | Adapted SRQ-A, SDI | 2014-2016 | IG: A curriculum-based 5-6 day residential outdoor learning course “research weeks” CG: teaching as usual |
| Dismore & Bailey, 2005 | 671 | Gr. 5 (age 9-10) | United Kingdom | Intervention study | Children’s work from follow-up activities, teachers’ views of the project and data collected during two research days at Arethusa Centre with Head teachers, teachers, children, and parents | N/A | Outdoor based activity days in a school nature-based environment followed up with classroom work |
Ekenga et al., 2019 | 53 | Age 10-14 | United State of America | Pre-test/ post-test within subjects study | HRQoL Questionnaire | 2017-2018 | Thirteen weeks; Weekly STEM-based environmental classroom activity and 4 monthly nature based outdoor activity days in nature-based environments |
Finn et al., 2018 | 44 | Gr. 4 (age 9-10) | United States of America | Intervention study | Digiwalker SW-701 pedometer, accelerometers | N/A | Five day, five-hour per day outdoor education program at a camp |
Gustafsson et al., 2012 | 230 | Age 6-12 | Sweden | Quasi-experimental nonequivalent design study | SDQ - For each child, the SDQ was completed by his or her parents (parent-version) | N/A | Six months; 15 days of outside education in school nature-based environments between 14:00 h and 18:00 h for at least one hour |
| Johnstone et al., 2019 | 137 | Gr. 3 (age~7) | Scotland | Randomized control trial | Actigraph GT3X accelerometer, TGMD-2, Flanker Test, one minute basic number facts Test | August 2017-December 2017 | Ten weeks; 1 hour activity session per week (30 min of facilitated games and 30 minutes of free play) in school nature-based environments |
Jørring et al., 2020 | 10 | Gr. 6 | Denmark | Case study | Standardized tests, basic skills in mathematics MG test, observations, pupil interviews | School year: 2014 - 2015 | Nine months; 5 hours of EOtC per week in school nature-based environments |
| Mygind, 2017 | 19 | Age 9-10 | Denmark | Multidimensional cross-scientific research project | CSA Accelerometer | December 2000-May 2003 | Three years: spent every Thursday teaching in the forest |
| O’Brien & Murray, 2007 | 24 | Age 5-9 | United Kingdom | Case study | Collected storyboards and reflection posters, observations, questionnaires (completed by 4 parents) | 2002-2005 | Eight months: observed children in forest schools |
Quibell & Charlton, 2017 | 223IG, 217CG | Age 8-11 | United Kingdom | Matched groups design study | N/A | N/A | IG: six weeks outdoor learning program at the National Trust Site CG: teaching as usual |
| Romar et al., 2019 | 21 | Gr. 1-6 (age 3-9) | Finland | Mixed methods design study | Accelerometers (GT3X+ and wGT3X+), non-participant observations | N/A | Physical activity and sedentary time were compared for the same students between school times spent in a nature-based environment vs. traditional school days. Students had five out-door learning school days, and 4 traditional school days. |
| Schneller et al., 2017 | 663 total participants: 346IG, 317CG Participants with validdata: 201IG, 160CG | Gr. 3-6 (age 9-13) | Denmark | Mixed methods , quasi-experimental cross-disciplinary study | Axivity AX3 Accelerometer | School year: 2014 - 2015 | IG: At least 300 minutes of EOtC (either one or two weekly sessions) through field trips and trips to the park CG: teachers were asked not to regularly practice EOtC with the class (less than 300 min of EOtC) |
| Sprague et al., 2020 | 122 | Age 10-15 | United States of America | Pre-test/ post-test within subjects study | Pre-test/ post-test self-administered survey that measures HRQoL and STEM capacity | School year: 2018-2019 | Fifteen weeks; weekly STEM-based environmental education classroom lessons and monthly nature based outdoor field trips |
| Trapasso et al., 2018 | 59 | Age 7-9 | United Kingdom | Mixed methods of design study | Anthropometric data (BMI and BMI-Z Scores), uniaxial accelerometer ActiGraph GT1M, and PAQ-C | Spring/Summer 2014 | Twelve weeks; weekly sessions of forest school (2 hours per session) |
| Wistoft, 2013 | 98Students ,6Instructors, 153Parents | Gr. 5 | Denmark | Evaluation Study | Field observations, workshop observations, individual interviews with principal organizers and municipal consultants, focus group interviews with teachers, students, instructors, and parents | September 2010-June 2011 | Eight months; observation of children through the ‘Garden for Bellies Program’ in a school garden setting |
Yildirim & Akamca, 2017 | 35 | Age 5 (ranged from 58 - 66 months old) | Turkey | Quasi-experimental pre-test and post-test Intervention study | Observation | N/A | Ten weeks; 4 hours daily for 5 days a week of outdoor education in a school nature-based environment |
| Zamani, 2016 | 58 | Age 4-5 | United States of America | Observational study | Interviews, behaviour mapping data collection | N/A | During preschool hours, children were observed on outdoor education in manufactured zones, mixed zones, and natural zones |
BMI = body mass index, cfDNA = cell-free deoxyribonucleic acid, CG = control group, CSA = computer science and applications, EOtC = education outside the classroom, Gr = grade, HRQoL = health related quality of life, IG = intervention group, PAQ-C = physical activity questionnaire for older children, SDI = self-determining index, SDQ = strengths and difficulties questionnaire, SRQ-A = academic self-regulation questionnaire, STEM = science, technology, engineering and mathematics, TGMD-2 = test of gross motor development-2
Table 6 Reported outcomes
| Study | Intellectual Development & Academic Motivation | Psychological Well Being, Affective Development & Social Development | Physical Activity |
|---|---|---|---|
| Becker et al., 2019 | Strong negative association in the IG for the relative amounts of light PA on the cortisol PR levels (posterior mean = -0.728). The CG had a negative association between sedentary behaviour and MVPA in the cortisol PR. In the IG, there was a positive association between sedentary behaviour and MVPA in the cfDNA PR. To summarize, the amount of sedentary time is lower and MVPA is higher in the outdoor teaching setting when compared to indoors and the outdoor setting seems to be strongly associated with a decline of cortisol levels, whereas no such decrease was observed in the indoor teaching setting. | ||
| Bølling et al., 2018 | Regular exposure to EOtC was significantly associated with better intrinsic motivation compared to CG. Regular exposure to EOtC was associated with a better RAI score. School motivation declined from baseline to follow-up. Also, the higher the level of baseline motivation, the stronger the association between EOtC and intrinsic motivation. | ||
| Bølling et al., 2019 | Regular exposure to EOtC significantly associated with an increase in prosocial behaviour (p = .045), reduction in hyperactivity-inattention (p = .045), and peer problems (p = .045). The reductions in hyperactivity-inattention were greater in pupils of low SES than pupils of high SES (but were non-significant). | ||
| Dettweiler et al., 2017 | BPN satisfaction for the outdoor setting is higher than for indoors, resulting in a higher SDI. SDI, competence, autonomy, and student-teacher relatedness are higher in the outdoor context compared to an indoor context. | ||
| Dismore & Bailey, 2005 | Mathematics and vocabulary improved and children developed a positive attitude toward learning. | Outdoor based activities encouraged a more cohesive environment promoting teamwork and encouragement among the children and their peers. | |
Ekenga et al., 2019 | The statement of “science is important in my daily life,” became statistically higher after the intervention. Family support scores post-intervention were significantly higher than family support scores at pre-intervention (p<0.001). The overall HRQoL score was significantly higher at post-intervention than at pre-intervention (p<0.01). | ||
| Finn et al., 2018 | Children used scientific concepts to track their physical activity. | Steps per hour and minutes of MVPA per hour increased approximately threefold from baseline to the post-test. The average sedentary minutes per hour also decreased post-test. Steps per hour and MVPA per hour were significantly higher during the outdoor education program compared to in-school. Children were significantly more physically active in the outdoor environment compared to the traditional school environment. Using tracking devices, children were motivated to increase their steps taken and distance travelled. | |
Gustafsson et al., 2012 | Decrease in mental health problems (emotional symptoms, conduct problems, hyperactivity) for boys at the intervention school, and a slight increase or no substantial difference for girls, as compared to the reference school. | ||
| Johnstone et al., 2019 | There was a decrease in sedentary behaviour and an increase in light intensity physical activity and MVPA in the IG. The CG also had a decrease in sedentary behaviour and an increase in light intensity physical activity and MVPA. Baseline fundamental movement skills were assessed using the GMQ. The IG had an increase in GMQ score, and there were no effects in GMQ score within the CG. The IG had an increase in locomotor skill score, locomotor percentile, object control score, and object control percentile. The CG did not have an increase in locomotor skill score and locomotor percentile, but their objective control score and object control percentile increased. | ||
Jørring et al., 2020 | Pupils who were challenged in the normal classroom also tended to be academically challenged during EOtC. Pupils are able to have more help from teachers inside of the classroom compared to EOtC. Several academically high-achieving pupils expressed their enjoyment of the autonomy that was demanded during EOtC. Low achieving pupils, particularly boys, were more likely to direct their attention away from the teacher towards surrounding objects. Low-achieving boys tended to display lower academic participation during EOtC compared to low-achieving girls. Interviews indicated that competitions and relay races hinder concentration in both low- and high-achieving pupils because concentration disappears (e.g. when math is added to relay races or heckling from peers). | EOtC allowed pupils to use other skills outside of classroom skills (athletic abilities such as running, being able to light a campfire during food science lessons etc.). EOtC promotes social wellbeing for lower-achieving pupils academically by offering opportunities to experience feelings of success within an academic context. Pupils perceived EOtC to be a more fun and more varied teaching method than classroom teaching. Group work and joint activities during EOtC can help rearrange existing social patterns in class for both academically low-achieving and high-achieving pupils. It is also necessary for the teacher to provide a structure for group work in which there is an opportunity to rearrange social relationships among classmates for group work in EOtC to have a positive influence on pupils’ social wellbeing. | |
| Mygind, 2007 | The mean activity level was more than twice as high during outdoor learning days as compared to traditional school days. | ||
O’Brien & Murray, 2007 | Forest schools contribute to an individual’s knowledge and understanding of the environment. Children became familiar with the woodland and were encouraged to use their senses to observe flora and fauna. Children became eager to discover things for themselves and were motivated to learn. | Forest schools increase self-esteem and confidence of those individuals who take part, improves an individual's ability to work cooperatively and increases awareness of others, increases motivation and concentration, and contributes to the development of language and communication skills. Children gained confidence in undertaking a range of activities (e.g. climbing trees or building dens). Parents also took more interest in FS due to their child’s enthusiasm and sometimes visited the outdoors more often because of this. The children who attended FS were noticed to transfer some of the skills they developed to different settings such as the home or school environment. | Forest schools improve physical motor skills by challenging children physically (due to rough terrain of the woodland floor and walking to and from the site). Also, because the children undertake activities for each session and also handle tools and equipment. FS helps develop gross fine motor skills as they have to move their entire body and may develop better range of movement and increased muscle strength. FS also develops fine motor skills as it involves small and intricate movements like tying knots, using tools, or using a stick to draw. |
Quibell & Charlton, 2017 | There is an improvement in the performance of the IG compared to the CG particularly with English reading. The mean scores in the IG were significantly different from the CG across all three core subjects (reading, writing, and math). The children in the IG improved at a faster rate than the CG particularly in English writing. WS may have a beneficial impact on children’s long-term learning. | ||
| Romar et al., 2019 | During the outdoor-learning days, the students engaged in less sedentary and Sed10 behaviour and in more light1, light2, moderate-intensity, and MVPA behaviour than on traditional school days. They also performed more steps on outdoor-learning days. The time spent on physical activity and sedentary variables increased from 15% to 58% between traditional and outdoor-learning school days. During traditional school days, the students typically sat at their desks. | ||
| Schneller et al., 2017 | EOtC participants spent more daily minutes in MVPA, VPA, and MPA than comparison participants. Boys in the EOtC group spent more minutes in MVPA, VPA, and MPA per day than boys in the comparison group. Girls of the EOtC group did not spend a significantly different amount of daily time in MVPA, VPA, MPA, or LPA compared to girls in the comparison group. There was a significant inverse association between one year increase in age and daily time spent in MVPA, VPA, MPA, and LPA (activity declined with age). | ||
| Sprague et al., 2020 | Each STEM-capacity domain (leadership, teamwork, science relevance, sustainability relevance, STEM self-efficacy, science interest, and overall STEM-capacity) significantly increased from pre-intervention to post-intervention. | There were improvements in all HRQoL domains (physical health functioning, school functioning, social functioning, family functioning, and overall HRQoL) after the intervention. Every NBE mentor and St. Louis public school teacher indicated that the NBE intervention was beneficial for the St. Louis public students due to enriched educational experiences, increased environmental awareness, novel experiences, and behavioral changes. | |
| Trapasso et al., 2018 | Majority of students preferred having lessons outdoors as opposed to indoors as it gave opportunities to play outdoors, whilst also being compliant to their school curriculum alongside nature-based learning activities. Children also reported that the behavior of their peers was different in FS sessions (peers were more helpful). The children did recognize FS as a place for learning, as well as a fun and playful experience. Children also enjoyed participating in FS sessions as it allowed free play and creativity leading to enjoyment and increased wellbeing. The children believed that FS improved their mood. | There was a significant difference between boys’ and girls’ average weekly MVPA levels. Boys averaged 75 min MVPA per weekday compared to the girls that averaged 55 minutes. Children had significantly greater levels LPA on a FS day compared to a regular non-PE school day and significantly more LPA on a PE day than a regular non-PE school day. Majority of the children expressed that FS sessions had increased the amount of physical activity they participated in at home. The main differences highlighted by children between their FS and their PE lessons were that FS sessions were perceived by the children to be more active and that PE lessons incorporated sports and sports equipment. Children felt that they had more freedom in FS compared to PE lessons | |
| Wistoft, 2013 | Children had an increased motivation and desire to learn through enjoyment and experience. The children were proud and ambitious about the level of learning achieved. Also, children learned sensory impressions (listening, smelling, seeing, tasting, touching). Children also developed insight into understanding ecology and natural relationships, growing vegetables, knowledge of different vegetables and herbs, understanding seed and the growth, new taste experiences and desire to taste, interest for and knowledge of healthy food, knowledge of ingredients and food preparation, ability to cook over a bonfire. | Children thought the ‘Garden for Bellies’ program was fun (experiencing nature with enthusiasm, preventing school fatigue, authentic experiences) Also, the children developed social competencies (better teamwork skills), strengthened feelings of belonging to a group - ability to establish and maintain a community), development of communication and interaction skills, learned to take responsibility, developed respect for others and their property). | |
Yildirim & Akamca, 2017 | The children’s cognitive skills, linguistic skills, socioeconomic skills, and psycho-motor skills significantly enhanced after the 10-week outdoor education (p < .001). | ||
| Zamani, 2016 | Different zones provided different cognitive play types. Play settings within the natural and mixed zones that support opportunities for hiding, imagining, creating games and playing in groups. Diversity of natural settings stimulates complex and exciting play opportunities for children with different learning styles. The natural zone offered higher levels of cognitive play, compared with the manufactured zone, and consequently provided fewer instances of functional play. The natural zone provided almost twice as much constructive play as the manufactured zone, along with the most exploratory and dramatic play compared with the other zones. | Children showed their preference for functional play, places to hide, climb or jump (inspiring their dramatic play), explorative play in nature, dramatic play and games in the natural zone, and loose elements for the diverse cognitive play stimulation. |
BPN = basic psychological needs, cfDNA = cell-free deoxyribonucleic acid, CG = control group, EOtC = education outside the classroom, FS = forest school, GMQ = gross motor quotient, HRQoL = health related quality of life, IG = intervention group, light1 = 101–1197 counts per minute, light2 = 1198–2295 counts per minute, LPA = light physical activity, MPA = moderate physical activity, MVPA = moderate-to-vigorous physical activity, NBE = nature-based environment, PA = physical activity, PE = physical education, PR = peak reactivity score, RAI = relative autonomy index score, Sed10 = sedentary periods lasting more than 10 min, SDI = self-determination index, SES = socioeconomic status, STEM = science, technology, engineering and mathematics, VPA = vigorous physical activity, WS = wild schooling
