Availability of nature and access to nature in school: evidence from a Danish longitudinal school study

ABSTRACT

There has been an increasing focus on test scores and rankings in schools. However, the focus on children’s standardized test scores has come at the cost of their social relations. As a reaction to this test-centric focus, new approaches claim that nature is important not only for children’s social relations but also for their scholastic achievement. Drawing on a longitudinal dataset following 1,087 Danish schoolchildren in the Municipality of Aalborg, this paper sets out to investigate the impact of the availability of nature and access to nature on social relations and scholastic achievement. We address the endogeneity of child quality by applying a value-added approach and the endogeneity of school quality using propensity score matching. Our analyses suggest that the availability of nature and access to nature in themselves are not enough to promote social relations.

KEYWORDS:

• Nature
• scholastic achievement
• social relations
• value-added approach
• propensity score matching

Notes

1. According to Rosa (2013) bildung has to be addressed in a school context and bildung is not primarily a matter of the acquisition of knowledge or development of individual skills. Bildung is about creating resounding relationships with the world.

2. The concept of resonance means to set oneself in swinging movement by responding to something happening in the world. Similar to the way musical instruments reinforce their sound through resonance, Rosa maintains that people can experience a resonance in themselves when they experience the world (Rosa, 2013).

3. In Denmark, sports courts are placed mainly within school grounds.

4. The endogeneity problem implies that there is a correlation between the error term and nature, which undermines causal inferences (Wooldridge, 2006). One major problem in empirical research is the omitted variable problem, as one does not properly account for unobserved or uncontrolled variables. Another major problem is the selection problem, which can arise due to the endogeneity of choices.

5. The participation rate was 94.4% in 2008 and 91.4% in 2010.

6. Our definition of wild nature schools, cultivated nature schools and asphalt schools resemble, to some extent, the specification used by Mannion et al. (2015): beyond school ground, residential area and within school.

7. For 2008, GPA was obtained at the school level. The school level GPA was computed from the mean grades of primary school leavers (9th form) at each school in 2008, whereas the 2011GPA was at the individual level. This solution was the only one possible in order to obtain a GPA measure in 2008, as GPA is only available in the register data for Danish students at their final examination in 9^th grade. We use GPA obtained at school level to construct the value-added analyses (Quinto Romani & Beldo Klausen, 2017).

8. Using the value-added approach; we address the problem of omitted variable bias, but not the selection bias, implying that parents do not randomly select nature schools. In contrast, propensity score matching enables us to address the selection problem, but not the omitted variable problem, as we only match on observable characteristics (Wooldridge, 2006).

9. Christie, Higgins, and McLaughlin (2014) consider the impact residential outdoor learning interventions have on scholastic achievement using quantitative and qualitative methods. According to their results, there may be an interplay between soft and hard GPA. We reran the regressions using total GPA. The results shows that the availability of nature and access to nature have a beneficial but non-significant impact (results can be sent upon request).

10. Amount of nature is divided into ‘asphalt’, ‘cultivated nature’ and ‘wild nature’. However, no further subdivision is possible for distance to nature.

11. In terms of area, the 25th percentile is at or above 997.29 km², the 50th percentile is at or above 2414.84 km² and the 75th percentile is at or above 5064.75 km². Considering distance, the 25th percentile is below 38.31 km², the 50th percentile is below 91.16 km² and the 75th percentile is below 241.2 km².

Additional information

Funding

This study was financed by a grant for the research project “Structural Factors in Relation to Youth Behaviour, Lifestyle, and Health” from the Danish Council for Independent Research [grant number 12-127741].

Notes on contributors

A. Quinto Romani

Annette Quinto Romani is an associate professor at the Department of Sociology and Social Work at Aalborg University in Denmark. Her main research interest is in exploring inequality in health and education using econometric methods. The co-author Rolf Lyneborg Lund is an assistant professor at the Department of Sociology and Social Work at Aalborg University in Denmark. His main research interest is in geographical modeling, machine learning and AI.