Forest school participation and ecological identity formation — a longitudinal case study

This longitudinal case study follows 24 children through six years of year-round forest-school programming, from kindergarten through grade 5, with the goal of examining how sustained immersion in a single ecological context shapes what the developmental literature calls ecological identity. Participants were drawn from a single cohort at a private forest school in central Ontario, with retention of 22 of 24 across the study period.

Mixed-methods data collection included annual semi-structured interviews, structured observation sessions, ecological literacy assessments adapted from existing instruments, and academic-outcome tracking against age-matched controls at conventional schools. Qualitative analysis used a grounded-theory approach informed by Chawla's framework for ecological identity development.

Findings: by grade 5, all retained participants demonstrated ecological identity profiles consistent with what the framework describes as "place-identified" rather than "place-aware" — a substantially deeper relationship to specific ecological contexts than the literature typically reports. Academic outcomes on provincial benchmarking were comparable to or slightly exceeded matched controls, despite the cohort spending no more than 30% of its instructional time on conventional indoor academic work.

Context. Forest schools — defined here as learning environments in which children spend the majority of their instructional time in a consistent outdoor ecological context — have grown rapidly in Canada over the past decade. The pedagogical claims made for forest schools include effects on cognitive, social-emotional, and academic outcomes, but the longitudinal evidence base remains thin. Most studies are cross-sectional, retrospective, or limited to single-year cohorts.

This case study was undertaken to address that gap. By following a single cohort across six years of programming, we hoped to surface developmental trajectories that shorter-window studies miss.

Participants and setting. The cohort comprised 24 children at a private forest school in central Ontario, all of whom enrolled in kindergarten and were tracked through grade 5. The program operates year-round, with children spending 70% of their instructional time outdoors in a 200-acre woodland setting. Retention across the six-year window was 22 of 24.

Methods. Annual semi-structured interviews were conducted with each child, each year. Structured observation sessions (four per year per child) were coded by two independent raters. Ecological literacy was assessed using an instrument adapted from the Children's Environmental Perception Scale (CEPS). Academic outcomes were tracked against age-matched controls at three conventional public schools in the same region.

Findings: ecological identity. By grade 5, all 22 retained participants demonstrated identity profiles consistent with what Chawla and colleagues describe as place-identified. Children referred to specific landscape features as having moral and personal significance; described continuity between past and future seasons in the same setting; and reasoned about ecological systems with a depth of detail that suggests cumulative, situated knowledge rather than instructed knowledge.

Findings: academic outcomes. Despite spending substantially less time on conventional indoor academic work, cohort academic outcomes on provincial benchmarking were comparable to or slightly exceeded matched controls. Reading outcomes were marginally higher; numeracy was comparable; science scores were substantially higher.

Discussion. The findings are consistent with two non-exclusive interpretations. First, sustained immersion in a single ecological context produces a developmental trajectory in ecological identity that is qualitatively different from what typically occurs in episodic or supplementary outdoor programs. Second, the time spent outdoors is not subtractive from academic learning — the cohort's outcomes suggest that whatever instructional efficiency is lost in covering less ground indoors is recovered through the depth of situated engagement.

Limitations and future work. The single-site case-study design limits generalisability. Comparison cohorts were age-matched but not randomly assigned. A multi-site replication with random assignment is the obvious next step, and is being scoped for a 2026 launch.