AD / CLIMATE PHYSICS AI← All research articles
Climate education · Proposed intervention and evaluation

Can organised sport produce measurable climate action?

A testable intervention linking Caribbean youth sport, climate literacy and independently verified environmental activities.

Adrian Dunkley · Climate Studies Group Mona, UWI Mona11 October 2026 · Design and evaluation protocol; no completed youth pilot or measured outcomes

Abstract

Climate education often measures what young people remember, although communities also need to know whether learning translates into safe, useful action. This article describes a proposed Caribbean intervention that embeds climate concepts in organised youth sport, accompanied by supervised local activities, independent verification of participation and transparent reporting of sponsorship funds. The model uses familiar team structures and regular practice sessions as opportunities for repeated exposure to weather risk, environmental stewardship and collective decision-making. It distinguishes three endpoints: changes in climate knowledge, verified participation in appropriately supervised activities and persistence of learning after the programme ends. A proposed cluster-randomised evaluation compares the intervention with ordinary sport programming across participating clubs or schools. It would estimate changes using pre-programme and delayed assessments, observation of activities and audited administrative records, while separately examining attendance, access, disability inclusion, safeguarding and unintended pressure to perform. Gamification is a design choice, not a demonstrated causal effect in this setting. No completed pilot, literacy improvement or environmental outcome is claimed. The scientific contribution is a research protocol specifying what evidence would distinguish temporary engagement from retained knowledge and independently verified community action.

Keywords · climate literacy · gamification · Caribbean · sport · youth safeguarding · evaluation

1. Introduction

Climate education faces two separate questions. Can a learner explain a climate-related hazard accurately? And can that learning contribute to a safe, beneficial action in the learner’s community? An awareness survey does not answer the second question. Neither does a tally of people attending a tree-planting event establish that a climate education programme caused a lasting change in behaviour.

Organised sport provides a setting in which young people often already cooperate, practice repeatedly and follow shared rules. In Jamaica and elsewhere in the Caribbean, the proposed intervention could be integrated into cricket, netball, football and other established activities. A climate exercise need not replace the sport session. A brief team challenge might ask players to distinguish the meaning of a weather forecast from a climate projection, identify heat-risk warning signs or map a recognised flood exposure near their school.

The programme described here remains a proposed intervention. There are no collected trial outcomes, claims of improved climate literacy or verified environmental gains from this particular programme. The immediate objective is to define an implementation that can be evaluated without confusing engagement, knowledge, physical activity and environmental impact.

2. Evidence and conceptual basis

Gamification is the use of game-design features in activities that are not games. Features may include points, progress indicators, teams, challenges and feedback. A meta-analysis by Sailer and Homner examined the effects of gamification on cognitive, motivational and behavioural learning outcomes across educational settings [1]. Its findings motivate further testing but do not establish that a particular Caribbean youth-sport programme will change climate behaviour. A lesson made enjoyable may increase participation without improving factual accuracy or long-term retention.

Climate risk is also not distributed uniformly across young people. UNICEF’s global assessment of children’s exposure and vulnerability to climate-related hazards documents the importance of child-sensitive climate risk [2]. A Caribbean implementation would need to account for differences in access to safe playing spaces, reliable internet, transport, disability-inclusive activities and parental support. Education about a hazard must not expose children to that hazard.

The proposed logic is therefore modest. Repeated, structured encounters with local climate problems may improve learning opportunities; cooperation within teams may support completion of supervised tasks; feedback may encourage return participation. Whether those mechanisms produce a difference beyond ordinary sport and standard climate lessons remains an empirical question.

3. Intervention design

3.1 A three-part learning cycle

First, participants encounter a short climate concept attached to a normal sport activity. For example, a cricket session could introduce rainfall intensity and the difference between a rate in millimetres per hour and a total accumulation in millimetres. A netball practice could explain the interpretation of a heat index and why risk depends on exertion, hydration, access to shade and local conditions. Coaches would use reviewed lesson cards rather than improvised medical or weather advice.

Second, teams examine a local problem. Examples include creating a school flood-risk sketch from existing maps, conducting a supervised waste audit, documenting a maintenance need at an established tree-care site or checking whether a designated cooling space is accessible. Children should not clear drains, enter floodwater, handle hazardous waste or inspect unstable structures. Third, teams submit evidence of an activity through an approved adult organiser. Evidence could be a signed checklist, a short activity record and a non-identifying image of an approved project, where consent permits.

Credits would recognise accurate explanation, cooperation and completion of verified tasks, rather than rewarding the most resources collected or the largest volume of photographs. The system should permit equivalent non-digital participation. It must not penalise a child for their economic circumstances, disability, lack of a device or decision not to appear in a photograph.

3.2 Verified environmental actions

Verification requires a named supervisor, an agreed task definition, a date and an observable completion criterion. In the case of tree care, the programme might record that a designated existing planting site was watered or checked according to an approved maintenance plan. A count of trees planted should not be equated with trees that survive. Survival would require a later independent assessment using a defined observation period. Similarly, completing a flood-risk map is a learning product, not evidence that flood exposure has been reduced.

3.3 Sponsorship and financial integrity

Funds raised through participation must remain separate from scores awarded for learning. A public financial ledger would identify donations received, amounts disbursed, permitted expenses and independent reconciliation. Participation must not be contingent on fundraising. The programme requires a written safeguarding policy, consent processes for minors, restricted access to personal data and an adult escalation route for complaints or unsafe activities.

4. A physical example for a sport lesson

Consider an introductory exercise that asks a team to predict the approximate flight of a ball thrown at a fixed angle. It is not itself a climate experiment. It introduces the difference between a physically defined relationship and a statistical association. For idealised motion without air resistance, the horizontal and vertical positions after time t follow the familiar constant-gravity approximation:

x(t) = x₀ + v₀ cos(θ)t ;   z(t) = z₀ + v₀ sin(θ)t − ½gt² (1)x and z are positions in metres; v₀ is launch speed in m/s; θ is angle; g is gravitational acceleration in m/s². This simplified model neglects aerodynamic drag and is not a model of storm physics.

Participants can then discuss why models depend on assumptions. Air resistance and wind alter a ball's trajectory. Atmospheric temperature and rainfall affect playing conditions through entirely different mechanisms. The lesson does not claim that altering a rainfall slider in an illustration physically changes a ball trajectory unless such effects are explicitly modelled and validated.

A second numerical exercise can explain accumulation: a constant rain rate of r millimetres per hour over an interval of Δt hours produces an idealised depth R = rΔt millimetres. A short burst of rain and a day-long average should not be treated interchangeably. This is an opportunity for careful unit handling, not a substitute for local hazard guidance.

5. Interactive teaching illustration

Figure 1. Projectile motion and separate weather-context controls
Synthetic ball motion above a sport field.DRAG TO ORBIT · EDUCATIONAL / SYNTHETIC
The ball follows an illustrative projectile arc. Weather inputs control the display of synthetic rain and a numerical context label; they are not real-time observations, physiological measurements, changes to the projectile equation or operational sport-safety thresholds. This animation supports discussion of modelling assumptions. Synthetic learning scene.

6. Proposed empirical evaluation

6.1 Study population and allocation

The preferred first-stage study assigns entire participating clubs or schools to the programme or an appropriate comparison condition. Cluster allocation reduces contamination between young people who train together. An ordinary-sport comparator is preferable to a no-activity group because it helps separate the educational addition from the benefits of participation in sport. If random assignment is not feasible, a matched comparison with pre-specified confounding variables should be used, and the more limited causal interpretation made explicit.

Eligibility criteria, the assent and guardian-consent process, age range and restrictions on the collection of identifying data must be reviewed before enrolment. Recruitment should not favour schools with high connectivity or well-funded facilities. Where possible, a design team with educators, child-protection specialists, coaches and young people should test the lesson materials for comprehension and safe implementation.

6.2 Outcomes

The primary educational outcome would be a pre-specified, age-appropriate climate-knowledge assessment scored against a validated answer key. Secondary outcomes might include delayed retention after the activity period, completion of independently verified tasks and changes in self-reported confidence. Self-reported intentions must be reported separately from observed acts. Actual environmental outcomes, such as survival at a maintained planting site, would require a later field assessment with baseline measurements.

A simple analysis of change could begin with:

Δ = (ȲI,after − ȲI,before) − (ȲC,after − ȲC,before) (2)Δ is a difference-in-differences contrast of mean outcome scores; I identifies intervention clusters and C comparison clusters. Its causal interpretation requires suitable allocation or defensible parallel-trends assumptions.

For an actual cluster trial, the final model should account for intra-club correlation, baseline ability, differential attendance and missing follow-up observations. A sample-size calculation must be based on a declared minimum educational effect, plausible variation and expected attrition. Counts of submitted tasks require audit sampling to identify duplicate, trivial or non-compliant activities. Safety incidents and complaints are outcomes to measure, not administrative notes to omit.

7. Threats to validity and safeguards

There are several credible reasons for a trial to fail. Teams may become better at answering the exact test questions without improving conceptual understanding. Clubs with more adult volunteers may complete more verified actions regardless of the learning design. Learners may withdraw if the scoring system feels punitive, or avoid participation if digital devices are required. A short-term increase in enthusiasm may disappear once the competition ends.

The analysis should therefore report implementation fidelity, missing data, results by access category and any displacement of normal coaching time. The control programme must not be described as lacking climate awareness merely because it does not use the proposed platform. Researchers also need to distinguish behavioural proxies from environmental consequences: an audited community activity is not equivalent to measured local risk reduction.

No children should perform dangerous environmental work to receive credit, and no child's personal location, photograph or identifiable performance history should be made publicly searchable. Clubs must retain a non-digital path to full participation. Any public reporting must use suitable aggregate categories and suppress small groups where re-identification is possible.

8. Conclusion

Sport may be a practical setting for climate learning because it provides repeated contact, structured cooperation and established community roles. The proposed contribution is not a claim that gamification works in Caribbean climate education; it is an experimental design that can test whether the addition produces retained knowledge and independently verified, safe action. Credible evidence requires comparisons, child protection, financial transparency and follow-up that continues after the novelty of competition has faded.

References

  1. Sailer, M. & Homner, L. (2020). The gamification of learning: a meta-analysis. Educational Psychology Review, 32, 77–112. https://doi.org/10.1007/s10648-019-09498-w.
  2. UNICEF (2021). The Climate Crisis Is a Child Rights Crisis: Introducing the Children's Climate Risk Index. https://www.unicef.org/reports/climate-crisis-child-rights-crisis.
  3. UNESCO (2021). Recommendation on the Ethics of Artificial Intelligence. https://www.unesco.org/en/artificial-intelligence/recommendation-ethics.