Find a project worth starting.

• Sustainability
• Neuroscience & learning
• Responsible technology
• Literature
• Social impact
• Chemistry
• Economics
• History
• Mathematics
• Psychology & education
• Biology
• Language
• Engineering

13 project ideas shown

• SustainabilityBeginner

  Less waste. Better systems.

  Which published food-waste reduction approaches could inform a feasible school proposal?

  • Literature review
  • 3–4 weeks

  First stepFind three published food-waste interventions and note where each was tested.

  See the plan

  Approach

  Compare published interventions, their context, costs and evidence quality. Use public information.

  Possible output

  An evidence-informed proposal with an evaluation plan.

  Watch out for

  Evidence from another setting may not transfer; implementation needs school permission.

• Neuroscience & learningIntermediate

  Sleep, attention & what we know

  How consistent is published evidence linking sleep and attention?

  • Literature review
  • 4–6 weeks

  First stepSearch Google Scholar for “sleep” AND “attention” AND “adolescents” and skim five abstracts.

  See the plan

  Approach

  Compare studies, their age groups, measures and possible confounders. Do not collect classmates’ health information.

  Possible output

  A critical literature review and an evidence map.

  Watch out for

  Observational associations cannot establish that sleep alone causes a change in attention.

• Responsible technologyStretch

  Whose data? Whose decisions?

  How do published evaluations reveal different patterns of AI error across groups?

  • Data analysis
  • 5–8 weeks

  First stepPick one published AI evaluation and list how it defines each group it compares.

  See the plan

  Approach

  Examine an appropriate public dataset or published evaluations; document definitions and sampling.

  Possible output

  A transparent comparison with a discussion of fairness.

  Watch out for

  A benchmark captures only part of a system’s real-world context.

• LiteratureIntermediate

  Reading fairness between the lines

  How do selected texts frame fairness and identity through narrative voice?

  • Close reading
  • 4–6 weeks

  First stepChoose two texts and mark three passages where the narrator judges a character.

  See the plan

  Approach

  Choose a small, justified set of texts and compare passages, context and alternative interpretations.

  Possible output

  A comparative essay with close readings.

  Watch out for

  Text selection shapes the argument; acknowledge other plausible interpretations.

• Social impactBeginner

  Access begins with evidence

  What barriers to a community resource are visible in public information?

  • Document analysis
  • 3–5 weeks

  First stepPick one local service and list every requirement a new user must meet to access it.

  See the plan

  Approach

  Review public reports and service information; compare proposed responses and their assumptions.

  Possible output

  An evidence-informed access plan with practical next steps.

  Watch out for

  Public information cannot stand in for lived experience; further engagement needs faculty guidance.

• ChemistryBeginner

  What’s really in the tap?

  How do water hardness and mineral levels vary across published water-quality reports for different districts?

  • Public data
  • 3–4 weeks

  First stepFind your local water authority’s latest quality report and list which minerals it measures.

  See the plan

  Approach

  Collect official water-quality reports, record the same measures for each area and compare them against published guideline values.

  Possible output

  A clear comparison chart with an explanation of what the numbers mean.

  Watch out for

  Reports may use different sampling dates and methods; compare like with like.

• EconomicsIntermediate

  The price of a lunch

  How have the prices of common lunch foods changed compared with overall inflation over the last ten years?

  • Data analysis
  • 4–6 weeks

  First stepDownload one food-category price index from your national statistics office.

  See the plan

  Approach

  Use official consumer price index data for food categories and compare their trends with headline inflation.

  Possible output

  A short data report with two or three well-labelled charts.

  Watch out for

  National averages may hide local differences; price indices are not the same as individual prices.

• HistoryIntermediate

  Archives in your pocket

  How did newspapers report a past public-health campaign, and how did the message change over time?

  • Archival research
  • 5–7 weeks

  First stepSearch one digitised archive for a campaign name and save ten articles with dates.

  See the plan

  Approach

  Search a free digitised newspaper archive, sample articles across several years and compare the language and claims used.

  Possible output

  A source-based essay with quoted evidence and a timeline.

  Watch out for

  Surviving and digitised papers are a partial record; note whose voices are missing.

• MathematicsStretch

  Fair queues

  Which queue arrangement would minimise the average waiting time at a school canteen?

  • Modelling
  • 4–6 weeks

  First stepSketch the canteen layout and write down three assumptions your model will make.

  See the plan

  Approach

  Build a simple queue model in a spreadsheet using made-up or anonymous counted arrival times, and compare one queue with several.

  Possible output

  A model, its assumptions and a recommendation with its limits.

  Watch out for

  Models simplify real behaviour; test how sensitive the answer is to your assumptions.

• Psychology & educationBeginner

  Does testing help?

  What does published research say about retrieval practice (self-testing) for secondary-school students?

  • Literature review
  • 3–4 weeks

  First stepFind two studies on retrieval practice and note the age of the participants in each.

  See the plan

  Approach

  Compare studies by age group, subject and how learning was measured. Any classroom trial needs teacher approval.

  Possible output

  An evidence summary with practical tips for students.

  Watch out for

  Many studies use university students; check whether results apply to your age group.

• BiologyBeginner

  Biodiversity on campus

  How does the variety of plants and insects differ between green and paved areas of the school grounds?

  • Field observation
  • 3–5 weeks

  First stepAsk a teacher which areas you may survey, then mark two plots of the same size.

  See the plan

  Approach

  Mark equal-sized plots, count species at the same time of day on several days and compare the totals.

  Possible output

  A field report with maps, tallies and a simple diversity comparison.

  Watch out for

  A few weeks of counts only capture one season; weather can change what you see.

• LanguageBeginner

  How we write, then and now

  How has the use of a common word or phrase changed in published books over the last century?

  • Text data
  • 2–4 weeks

  First stepCompare two related words in the Ngram Viewer and screenshot the chart.

  See the plan

  Approach

  Use the free Google Books Ngram Viewer, compare several related words and explain the patterns with historical context.

  Possible output

  A short illustrated essay with annotated trend charts.

  Watch out for

  The book corpus over-represents some kinds of writing; trends in books are not everyday speech.

• EngineeringIntermediate

  Shade that works

  Which low-cost shade designs best reduce the surface temperature of a sunny wall or bench?

  • Small experiment
  • 3–5 weeks

  First stepChoose a sunny surface and record its temperature at the same time for three days.

  See the plan

  Approach

  Build two or three simple shades, measure surface temperature with a thermometer at set times and repeat on several days.

  Possible output

  A design report with test data, photos and a recommendation.

  Watch out for

  Small, short tests vary with weather; repeat measurements and report the range.