Survey and observation
A survey is a planned look at the whole field. An observation is a careful note of what you see, hear, smell or measure.
- Plan: walk a zig-zag or X path so every part is seen. Pick 5 or more spots, not just the nicest one.
- Look at: plant height, leaf colour, gaps, flowering, pests, weeds, soil wetness, nearby trees or water.
- Record: write the date, spot, and what you saw. Take photos. Use numbers where you can (cm, count, %).
- Ask people: what was grown here before, what was added, how much rain fell.
Observation shows what is happening. It does not yet tell you why.
Diagnosing growth and environment
To diagnose is to find the cause of a problem from clues. Use a simple table of clues.
- Old (lower) leaves turn pale yellow: nitrogen low.
- Leaf edges purple or reddish: phosphorus low (often in cold soil).
- Leaf edges brown and dry: potassium low, or fertiliser burn or salt.
- Young leaves yellow with green veins: iron or similar minor nutrient low, often in alkaline soil.
- Droopy leaves, dry soil: water short. Droopy leaves, wet soil: roots lack air or rotted.
- Holes, curled leaves, sticky patches: insects. Spots with fuzzy growth: fungus.
Clues can look alike. So we measure too: soil pH, soil wetness, air and soil temperature, rainfall, day length. Then link clue + measurement + history. Only then pick a fix, and test it.
Experiment and verification
An experiment tests a guess (a hypothesis), for example "more water gives taller bean plants".
- Change one thing (the variable, here water). Keep soil, light, seed and pot the same. This is a fair test.
- Keep a control: one plot with no change, for comparison.
- Repeat: use 3 or more plants per treatment, so one odd plant does not fool you. Take the mean (average).
- Measure the same way each time (same day, same ruler).
- Record and graph: table first, then a bar or line graph.
- Verify: does the result support the guess? Repeat once more or in a second season. If it holds, you can trust it.
Remember: if two things changed together, you cannot tell which one caused the result.
Try it: yellow leaf detective
Find a plant with some yellow leaves in your garden or pot. Fill a table: which leaves (old or young), soil wet or dry, sunny or shady. Make a guess for the cause. Then test ONE fix on one plant only (for example, water a dry pot, or add a spoon of compost) and keep a similar plant as your control. Measure leaf colour by photo after 7 days. Did the fix work?
Key formulas and definitions
- Mean = sum of all measurements ÷ number of measurements
- Percent change = (new − old) ÷ old × 100
- Fair test: change ONE variable, keep all others the same
- Sampling: more spots and repeats = more trustworthy result
Worked examples
1. Three bean plants in the same pot group measure 18 cm, 20 cm and 22 cm. What is the mean height?
Sum = 18 + 20 + 22 = 60. Mean = 60 ÷ 3 = 20 cm.
2. A plant grew from 10 cm to 16 cm in a week. What is the percent increase?
Change = 16 − 10 = 6 cm. Percent = 6 ÷ 10 × 100 = 60%.
3. A student gives Pot A more water AND more light than Pot B and finds A taller. Can she say water caused it?
No. Two things changed together, so we cannot tell which one caused the difference. It is not a fair test. Change only one thing.
Common mistakes
- Looking at only the best or worst spot. Survey many spots across the field.
- Jumping to a fix from one clue. Yellow leaves can come from water, nitrogen, iron or disease.
- Changing two things in one experiment, so the cause is unclear.
- Skipping the control and the repeats, then trusting a single plant.