Skills, Data Analysis and Fieldwork
Working scientifically
Candidates must design investigations, identify variables (independent, dependent, controlled), write testable hypotheses, assess reliability and validity, and distinguish accuracy from precision. Know how to select equipment, sample representatively and reduce bias.
Data and graphs
- Read axes, units and scale before interpreting trends.
- Describe patterns (direction, rate, anomalies) before explaining them.
- Distinguish correlation from causation; identify confounding variables.
- Calculate means, percentages, rates and simple ratios; use appropriate significant figures and units.
- Interpret uncertainty, error bars and anomalous points.
Fieldwork
Use transects and quadrats for vegetation, stratified or random sampling for representativeness, and appropriate safety and ethical protocols. Record site, date, time, conditions and method so work is repeatable. Consider spatial and temporal scale, and limitations such as access, weather and sampling effort.
Common traps
- Describing a trend without quoting data values.
- Explaining before describing, or inventing a mechanism unsupported by the data.
- Confusing accuracy (closeness to true value) with precision (reproducibility).
- Using a non-representative sample, such as quadrats placed only where vegetation is obvious.
- Omitting units, or mixing units in a calculation.
- Claiming causation from a correlation.
Worked example: rate calculation
A quadrat survey records 12, 15, 9, 14 and 10 plants in five 1 m2 quadrats. Mean = 60/5 = 12 plants m-2. Rule of thumb - calculate the mean, then scale to the area, e.g. a 500 m2 site gives about 6 000 plants. State this as an estimate, noting sampling variability, and never report more significant figures than the data justify.