Paper 1B study guide
Biology experimental skills checklist
Use this to interrogate an unfamiliar investigation: what was sampled, what was measured, how variable the system is, and what the evidence actually supports.
Microscopy & image analysis
Use scale bars, magnification and repeated fields of view; make clear what dimension or area was measured.
Sampling organisms
Use random or justified systematic sampling, quadrats, transects or capture–mark–recapture; consider representativeness and independence.
Replicates & variation
Distinguish technical/repeated measurements from biological replicates; summarize spread with an explicitly named statistic.
Statistical testing
State a null hypothesis, choose a test appropriate to the design and data, and interpret p-values without claiming causation.
Graphs & models
Choose suitable graph types, label axes and units, use best-fit relationships, r and R², and distinguish interpolation from extrapolation.
Controlled experiments
Identify independent, dependent and controlled variables and evaluate whether controls isolate the proposed causal factor.
Databases & secondary data
Check provenance, definitions, sample sizes and uncertainty before treating secondary data as directly comparable.
Ethics & biological systems
Consider welfare, consent where relevant, environmental impact and whether the measurement itself could alter the system.
Data-quality language
| Concept | Useful interpretation |
|---|---|
| Random variation | Creates scatter among observations; replication helps reveal the distribution and improves estimation of a mean. |
| Systematic bias | Shifts observations in a common direction; more repeats do not remove it. |
| Precision | Closeness of repeated measurements and/or the resolution with which a quantity is recorded. |
| Reliability | Consistency under repetition or replication. |
| Validity | Whether the design and measurement genuinely address the biological question. |
| Representativeness | Whether a sample reasonably reflects the population or condition to which conclusions are generalized. |
Graph checklist
- Is the graph type appropriate for continuous, categorical or frequency data?
- Are axes labelled with quantities and units?
- Are error bars identified as SD, SE or something else?
- Does the relationship look linear, curved, thresholded or saturated?
- Are r and R² being interpreted as association/model fit rather than proof of mechanism?
- Is the requested value interpolated within the observed range or extrapolated beyond it?
Evaluation checklist
A strong improvement names the limitation, explains how it could affect the measured or processed result, and proposes a change that directly addresses that mechanism. “Human error” is usually too vague to be useful.