Planning, Analysis and Evaluation
Planning: state the independent, dependent and standardised (controlled) variables, how each is changed, measured or kept constant, and the range and intervals of the IV. Include a control (e.g. boiled enzyme) and repeats (at least 3). Assess risks. Statistical tests (with a null hypothesis of no difference/association): t-test compares two means; χ² compares observed with expected frequencies; Spearman's rank or Pearson's test correlation. If the calculated value is greater than the critical value at p = 0.05, reject the null hypothesis. Evaluation: validity (measures what it should, all other variables controlled), reliability/repeatability (consistent repeats), accuracy, precision; identify anomalies; suggest specific improvements (e.g. use a thermostatically controlled water bath).
Learning objectives
- Plan investigations with variables, controls and repeats.
- Choose the right statistical test.
- Interpret test results using critical values.
- Evaluate validity and suggest improvements.
You want to know if leaf length differs between shade and sun plants. Which test?
Core idea
Two plant groups differ in mean height. Statistics tells you whether chance alone could explain it.
Plan to change one variable and control the rest, then test results against a null hypothesis.
Match the test to the question.
Means 12.0 and 10.0; variances 4.0 and 5.0; n = 10 each.
- Goal
- t.
- Strategy
- Difference ÷ SE of difference.
- Standard error of difference.
- Compare with 2.10 (df 18).
Summary
- Variables and controls.
- Choose the right test.
- Validity and improvements.
- Null hypothesis
- Statement that there is no significant difference or correlation.
- Validity
- Whether an experiment measures what it intends to.
Create a free account to read the rest of this lesson: key formulae, worked examples, explore the graph, common mistakes, exam tip, guided practice and more. Your answers are checked and your progress is saved.