Inductive Reasoning and Scientific Method
What this domain tests
Generalisation, analogy, causal reasoning through Mill's methods, and the stages of scientific method. Answers must name the method used and state why the conclusion is only probable.
Forms of induction
- Enumeration: observed instances support a generalisation; strength depends on sample size and representativeness.
- Analogy: two things share known properties, so they probably share another.
- Causal inference: Mill's five methods - Agreement, Difference, Joint Method, Residues, Concomitant Variation.
Scientific method
Observation → statement of the problem → hypothesis → deduction of consequences → experiment and verification → acceptance, revision or rejection. A good hypothesis is testable, fruitful and simple, and a crucial experiment decides between rival hypotheses.
Worked example (Method of Difference)
A patient takes a drug and develops a rash. On a second occasion every circumstance is identical except that the drug is withheld, and no rash appears. The single varying factor, the drug, is the probable cause. The Method of Agreement would instead compare several cases that share only the rash-producing factor.
Traps that cost marks
Post hoc ergo propter hoc - treating sequence as causation; confusing Agreement (one common factor across cases) with Difference (one factor varied between two cases); generalising from a tiny or biased sample; presenting a confirmed prediction as proof rather than as support; forgetting that inductive conclusions are probable, never certain.