What makes an inductive argument strong, and how does inductive support differ from the certainty of deduction?
distinguish inductive from deductive reasoning and evaluate inductive arguments for strength and cogency rather than validity
A focused QCE Unit 3 answer on inductive reasoning. Covers the difference between deduction and induction, why inductive arguments are assessed for strength and cogency rather than validity, the role of probability, and how added evidence can defeat an otherwise strong inference.
Reviewed by: AI editorial process; not yet individually human-reviewed
Have a quick question? Jump to the Q&A page
Jump to a section
What this dot point is asking
QCAA wants you to handle the second great family of reasoning. Deductive arguments aim at validity; inductive arguments aim only at making a conclusion probable. You need to define induction, contrast it with deduction, and assess inductive arguments using the right vocabulary: strength and cogency, not validity and soundness. This underpins the whole scientific reasoning strand of Unit 3.
The answer
Deduction versus induction
A deductive argument claims that its conclusion follows necessarily: if the premises are true, the conclusion cannot be false. An inductive argument claims only that its premises make the conclusion likely. The conclusion of an inductive argument always goes beyond the information in the premises, which is why it can be wrong even when every premise is true.
Compare:
- Deductive: All swans in this lake are white; that bird is a swan in this lake; therefore it is white.
- Inductive: Every swan observed so far has been white; therefore all swans are white.
The inductive version was historically refuted when black swans were found in Western Australia. No deductively valid argument can be overturned by new evidence in that way, because validity is not a matter of degree.
Strength, not validity
Because induction is a matter of degree, we do not call inductive arguments valid or invalid. We call them strong or weak. An inductive argument is strong when, if the premises are true, the conclusion is probably true; it is weak when the premises give the conclusion little support. Strength comes in degrees: a sample of ten thousand supports a generalisation more strongly than a sample of three.
Cogency, the inductive cousin of soundness
Soundness combines validity with true premises. The inductive parallel is cogency. An inductive argument is cogent when it is strong and its premises are actually true. A cogent argument gives you a genuinely well-supported conclusion. A strong argument with a false premise is uncogent, just as a valid argument with a false premise is unsound.
So the parallel structure is:
- deduction: valid plus true premises equals sound;
- induction: strong plus true premises equals cogent.
Defeasibility
A central feature of induction is that it is defeasible: adding new true premises can weaken or destroy a previously strong argument. "Most birds fly; Tweety is a bird; so Tweety probably flies" is strong, until we add "Tweety is a penguin." Deductive validity is monotonic by contrast: adding premises can never make a valid argument invalid. Recognising defeasibility is essential when you evaluate scientific and everyday reasoning, because real evidence keeps arriving.
Common inductive forms
The inductive strand of Unit 3 includes several recurring patterns you will study in their own right:
- Inductive generalisation: from a sample to a population.
- Argument from analogy: from similarities between cases to a further shared feature.
- Causal reasoning: inferring causes, often via Mill's methods.
- Inference to the best explanation (abduction): inferring the hypothesis that best accounts for the evidence.
- Statistical syllogism: from "most F are G" and "x is F" to "x is probably G."
Each is assessed for strength: how probable does the conclusion become, given the premises?
Why this matters in philosophy
Almost all empirical knowledge rests on induction: every prediction that the future will resemble the past is inductive. David Hume argued that this reliance cannot be justified without circularity, which is the problem of induction. Karl Popper responded by trying to base science on deductive falsification instead. You cannot follow either debate without first being clear that induction trades certainty for probable, defeasible support.
Try this
Q1. Distinguish a strong inductive argument from a valid deductive argument. [3 marks]
- Cue. Validity makes a false conclusion impossible given true premises; strength makes a false conclusion merely improbable.
Q2. Explain what cogency adds to strength. [2 marks]
- Cue. Cogency requires the argument to be strong and to have actually true premises.
Q3. Give an example showing that an inductive argument is defeasible. [3 marks]
- Cue. "Tweety is a bird, so Tweety probably flies" is weakened by adding "Tweety is a penguin."
Related dot points
- distinguish validity from soundness, and evaluate deductive arguments for both, using premises and conclusions
A focused QCE Unit 3 answer on validity and soundness. Covers the structure of deductive arguments, the difference between truth and validity, what soundness adds, common valid forms such as modus ponens and modus tollens, and how to test arguments by counterexample.
- analyse and evaluate arguments from analogy, assessing the relevance and number of similarities and the presence of relevant disanalogies
A focused QCE Unit 3 answer on analogical reasoning. Covers the structure of an argument from analogy, the criteria that make one strong (relevance, number and variety of similarities), how relevant disanalogies weaken it, and famous philosophical analogies such as Paley's watch and Thomson's violinist.
- evaluate inductive generalisations by assessing sample size, representativeness and the dangers of hasty generalisation and biased sampling
A focused QCE Unit 3 answer on inductive generalisation. Covers the structure of generalising from a sample to a population, the criteria of sufficient size and representativeness, the fallacies of hasty generalisation and biased sampling, and why anecdotes and self-selected samples mislead.