Discover how inductive arguments work with probability, why new evidence can always change the picture, and how to judge the strength of an inductive case.
In Lesson 2 you learned that deductive arguments aim to guarantee their conclusions. Inductive arguments work differently. They do not claim to guarantee anything. They claim to make a conclusion probable given the evidence.
This is not a weakness. It is simply how most real-world reasoning works. Science, medicine, business, and everyday prediction all rely on inductive reasoning.
Because inductive arguments cannot be valid or invalid in the deductive sense, we evaluate them differently. The key question is: how much do the premises raise the probability of the conclusion?
One of the most important properties of inductive arguments is that they are defeasible: new information can defeat or weaken an argument that previously seemed strong.
Consider this example. You observe that every swan you have ever seen is white. You inductively conclude: all swans are white. This was a strong argument given your evidence. Then you travel to Australia and encounter a black swan. The argument is defeated by new evidence.
This is not a failure of inductive reasoning. It is how it is supposed to work. Good inductive reasoners stay open to revision.
Click each scenario to see the inductive structure and a strength assessment.
"It has rained every afternoon for the past two weeks during summer. It will probably rain this afternoon."
Premises: Consistent two-week pattern of afternoon rain during summer.
Strength: Moderate to strong. The pattern is consistent but limited to two weeks. Seasonal data and weather models would strengthen this further.
Defeasible: A high-pressure system moving in could defeat the conclusion.
"In a randomised controlled trial of 10,000 patients, Drug X reduced symptoms in 78% of cases. Drug X is likely effective."
Premises: Large, randomised trial with clear results.
Strength: Strong. Large sample, random assignment, and a clear effect size all contribute. This is near the top of the evidence hierarchy.
Defeasible: Conflicts of interest, publication bias, or a later meta-analysis could revise this.
"My friend's car has broken down three times this month. It is probably unreliable."
Premises: Three breakdowns in one month, one observer.
Strength: Weak. Small sample (one car, one month), no comparison baseline, unknown causes. The pattern is suggestive but not yet convincing.
Defeasible: If all three breakdowns had the same cause (a faulty batch part now replaced), the car may be fine going forward.
"Sales have grown 20% every quarter for three years. We expect continued growth next quarter."
Premises: Twelve consecutive quarters of 20% growth.
Strength: Moderate. Consistent long-term trend is a good sign. But business environments change, and past performance does not guarantee future results.
Defeasible: A new competitor, market saturation, or economic downturn could break the pattern.
Answer each question correctly to unlock the next one.
You will be shown a prediction scenario. Choose the piece of evidence that would make the inductive argument strongest. Score 4 or more out of 6 to pass.
Think of a prediction you make regularly: about a person, a team, a market, or a season. What evidence is it based on? How strong is that inductive argument? What new information could defeat it?
This is just for you. Nothing is saved or submitted.