Evaluating Inductive Arguments and Probabilistic and Statistical Fallacies
Inductive Arguments and Statistical Generalizations
Inductive arguments have conclusions that follow with high probability, but not certainty.
They are "defeasible," meaning additional premises can weaken them.
Statistical generalizations are based on empirical observations:
Universal generalizations apply to 100% of a class.
Partial generalizations apply to a percentage of a class.
Deductive arguments use universal generalizations; inductive arguments use partial generalizations.
Evaluating statistical generalizations involves assessing the truth of premises.
Conditions for a good statistical generalization:
Adequate sample size
Non-biased sample
A sample is a portion of a population. A population is the totality of members of some specified set of objects or events
A representative sample mirrors the population's characteristics.
Hasty Generalization
Hasty generalization is an informal fallacy that infers a statistical generalization from too few instances.
Sampling Bias
Sampling bias occurs when a sample is not representative of the population.
Bias can arise from:
How the sample is collected.
The questions asked.
Random Sampling
Random sampling aims to avoid bias by giving everyone in the population an equal chance of being selected.
Techniques must adapt to changing social and technological landscapes.
Inference to the Best Explanation
Inference to the best explanation is an inductive argument that concludes a hypothesis is true because it best explains observed facts.
Form:
Observed facts
Hypothesis explaining the facts
Comparison of competing explanations
Conclusion that the hypothesis is true
Explanatory Virtues
Criteria for a good explanation:
Explanatoriness: Explains all observed facts.
Depth: Doesn't raise more questions than it answers.
Power: Applies to similar contexts.
Falsifiability: Can be proven incorrect.
Modesty: Claims no more than necessary.
Simplicity: Posits fewer entities or processes (Ockham’s razor).
Conservativeness: Conflicts with fewer well-established beliefs.
Analogical Arguments
Arguments from analogy infer that if things x and y share similar properties, and y has characteristic A, then x probably has characteristic A.
Key conditions for strong arguments from analogy:
Relevant similarities between compared things.
Absence of relevant disanalogies.
Causal Reasoning
Causal reasoning identifies causes that produce specific effects.