Deductive and Inductive Reasoning Flashcards

An investigator reasons: 'Every witness who recants their statement in this jurisdiction has done so within 72 hours of giving it. This witness recanted after 96 hours. Therefore, this recantation likely has a different underlying cause than the typical witness.' Which flaw, if any, exists in this inductive reasoning?

Answer: The generalization may rest on an insufficiently examined sample or incomplete case records, making the 72-hour pattern unreliable.

Inductive generalizations depend on the representativeness and completeness of the sample. If the investigator's 'every witness' claim is based on a small or unrepresentative set of cases, the 72-hour pattern may not be a reliable baseline. The conclusion may be reasonable, but the hidden weakness is the unverified breadth of the underlying sample — a classic inductive limitation.

A detective constructs the following argument: 'If the suspect was at the scene, his fingerprints would be present. His fingerprints are present. Therefore, he was at the scene.' A senior officer correctly identifies this as logically invalid. Which formal fallacy does the detective's reasoning commit?

Answer: Affirming the consequent

The argument has the form: If P then Q; Q; therefore P. This is 'affirming the consequent,' a formal deductive fallacy. Fingerprints being present is consistent with many explanations beyond the suspect having been at the scene (e.g., transferred prints, planted evidence). A valid deductive form would be: If P then Q; not-Q; therefore not-P (modus tollens).

During a criminal analysis briefing, an analyst presents inductive evidence that a serial offender operates exclusively on weekends. The lead investigator argues this conclusion would be strengthened most by which of the following additions?

Answer: Showing that the pattern holds across a larger dataset, including incidents initially excluded due to incomplete reports.

Inductive arguments are strengthened by increasing the breadth and representativeness of the evidence base. Including previously excluded incidents (which may have been omitted due to incomplete data rather than irrelevance) directly addresses the sample completeness problem. Option D would convert it to a deductive argument — a different logical category. Option B introduces a separate inductive claim. Option A merely confirms one data point without strengthening the pattern.

An officer applies the following deductive syllogism to a use-of-force decision: 'All situations where a subject is armed and advancing on an officer justify defensive action. This subject is armed and advancing. Therefore, defensive action is justified.' A use-of-force review board later finds the reasoning procedurally flawed. The most likely basis for this finding is:

Answer: The major premise oversimplifies a legally and contextually nuanced standard, making it an unreliable basis for operational decisions.

The syllogism is formally valid — the conclusion does follow from the premises. The flaw lies in the major premise itself, which reduces a complex, multi-factor legal standard (proportionality, necessity, immediacy of threat, officer training) to an oversimplified categorical rule. Valid deductive arguments with false or oversimplified premises produce unreliable real-world conclusions. Review boards scrutinize whether the operative premise accurately reflects legal and policy standards.

Investigators note that in 80% of observed residential break-ins in a district, the point of entry was an unlocked rear window. They inductively conclude that unlocked rear windows are a primary vulnerability. A criminologist challenges the conclusion, arguing the investigators are committing a sampling error. The most precise description of this error is:

Answer: Base rate neglect, because the investigators failed to account for the overall proportion of homes with unlocked rear windows.

While survivorship bias and confirmation bias are real concerns, the most precise challenge the criminologist is raising is base rate neglect. If 95% of homes in the district have unlocked rear windows, a 80% window-entry rate may simply reflect the prevalence of that vulnerability rather than its disproportionate exploitation. Without knowing the base rate of the vulnerability in the population, the inductive conclusion about 'primary vulnerability' is statistically unsupported.

An analyst builds the following chain of reasoning during a fraud investigation: (1) Inductive: Based on 40 past cases, invoice fraud in this sector always involves a supervisory override. (2) Deductive: This transaction had no supervisory override. (3) Conclusion: This transaction is not invoice fraud. What is the most accurate logical assessment of this combined argument?

Answer: The deductive step is valid (modus tollens), but the conclusion's reliability depends entirely on the strength of the inductive generalization in step 1.

The deductive step applies modus tollens correctly: if invoice fraud → supervisory override; no override → not invoice fraud. This is formally valid. However, the conditional premise itself was derived inductively from 40 cases. The entire chain's reliability therefore hinges on whether that inductive generalization is strong enough to be treated as a reliable universal rule. This is the key epistemological vulnerability in mixed inductive-deductive reasoning chains used in investigative analysis.