Logical Fallacies

LogFall

A practical logical-fallacies reference with clear explanations, usable examples, and teaching tools.

Fallacy profile

Hasty generalization

Occurs when someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample.

EvidentialMathematical

Definition

Occurs when someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample.

Illustrative example

I interviewed four students about AI and two admitted cheating, so generative AI is ruining education.

Teaching gauges

These 0-100 gauges are teaching aids for comparing fallacies. They are editorial classroom estimates, not measured statistics.

Near-constant

85

Common in today's rhetoric

Shows up constantly in current politics, media, and online argument.

Moderate

65

Easy to spot

Recognizable, but easy to miss in a fast or heated exchange.

Almost automatic

85

Easy to innocently commit

Very easy for well-meaning people to commit without noticing.

Foundational

25

Difficulty

Usually approachable without much prior logic background.

Middle school+Formal logic

Reference

Family

Statistical/Sampling Fallacy

The reasoning misuses rates, probabilities, samples, distributions, or other quantitative expectations.

Aliases

fallacy of insufficient statistics, fallacy of insufficient sample, fallacy of the lonely fact, leaping to a conclusion, hasty induction, secundum quid

Quick check

What evidence is missing, selected, or overstretched here?

Why it misleads

A fuller explanation of how the fallacy works and why it can look persuasive.

The problem is not moving from sample to population; all induction does that. The problem is pretending a weak or unrepresentative sample carries more weight than it does.

That's like saying...

Instead of leading with the label, this analogy answers the shape of the reasoning move directly so the mistake is easier to see in plain language.

Fallacious claim

I interviewed four students about AI and two admitted cheating, so generative AI is ruining education.

That's like saying...

That's like tasting one spoonful from a burnt corner of the soup and condemning the entire pot. The leap from small sample to broad conclusion outruns what the evidence can support.

Caveat

This label is easy to overuse. The point here is not to call every weak argument by this name, but to reserve it for the exact misstep it describes.

Common misapplication

Do not use this label every time someone draws a general conclusion from limited evidence. Sometimes the sample really is enough for a modest claim; the problem is overreaching beyond what the sample can support.

Use the label only when...

Use this label only when someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample. If the real problem is that destruction or forced replacement is treated as an economic gain because the visible spending is counted while the unseen losses and forgone alternatives are ignored, the better label is Broken window fallacy.

Often confused with

These near neighbors are easy to mix up, so use the comparison to see the exact difference.

Comparison

Broken window fallacy

Why people mix them up: Both often look like evidential and mathematical mistakes at first glance.

Exact difference: Hasty generalization happens when someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample. Broken window fallacy happens when destruction or forced replacement is treated as an economic gain because the visible spending is counted while the unseen losses and forgone alternatives are ignored.

Quick split: What evidence is missing, selected, or overstretched here? Then compare it with What evidence is missing, selected, or overstretched here?

Comparison

Survivorship bias

Why people mix them up: Both often look like mathematical and evidential mistakes at first glance.

Exact difference: Hasty generalization happens when someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample. Survivorship bias happens when conclusions are drawn from the visible successes that made it through a filter while the failures, dropouts, or non-survivors are ignored.

Quick split: What evidence is missing, selected, or overstretched here? Then compare it with What numbers, rates, or probabilities are being ignored or mishandled?

Practice And Repair

Extra teaching tools that show why the fallacy is persuasive, what to look for, and how to correct it.

Why it matters

Why this mistake matters

Hasty generalization threatens rationality because someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample.

Main reasoning problem

Someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample.

Why this kind of mistake matters

It overstates, understates, cherry-picks, or misallocates the force of evidence.

Check yourself

The assessment area now uses mixed 10-question sets, so the fallacy is not announced in the title before the quiz begins.

What the assessment does

You will work through a mixed set of fallacy-identification questions. Focused links from a fallacy page will quietly include this fallacy among nearby look-alikes without announcing the answer in the page title.

Questions to ask

Use these category-based prompts to audit similar arguments.

Prompt 1

What evidence is missing, selected, or overstretched here?

Prompt 2

What numbers, rates, or probabilities are being ignored or mishandled?

Case studies

Each case study explains why the example fits the fallacy and links back to its source whenever source information is available.

Noncitizen voting, already illegal in federal elections, becomes a centerpiece of 2024 GOP messaging

AP's May 18, 2024 overview of noncitizen-voting rhetoric documented how a politically useful intuition about election fraud kept being treated as if it were established by the evidence. The report is especially useful for seeing how tiny counts, suggestive language, and moral urgency can be stretched into system-wide claims. The fallacy here is Hasty generalization: someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample. That matters here because the problem is not moving from sample to population; all induction does that. A better analysis would remember that the problem is pretending a weak or unrepresentative sample carries more weight than it does.

Associated Press · 2024-05-18

FACT FOCUS: Here's a look at some of the false claims made during Biden and Trump's first debate

AP's June 27, 2024 fact check of the first Biden-Trump debate is a dense collection of real argumentative shortcuts: statistics pulled loose from context, emotionally loaded immigration claims, and repeated assertions that did more rhetorical than evidential work. It is one of the best single-source stress tests in the library. The fallacy here is Hasty generalization: someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample. That matters here because the problem is not moving from sample to population; all induction does that. A better analysis would remember that the problem is pretending a weak or unrepresentative sample carries more weight than it does.

Associated Press · 2024-06-27

AP Explains: Migration is more complex than politics show

AP's migration explainer from September 20, 2024 is useful because it deliberately widens the frame beyond debate slogans and viral rumors. That makes it a strong case for fallacies that depend on flattening a complicated policy landscape into one cause, one image, or one moral punchline. The fallacy here is Hasty generalization: someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample. That matters here because the problem is not moving from sample to population; all induction does that. A better analysis would remember that the problem is pretending a weak or unrepresentative sample carries more weight than it does.

Associated Press · 2024-09-20

A single viral video of a fight, theft, or protest is often used online as if it reveals what a whole city, campus, or demographic group is like. The fallacy here is Hasty generalization: someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample. That matters here because the problem is not moving from sample to population; all induction does that. A better analysis would remember that the problem is pretending a weak or unrepresentative sample carries more weight than it does.

Election commentary frequently treats one poll, one county, or one diner focus group as proof of a national shift, even when the broader polling picture is mixed. The fallacy here is Hasty generalization: someone draws a broad conclusion from too little evidence, too small a sample, or a badly skewed sample. That matters here because the problem is not moving from sample to population; all induction does that. A better analysis would remember that the problem is pretending a weak or unrepresentative sample carries more weight than it does.

Related fallacies

Nearby entries chosen by shared categories and family resemblance.