Forbidden fallacies: Hasty generalization

At some point, a teacher or professor probably told you to avoid fallacies in your writing.

What are fallacies? Put simply, fallacies are arguments or conclusions based on faulty logic. Avoiding them will strengthen your writing and help gain readers’ trust. There are many fallacies, but today I’ll highlight one: hasty generalization.

What is hasty generalization? A hasty generalization is a conclusion based on an insufficient or nonrepresentative sample. In other words, you jumped to a conclusion too quickly based on just a few examples that don’t necessarily represent the whole. As such, your logic is flawed.

Here are a few examples:

There are no good restaurants in New York; I tried two in Times Square, and they were awful.

LAX is the worst airport; the last time I flew out of there, my flight was delayed.


I just heard about a guy from Mexico who was caught smuggling drugs across the U.S. border. As you can see, all Mexicans are bad hombres.

Why it’s wrong: These isolated examples likely do not represent the larger groups mentioned.

How to fix it: Mention only the specific instances; don’t generalize to a larger group. Or, if you do have data to back up a larger generalization (which is rare; see below), cite the research.

Generalizations based on study samples

Some generalizations are less overt. Often, we may misrepresent a finding by extrapolating it to a group it doesn’t necessarily represent. Most study samples are not statistically representative, meaning they don’t contain the same proportions of demographics and other characteristics as the general population. Unless the research used a statistically representative sample, then we can’t generalize the results to a wider group.

Example #1

Say you’re writing about a case report of a child who died after being bitten by a garden spider. You may be tempted to write a lede like this: Garden-spider bites may be deadly, a case report shows.

Why it’s wrong: It’s misleading to imply that it’s common for the bites of these spiders to be deadly just because this one child died from the bite. (In fact, these spider bites are harmless to humans.) This particular child may have had a severe allergy or other condition that worsened the effects of this spider bite. Or, there might have been other conditions surrounding the incident that ultimately led to his death. There are millions of possibilities, and we can’t generalize just from this one example.

How to fix it: Be more specific, based on the information from the report. Make it clear that we’re reporting on just one instance:

In an unusual case, a child died after being bitten by a garden spider. Although these creepy creatures are almost always harmless to humans, this 3-year-old boy had a bizarre allergic reaction to the spider that ultimately led to his death, a new case report shows.

Example #2

You’re writing about a survey of 50 Wall Street executives. Forty of them (80 percent) said they work more than 50 hours per week. You incorrectly write the following lede:

More than three-quarters of employees are working more than 50 hours per week, new research finds.

Why it’s wrong: This survey had a small, homogenous sample. There are likely many aspects of this particular profession and the people in this particular group (location, gender, income level, etc.) that biased the results. Just because these Wall Street execs worked this many hours doesn’t mean all people do. It’s also a very small sample size. Therefore, we cannot generalize the results to everyone.

How to fix it: Be specific: Eighty percent of Wall Street executives surveyed in a new study said they work more than 50 hours per week.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s