The LSAT’s Most Common Flaws

Confusing Sufficient with Necessary

Confusing sufficient and necessary conditions is hands-down the most common flaw on the LSAT. It’s also a flaw that tends to trip up novices the most. But understanding the difference between sufficient and necessary is a lot simpler than you might think.

Sufficient-necessary flaws are so crucial to understand that they deserve an article of their own. Learn how to interpret conditional statements in commonsense terms and how to spot the sufficient-necessary flaw every time here.

‍

‍

Confusing Correlation with Causation

Inferring causation from mere correlation in the second most common flaw tested on the LSAT. 

Two things are correlated when they are associated with or connected to each other in some way. In short, when one thing happens, the other thing also tends to happen.

Causation, on the other hand, means that one thing’s occurrence actively brings about the occurrence of the other.

The mere fact that two things (A and B) are correlated doesn’t prove that one (A) causes the other (B). There could be an alternative explanation for the observed correlation:

1. The causal relationship might go in the opposite direction (instead of A causing B, B causes A). 

2. Both things might be caused by some third variable (C causes A, and C also causes B).

3. Less frequently, the correlation might be due to random coincidence (A and B are not related).

Let’s put these concepts into concrete terms using an example argument:

When ice cream sales go up, murder rates increase. Therefore, eating ice cream must cause people to kill each other.

This argument is flawed because the existence of a correlation between ice cream sales and murder rates is not enough to establish that ice cream sales drive murder rates. More likely, a third factor is responsible for the observed correlation. Ice cream sales increase in the summer because it’s hot out. Murder rates also increase in the summer because people spend more time outside around one another, resulting in more opportunities for conflict.

Summer, in this scenario, is a third variable that drives both ice cream sales and murder rates. Ice cream sales and murder rates are thus correlated but not causally connected to each other.

How to Tell a Causal Statement from a Correlative Statement

To figure out whether a sentence asserts causation or mere correlation, focus on the verb of the sentence. Does it say that the first thing actively affects the second thing? If so, that’s causation. Or does it say that the first thing just has a tendency to vary with the second thing? That’s correlation.

The following sentences are causal because they claim that one thing affects the other:

• Sodium not used by the body will increase blood pressure.
• His behavior sometimes leads to overall poor performance in his job.
• Situational factors account for most code-switching.
• Facebook is a factor in shaping the social morals of modern America.
• These vegetables were depleted of nutrients because of an earlier failure to rotate crops.

The following sentences are only correlative because they say only that two things happen together:

• Democracies are more likely than nondemocratic forms of government to have policymakers who understand the complexity of governmental issues.
• Those who buy expensive running shoes tend to exercise more often than those who buy cheap ones.

In the correlation examples above, we don’t know why the two things happen together, only that they do. It would be a logical fallacy to claim that just because they happen together, one causes the other.
‍

‍

Other Common Flaws

Nonrepresentative Samples

Any sample from which you draw your conclusions must be representative of the population to which you attribute your findings. If this sounds like gobbledygook, consider a concrete example: 

Let’s say you want to figure out the average lifespan of all human beings. A representative sample of such a broad population would need to reflect the vast diversity of the human population. Attributes like gender, race, socioeconomic status, culture, health, ability status, and lifestyle should all be represented. 

If a sample is not representative, then observations drawn from it are not generalizable to the broader population.

Here are the two main (often related) ways a sample can be nonrepresentative:

1. The sample is nonrandom. For instance, if you’re trying to draw conclusions about human beings in general, but your sample is composed entirely of Portuguese men, then your sample is biased. It fails to represent the great racial and gender variation that exists across all human beings.
   ‍
2. The sample is too small. The smaller the sample size, the less likely it is to accurately reflect the diversity of the broader population. For instance, you cannot claim that all humans can live to the age of 105 merely because your great-uncle just turned 105 years old. He might be an outlier!

Remember a lawyer’s currency is the strength of their argument. Arguments often rely on empirical evidence for their factual basis. And strong evidence requires strong, valid methodology, including the use of representative samples.

Don’t let opposing counsel get away with presenting evidence that’s rotten because it’s based on a nonrepresentative sample. Catch that flaw and bust it wide open.

Numbers Shifting Between Proportional and Absolute

Let’s say you want to describe how much your investment portfolio has grown this year. You may express a numerical change in one of two ways:

1. Using proportional (relative) language: “My portfolio has grown by 25% of last year’s value.” This statement doesn’t say what your portfolio was worth this year or last year, nor does it communicate your portfolio’s absolute growth. But it does tell you the change in this year’s value relative to last year’s.
   
   Let’s say your initial investment was $5. A “25% increase” might sound significant, but really it’s a difference of only $1.25. If, on the other hand, your initial investment was $10 million, a 25% increase would amount to $2.5 million! Describing numbers only in proportional terms says nothing about their absolute values. 
   ‍

2. Using absolute language: “My portfolio’s value has increased by $50 this year.” This statement communicates exactly how much your portfolio has grown in absolute terms. But $50 could be a little or a lot relative to the total value of your portfolio.
   
   If your initial investment was $5 and has since grown to $55, you’d be over the moon—that’s a 1,000% increase! But if your initial investment was $10 million, a $50 increase would be trivial.

The way that numbers are communicated—in relative versus absolute terms—can make a numerical change sound bigger or smaller than it actually is. Beware of LSAT arguments that draw relative conclusions from absolute numbers or vice versa.

Consider the following flawed argument:

The Halifax Police Department made 50 more arrests this month than they made last month. But 100 more crimes occurred this month than occurred last month. Therefore, the crime rate has increased more rapidly this month than has the arrest rate.

This argument draws a relative conclusion (about the relationship between the arrest rate and the crime rate) based on absolute numbers. The argument is invalid because it does not provide enough information to soundly draw a relative conclusion.
‍

‍