Number vs. Percentage Flaw on the LSAT: The Confusion That Costs Points

Numbers feel authoritative. When an LSAT argument says 'City A had 500 more car thefts than City B last year,' most readers nod and accept that City A has a worse theft problem. But that conclusion does not follow — not without knowing how large each city is.

This is the number versus percentage flaw, and it is one of the most reliably tested mistakes in LSAT Logical Reasoning. Once you know its two forms, you will spot it in seconds.

The flaw works in both directions. An argument can move illegitimately from a raw number to a rate, or from a rate to a raw number. Both moves are invalid without additional information about the size of the underlying group.

An argument commits this flaw when it observes an absolute count — a number of events, people, or instances — and then draws a conclusion about frequency, likelihood, or proportion.

Example: 'Last year, Northside High School reported 40 disciplinary incidents, while Southside High School reported only 25. Therefore, students at Northside are more likely to misbehave.'

The flaw: the argument compares raw counts but draws a conclusion about likelihood — which is a rate. If Northside has 2,000 students and Southside has 400, Southside's rate is actually higher (6.25% vs. 2%). You cannot move from count to rate without knowing the base population.

What weakens this argument? Any answer choice that introduces information about different population sizes. What would be an assumption? That the two schools have roughly the same number of students.

The same flaw runs in the other direction. An argument observes that one group has a higher rate — percentage, proportion, or likelihood — and concludes that the group has a higher absolute count.

Example: 'Motorcycle riders are three times more likely to be involved in a fatal accident than car drivers. Therefore, motorcycles account for more fatal accidents than cars each year.'

The flaw: a higher rate does not guarantee a higher raw count if the groups are very different in size. There are far more car drivers than motorcycle riders in the country. Even at three times the rate, motorcycles could produce far fewer absolute fatalities. You cannot move from rate to count without knowing the size of each group.

The signal is a mismatch between the type of evidence and the type of conclusion. Ask yourself: is the evidence a raw number or a rate? Is the conclusion a raw number or a rate? If they are different types, look hard for the flaw.

A rate means any expression involving a proportion: percent, per capita, likelihood, odds, frequency, average, ratio. A raw number is just a count: 400 cases, 30 students, 12 incidents.

The wrong-answer traps on flaw questions for this stimulus type will often describe real flaws but not the correct one. You will see answers mentioning causation, circular reasoning, or false dilemma — none of which apply. The correct answer will specifically name the error of treating a count as if it established a rate (or vice versa), often phrased as 'the argument fails to consider that the comparison groups may differ in size' or 'concludes that one thing is more prevalent when it only shows it is more common in relative terms.'

Stimulus: 'A recent study found that people who exercise regularly report 30% more energy in daily tasks than people who do not. Therefore, the exercise group must be completing more total hours of productive work each day.'

Flaw: the stimulus moves from a rate (30% more energy) to a raw-count conclusion (more total hours of productive work). But the exercisers and non-exercisers may work very different baseline hours. A 30% energy boost on a 4-hour workday produces less absolute productive time than a modest increase on a 10-hour workday.

Correct flaw answer: 'The argument overlooks the possibility that the two groups may not start from the same baseline level of productive activity.'

What strengthens: Evidence that both groups have similar baseline work schedules.

What weakens: Evidence that non-exercisers work significantly longer hours than exercisers.

This mistake appears in Weaken, Strengthen, Assumption, and Paradox questions as well. On a Weaken question, the correct answer will often introduce a population-size discrepancy that undermines a rate-based conclusion built on raw numbers. On an Assumption question, the correct answer will supply the hidden premise that the two groups are comparable in size.

On Paradox questions, an apparent paradox sometimes involves one statistic being a rate and another being a count — and the resolution is showing why a high rate can coexist with a low absolute count.

Training yourself to notice number-versus-rate mismatches is one of the highest-leverage logical reasoning skills because the flaw appears across multiple question types.