Evaluating Statistical and Numerical Claims

LNAT passages frequently include statistical and numerical evidence — percentages, averages, survey results, comparisons, and trends. While the LNAT is not a maths test, it does test your ability to evaluate whether numerical claims genuinely support the conclusions drawn from them. Misleading use of statistics is one of the most common reasoning errors in argumentative writing.

---

Why Statistics Can Be Misleading

Statistics are powerful because they appear objective and precise. But the same data can be presented in ways that support very different conclusions. On the LNAT, you need to ask not just "What does this statistic say?" but "Does this statistic actually support the argument?"

---

Key Concepts

1. Percentages vs Absolute Numbers

A percentage can be impressive or alarming while the absolute number is trivial — and vice versa.

"The number of shark attacks in British waters has increased by 200% in the past decade."

This sounds terrifying. But if the number went from 1 to 3, the absolute increase is negligible. A 200% increase of a tiny number is still a tiny number.

Rule: Always consider the base number when evaluating percentage changes. A large percentage change from a small base may be statistically insignificant.

Conversely:

"Only 2% of applicants were rejected due to the new policy."

If there are 500,000 applicants, 2% means 10,000 people — a significant absolute number.

Claim	Percentage	Absolute number	Misleading?
"Infections doubled"	+100%	From 5 to 10	Potentially — small base
"Only 1% of residents affected"	1%	8,000 people (in a city of 800,000)	Potentially — large absolute number
"Crime up 50%"	+50%	From 20 to 30 incidents	Potentially — small base

2. Base Rate Neglect

Base rate neglect occurs when you focus on specific information and ignore the underlying probability (the base rate).

"A new medical test is 95% accurate. John tested positive. Therefore, there is a 95% chance John has the disease."

Why this is wrong: If the disease affects only 1 in 10,000 people, then even with a 95% accurate test, most positive results will be false positives. The base rate of the disease is so low that the 5% false positive rate generates more positive results than the actual cases do.

This is a sophisticated concept, but the LNAT may present simplified versions of base rate neglect in passages about screening, testing, or risk assessment.

3. Sample Size and Representativeness

"A survey of 50 people in central London found that 80% support the congestion charge."

Problems:

• Sample size: 50 people is far too small to draw reliable conclusions about public opinion
• Representativeness: People in central London are not representative of the wider UK population — they may benefit directly from the congestion charge

Factor	Good evidence	Poor evidence
Sample size	Large (thousands)	Small (dozens)
Selection method	Random sampling	Self-selected or convenience sample
Representativeness	Reflects the target population	Drawn from a biased subgroup
Response rate	High (>70%)	Low (<30%)

4. Averages: Mean, Median, and Mode

The word "average" can refer to three different measures, each giving a different picture:

Measure	Definition	When it is misleading
Mean	Sum of all values ÷ number of values	Distorted by extreme outliers
Median	The middle value when all values are ranked	Can hide the spread of data
Mode	The most frequently occurring value	May not represent the typical case

Example

"The average salary at this company is £75,000."

If this is the mean, it could be distorted by a few very high earners (the CEO earning £500,000 pulls the mean up significantly). The median salary might be £35,000 — a very different picture.