Flaw II: Sampling, Representativeness & Statistical Flaws
Recap: Flaw, Part 2 — Statistics and Sampling
A second major Flaw cluster, distinct from causal flaws: arguments that misuse a statistic or generalize from an unrepresentative sample.
1. Unrepresentative Sample
A survey or study group that isn't representative of the population the conclusion is about (self-selected respondents, a narrow demographic, a non-random sample).
2. Percentage vs. Raw Number Confusion
An argument that treats "the percentage increased" as if it proves "the raw number increased" (or vice versa) — these can diverge if the total population size also changed.
3. Small Sample Size
Drawing a general conclusion from too few data points to be statistically meaningful.
4. Survivorship-Style Bias
Drawing conclusions only from the cases that remain observable, ignoring cases that dropped out or failed silently.
Practice items on this page are curriculum-authored in LSAT style; they are not official LSAC questions. Real drilling happens on LawHub — see the assignment below.
Q1. Argument: "The percentage of the company's revenue from software rose from 20% to 40% last year. So software revenue must have doubled." What's the flaw?
LawHub drill-tier LR section, untimed. Scan for Flaw stems involving statistics or samples.
An RC maintenance day tomorrow — Main Point and Primary Purpose in RC — before the third and final Flaw day, on conditional-logic flaws.