Causal Reasoning: Correlation, Alternate Causes & Reversal
Recap: A Different Kind of Logic
Conditional logic is about guarantees: if p, then q, no exceptions, full stop. Causal logic is about explanations: this correlation exists, and the author is proposing a cause for it. Causal claims are inherently weaker and more attackable than conditional ones — which is exactly why they're the backbone of Weaken questions and half of Flaw questions.
CS Bridge
Causal flaws are confounders in an A/B test. If a metric moved after you shipped a feature, that doesn't prove the feature caused the move — there could be a confounder (something else changed at the same time), reverse causation (the metric change caused the feature to ship, e.g. via a feedback loop), or plain coincidence. LSAT causal-argument critique is the same skepticism, applied to prose instead of dashboards.
1. Recognizing Causal Conclusions
Watch for language like "causes," "leads to," "is responsible for," "the reason for," or simply an inferential leap from "X and Y are correlated" straight to "X causes Y." Any time an argument concludes a causal relationship from an observed correlation, you're looking at a causal argument — and causal arguments are attackable in five standard ways.
2. The Five Standard Attacks (Weaken)
- Alternate cause — something else could explain the effect.
- Reversed causation — maybe Y causes X, not X causes Y.
- Cause without effect — cases where X happened but Y didn't follow.
- Effect without cause — cases where Y happened without X.
- Coincidence — the correlation might just be chance, especially with small samples.
3. The Five Standard Supports (Strengthen)
Strengthen answers for causal arguments typically do the mirror image: rule out an alternate cause, rule out reversed causation, show the effect tracks the cause across many cases (dose-response), or report a controlled experiment where the cause was manipulated and the effect followed.
4. Plan and Prediction Arguments
A special case: "doing X will cause Y" (a policy proposal or prediction). These get weakened by anything showing X won't actually happen, will happen but won't cause Y, or will cause Y but also cause a worse side effect Z that undermines the point of doing it.
5. Where Causal Reasoning Feeds Other Types
Flaw questions frequently are just "the argument mistakes correlation for causation." Weaken and Strengthen questions built on causal stimuli are the single largest recurring stimulus type in LR. Resolve-the-Paradox questions often present two causally-in-tension facts and ask for a third fact that lets both be true.
6. Traps
The trap is treating "correlation implies causation" as the flaw label to reach for reflexively, without actually checking whether the stimulus supports one of the five specific attacks. Vague pattern-matching gets you to the right family but the wrong specific answer.
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.
Original Practice Items
Q1. A study finds that employees who eat lunch away from their desks report higher job satisfaction, and the author concludes that eating away from one's desk causes higher job satisfaction. Which best weakens this?
Q2. Which of the following, if true, would most strengthen a claim that a new fertilizer causes increased crop yield?
Official Drill Assignment
LawHub free lessons covering causal reasoning arguments.
Peek into Tomorrow
Back to Reading Comprehension: how to build a passage map on the fly, with a worked example, plus your first official RC drill from the practice tier.