Mastering Cause & Effect Reasoning
The Foundation of All Critical Reasoning
A guide to identifying, weakening, and strengthening causal arguments.
What is Causal Reasoning?
It's the human brain's favorite shortcut—and its biggest flaw.
Causal Reasoning is any argument that claims one thing *causes* another.
Premise (The Observation): "After the city launched a new ad campaign, tourism increased by 20%."
Conclusion (The Causal Claim): "Therefore, the ad campaign *caused* the increase in tourism."
The "Master Key" of CR
Understanding causality is *not* just one question type. It is the key to 4 other types.
WEAKEN
...by showing an alternate cause.
STRENGTHEN
...by ruling out alternate causes.
ASSUMPTION
...the argument *assumes* no alternate cause.
EVALUATE
...by asking, "Does an alternate cause exist?"
The #1 Flaw: Correlation ≠ Causation
This is the most famous flaw in all of logic. It's the assumption that because two things happen together (a correlation), one must have *caused* the other.
The Classic Example
Observation 1
"In summer, ice cream sales increase."
Observation 2
"In summer, crime rates increase."
The Flawed Conclusion: "Therefore, ice cream sales *cause* crime."
The Real Answer: The Third Factor
The flaw is ignoring the *real* cause. A "Third Factor" (also called a "Lurking Variable") is causing *both* observations.
THIRD FACTOR: Hot Weather
Observation 1
Hot weather causes people to buy ice cream.
Observation 2
Hot weather causes more people to be outside, leading to more crime.
Anatomy of a Causal Argument
Premise: `A` is correlated with `B`.
Conclusion: `A` *causes* `B`.
Causal Signal Words
When you see these words in a *conclusion*, you are looking at a causal argument.
The Causal Assumption
Every causal argument makes a *huge* assumption. It's the lynchpin of the entire argument.
"The argument *assumes* that there is NO OTHER explanation for the correlation."
This includes...
• No Third Factor
• No Reverse Causality
• It's Not Just a Coincidence
How to Weaken a Causal Argument
To weaken a causal claim (`A ➔ B`), you must show that `A` is *not* the cause. There are 3 main ways.
Weaken #1: Introduce a Third Factor (Z)
This is the most common method. You provide an alternate cause (`Z`) that caused *both* `A` and `B`.
Example: "The ad campaign (A) didn't cause sales (B) to rise. A new national holiday (Z) was declared, which caused *both* more ad sales (A) and more product sales (B)."
Weaken #2: Show Reverse Causality
You show that the argument got it backwards: `B` actually caused `A`.
Example: "The ad campaign (A) didn't cause sales (B). The company *saw* that sales were already rising (B), so it *decided* to launch an ad campaign (A) to capitalize on it."
Weaken #3: Show It's a Coincidence
You show that the two events are statistically unrelated or just a fluke.
Example: "The ad campaign (A) ran in February, but the sales increase (B) didn't happen until November. The two events are unrelated." (Less common, but possible).
How to Strengthen a Causal Argument
To strengthen a causal claim (`A ➔ B`), you must *defend* it from the 3 attacks. You are showing that `A` is, in fact, the *only* likely cause.
Strengthen #1: Rule Out Third Factors
You provide evidence that *disproves* a specific alternate cause.
Example: "The ad campaign (A) *did* cause sales (B) to rise, because we have data showing that our main competitor *did not* go out of business (ruling out Z-factor #1) and the economy *did not* improve (ruling out Z-factor #2)."
Strengthen #2: Rule Out Reverse Causality
You provide evidence (usually timing) showing that `B` could not have caused `A`.
Example: "The ad campaign (A) must have caused sales (B) to rise, because sales were *falling* for six straight months *before* the campaign (A) was launched."
Strengthen #3: Provide a "Control Group"
This is a *very* strong method. You show that when `A` *doesn't* happen, `B` *also* doesn't happen.
Ran Ad (A) ➔ Sales Up (B)
No Ad (~A) ➔ Sales Flat (~B)
Example: "In cities where we ran the ad, sales rose. In similar cities where we *did not* run the ad, sales were flat. This shows the ad was the cause."
Strengthen #4: Show the Mechanism
You provide data *how* `A` led to `B`, closing the logical gap.
Example: "The ad (A) caused the sales (B) because post-ad surveys showed a 50% increase in brand recognition, and 30% of new customers cited the ad as their reason for purchasing."
Practice Set 1 (4 Options)
Passage 1 (Weaken): "A study showed that people who own pets have lower blood pressure. Therefore, pet ownership *causes* a reduction in stress, which leads to lower blood pressure."
Which of the following, if true, most seriously *weakens* this argument?
Claim: Pets (A) ➔ Lower Stress (B).
(A) Strengthens: This shows a *mechanism* for how pets (A) cause the effect (B).
(B) & (D) Irrelevant: These don't address the causal link.
(C) Correct. This is a classic Reverse Causality (or Third Factor) argument. It states that low stress (B) *causes* people to get pets (A). This breaks the argument's claim.
Passage 2 (Strengthen): "After our company made its new 'GetFit' app free, downloads soared. This shows that the price-drop *caused* the increase in downloads."
Which of the following, if true, *strengthens* this conclusion?
Claim: Price Drop (A) ➔ More Downloads (B).
(A) Weakens: This provides a Third Factor (magazine feature) that could have caused the downloads.
(B) Correct. This is a perfect Control Group. We see that when the cause (A) is absent, the effect (B) is also absent. This *rules out* third factors (like "fitness is suddenly popular").
(C) & (D) Irrelevant: These describe what happened *after* the download, not what *caused* it.
The Master Key: Causal Links
Let's see how this one concept unlocks all the other CR questions.
Causality in... Assumption
Argument: "Our new ad campaign (A) led to a 20% rise in sales (B)."
The question asks: "The argument *assumes* which of the following?"
The Answer Will Be:
• "The rise in sales was not caused by a new national holiday." (Assumes no Third Factor)
• "Sales were not already rising *before* the campaign began." (Assumes no Reverse Causality)
An assumption question *is* a causal question.
Causality in... Evaluate
Argument: "Our new ad campaign (A) led to a 20% rise in sales (B)."
The question asks: "The answer to which question would be most useful in *evaluating* the argument?"
The Answer Will Be:
• "Did a major competitor go out of business during the same period?" (Tests for a Third Factor)
• "What were the sales trends in the month *before* the campaign?" (Tests for Reverse Causality)
An evaluate question *is* a causal question.
Causality in... Paradox
Paradox: "We launched a new ad campaign (A), but our sales *fell* (B)."
The question asks: "Which of the following *resolves* this paradox?"
The Answer Will Be:
• "The ad campaign (A) was accidentally offensive, which *caused* a consumer boycott (C), *leading to* falling sales (B)."
The resolution is often a *new causal chain*.
Causality in... Flaw
Argument: "After we launched the ad campaign (A), sales rose (B). Therefore, the campaign (A) *caused* sales to rise (B)."
The question asks: "The argument is flawed because it..."
The Answer Will Be:
• "...mistakes a temporal sequence for a causal relationship."
• "...overlooks the possibility that a third factor caused both events."
A flaw question *is* a causal question.
Padding Slide: Causal Chain
Sometimes, the argument is a "causal chain."
A ➔ B ➔ C
Argument: "The new law (A) will cause businesses to hire fewer people (B), which will cause unemployment to rise (C)."
To Weaken: You only need to break *one link*. (e.g., "The law (A) will *not* cause businesses to hire fewer people (B)."
To Strengthen: You must defend *both* links.
Padding Slide: Multiple Causes
Sometimes, there are multiple causes for one effect.
(A + B) ➔ C
Argument: "To get a promotion (C), you need both high sales (A) and good reviews (B)."
Weaken: "Show that one is not necessary." (e.g., "John got a promotion (C) with high sales (A) but *bad* reviews (~B).")
Strengthen: "Show that both are required." (e.g., "Mary had high sales (A) but was *not* promoted (C) because she had bad reviews (~B).")
Padding Slide: The "Post Hoc" Flaw
This is the formal Latin name for the causal flaw: *Post Hoc Ergo Propter Hoc*.
"After this, therefore because of this."
This is simply a fancy name for mistaking *temporal order* (what came first) with *causation*. It's the same flaw as Correlation vs. Causation.
Example: "I wore my lucky socks (A), and then my team won (B). Therefore, my socks (A) *caused* the win (B)."
Practice Set 2 (5 Options)
Passage 3 (Assumption): "A new study concludes that drinking coffee *leads to* higher productivity, noting that employees who drank 3+ cups of coffee per day had the highest output."
The argument relies on which of the following assumptions?
Claim: Coffee (A) ➔ Productivity (B).
This is an Assumption question, which *is* a causal question. We are looking for an answer that *rules out* an alternate explanation.
(D) Correct. This rules out a Third Factor/Reverse Causality. It denies that a high-pressure workload (Z) *caused both* high productivity (B) and high coffee-drinking (A). If this assumption is *false*, the argument dies.
Passage 4 (Flaw): "The university's new, expensive library was built to improve student grades. However, in the two years since it was built, the average GPA at the university has not increased. Therefore, the new library has failed to improve student grades."
The argument is flawed primarily because it overlooks the possibility that...
Claim: Library (A) ➔ Better Grades (B). The argument claims this is *false* because B didn't happen.
Flaw: It assumes *only* the library (A) affects grades (B).
(A) Correct. This points out the flaw. It provides a Third Factor (Z: higher standards) that is *suppressing* the effect. The library (A) *is* causing grades to go up, but the new standards (Z) are pushing them down, so the *net* effect is zero.
Padding Slide: Causal vs. Conditional
This is a subtle but important difference.
Causal: "Watering the plant *causes* it to grow." (This is a scientific claim about the world. It can be weakened by a Third Factor, like "sunlight.")
Conditional: "If you water the plant, it will grow." (This is a 100% certain logical statement. It cannot be weakened by a Third Factor. It can *only* be proven false by a counterexample: "I watered it, and it did not grow.")
Most CR arguments are causal, not conditional, because they are real-world claims about what *leads to* what.
Padding Slide: Practice 5
Argument: "A new law requires motorcyclists to wear helmets. Since the law's passage, deaths from motorcycle accidents have decreased 30%. Therefore, the helmet law is working."
Which of the following, if true, most *weakens* the argument?
(A) "During the same period, the city launched a major campaign against drunk driving."
(B) "The number of people riding motorcycles also decreased by 30%."
Analysis: Both (A) and (B) are Third Factors that weaken the argument. (A) provides an alternate cause for *fewer deaths*. (B) provides an alternate cause for *fewer deaths* (fewer riders = fewer accidents). Both are valid weakeners.
Padding Slide: Practice 6
Argument: "Our company's wellness program (A) has been a great success. Last year, employees who *voluntarily* signed up for the program (Group 1) had 50% lower healthcare costs than employees who did not (Group 2)."
This argument is flawed because it fails to consider that...
Flaw: This is Reverse Causality (or self-selection bias). The program (A) didn't *cause* good health (B). The employees who were *already* healthy (B) were the ones who *voluntarily* signed up (A). Group 2 is not a valid control group.
Padding Slide: Summary of Attacks
CLAIM: A ➔ B
"But what about Z?" (Third Factor)
"But what if B ➔ A?" (Reverse Causality)
"We checked, and Z did not happen." (Rule out Third Factor)
"We checked, and A happened before B." (Rule out Reverse Causality)
Final Interactive Quiz
Test your mastery. Find the causal link.
Summary & Next Steps
Key Takeaways
- Correlation is not Causation. This is the #1 flaw.
- To WEAKEN: Introduce a Third Factor or show Reverse Causality.
- To STRENGTHEN: *Rule out* a Third Factor or *use a Control Group*.
- This logic is the "Master Key" for Assumption, Evaluate, and Flaw questions.
Practice Plan
- Become a Skeptic: Read news headlines ("This 1 simple trick *causes* X!"). Pause and ask: "What's the Third Factor?" (e.g., People who do the trick are also wealthy, etc.)
- Drill Weakeners: For every causal argument you find, try to invent *both* a Third Factor and a Reverse Causality scenario.
- Look for Control Groups: In Strengthen questions, the *best* answer is often the one that provides a control group.