Module 6 - Manley Ch 7 - Causes Flashcards
What is the fallacy of post hoc ergo propter hoc?
Assuming that because one event follows another, the first caused the second.
Why are humans prone to seeing causal connections even when they don’t exist?
Our minds are pattern-seeking and evolved to err on the side of caution in detecting potential threats or opportunities.
What is a correlation?
A relationship where one factor occurs at a higher rate when another factor is present, but without necessarily implying causation.
What does it mean for a correlation to be symmetrical?
If A is correlated with B, then B is also correlated with A.
How can selective noticing lead to illusory correlations?
By focusing on instances that confirm a perceived relationship while ignoring counterexamples.
What are distal and immediate causes?
Immediate causes directly trigger an event, while distal causes are broader, underlying factors.
Why is it misleading to ask “What is the cause?” for complex events?
Complex events typically result from multiple interconnected causes, not a single factor.
Why are scientific findings often misinterpreted in popular media?
Media tends to oversimplify results, focus on surprising findings, and interpret correlations as causal relationships.
How can we better understand the validity of a study reported in the media?
By consulting the original study, expert opinions, or meta-analyses that synthesize evidence from multiple studies.
What is reverse causation?
When we incorrectly assume A causes B, but in reality, B causes A.
How can reverse causation be identified?
By studying changes over time or designing experiments where causation can be tested.
What is a common cause?
A third factor, C, that causes both A and B, creating a misleading correlation.
Give an example of a misleading correlation due to a common cause.
Broccoli consumption and health—both may be influenced by health-conscious behavior (common cause).
What is the placebo effect?
When a treatment’s effectiveness is due to the expectation of improvement rather than the treatment itself.
Why are placebo-controlled trials important?
They isolate the true effect of a treatment by accounting for the placebo effect.
What is regression to the mean?
The tendency for extreme data points to move closer to the average on subsequent measurements.
How can regression to the mean mislead conclusions about interventions?
Improvements after an intervention might occur naturally due to regression rather than the intervention’s effectiveness.
What is an example of a spurious correlation?
Divorce rates in Maine correlating with margarine consumption per capita.
How can spurious correlations be avoided?
By evaluating causal plausibility and considering the role of selection effects in identifying patterns.
What are the three steps in an argument from correlation to causation?
We observed a correlation between A and B.
There is a general correlation between A and B.
A causes B.
What kind of evidence is needed to support the hypothesis “A causes B”?
Evidence with a high strength factor, which is:
Far more likely if A causes B than if it doesn’t.
Far more likely if A does not cause B than if it does.
What is the gold standard for establishing causation from correlation?
A double-blind randomized controlled trial (RCT).
What does “randomized” mean in a randomized controlled trial?
Subjects are randomly selected and placed into two groups: the treatment group and the control group.
What does “double-blind” mean in a randomized controlled trial?
Neither the subjects nor the experimenters interacting with or assessing them know which group the subjects are in.
Why is a placebo treatment used in a double-blind trial?
To ensure that the placebo effect is accounted for equally in both the treatment and control groups.
What sources of error does a double-blind randomized controlled trial eliminate?
Bias in the sample.
Reverse causation.
Common cause.
Placebo effect.
Regression to the mean.
How does randomization in an RCT eliminate sampling bias?
It ensures that subjects are randomly assigned, preventing systematic differences between the groups.
Why is a difference between groups in an RCT strong evidence for causation?
Because all other factors (e.g., bias, placebo, regression to the mean) are controlled for, leaving the treatment as the most likely cause of the difference.
What strengthens the evidence from a study beyond a single randomized trial?
Robust evidence from multiple studies with consistent results.
