What makes a claim causal?
A causal claim says that changing one factor would make a difference to an outcome, rather than merely occurring alongside it.
Study 4 Causation and Scientific Inquiry with 12 free online flashcards. Review key terms, definitions, and concepts with this interactive flashcard deck.
What makes a claim causal?
A causal claim says that changing one factor would make a difference to an outcome, rather than merely occurring alongside it.
What do potential outcomes represent?
A causal effect is a contrast between the outcomes a unit would have under different conditions.
Why is a causal effect hard to observe for one unit?
For one unit, only the outcome under the condition actually experienced can be observed; the alternative is the counterfactual.
How does random assignment support causal inference?
It assigns units to conditions by chance, tending to balance known and unknown background causes between groups.
Does randomization guarantee identical groups?
Chance imbalances can still occur, especially in small samples; good measurement, follow-up, and attention to missing data and noncompliance also matter.
What defines an observational study?
Researchers record exposures and outcomes without assigning the exposure.
What is confounding?
A confounder influences both exposure and outcome, creating or distorting their association.
What is reverse causation?
The outcome, or an early stage of it, influences the supposed cause; illness changing a behavior is one example.
What is selection bias?
It occurs when study inclusion depends on factors related to exposure or outcome, making included people systematically different from those left out.
How can measurement error weaken a causal study?
It arises when exposures, outcomes, or confounders are recorded inaccurately.
Why can an ambiguous treatment definition be a problem?
A label may cover many different versions of an intervention, making it unclear which causal comparison is being made.
What does positivity require?
Each relevant type of unit must have some chance of receiving each condition being compared.