Association & Causation Flashcards
What are variables?
Variables are usually what we measure or manipulate
What are the 4 types of variables?
-Dependent
-Independent
-Confounding/extraneous
-Control
What is association?
Typically = statistical relationship between two variables
What is correlation?
Refers to extent pair of variables are linearly related
What can association & causation be useful for?
-Can allow us to infer potential causal relationships (maybe..)
-Predictive relationships more so
–> BUT need v. solid empirical design & justification
What are spurious correlations?
Relationship between 2 otherwise completely unrelated variables
–> Variables can be correlated or associated without actually being related
What are 3rd variable problems?
Variables might be linked as a third problem is causing both (confounds the data)
What is Anscombe’s quartet?
-Statistical relationship is not indicative of distribution
-Careful about summary statistics
What is the most important thing to remember about correlation?
Correlation does NOT imply causation!!!
Fill in the table about correlations.
What is cause & effect for association?
Exposure associated with outcome
What is cause & effect for causation?
Exposure leads to outcome
Fill in this common single comparison test chart - for causation.
What are some of the issues surrounding causation?
-Finding a ‘result’ does not always reflect reality
-Comes from good design
What are confounding variables?
-Influences exposure & outcome
-Should control for
—> to ensure exposure causes outcome & not confounder