Causal Inference Flashcards
Define: dependent variable
the affected variable, the outcome of some process, ex. divorce rate
Define: independent variable
affects the dependent variable(s), ex. gender, race, income
Define: variables
factors that characterize people, have different values, seen as important to why people do the things they do and are the way they are
Define/explain: intervening, mediator, and mediating variables
variables that are affected by an independent variable in a way that affects the dependent variable’s outcome, ex. divorce rate (D) is affected by low income (I), because as income goes down stress (mediating) goes up, in all cases of stress (ex. death, low income, illness) divorce is higher
Define/explain: control, moderator, and moderating variables
a variable that changes the relationship between dependent and independent variable, can also change how mediating variables affect dependent variables
Define: positive/negative causal relationship between variables
as one goes up, the other goes up, and vice versa
Define: inverse causal relationship between variables
as one rises, the other goes down
Define: correlation
there exist an association between two variables that is not causal, one is not causing the other
Define: spurious effects
where a 3rd variable is affecting both independent and dependent variables and they have no direct relationship
What are 3 characteristics of a causal relationship?
- temporal precedence
- constant conjunction
- nonspuriousness
Define: temporal precedence
the independent variable is causing the effect in the dependent variable
Define: constant conjunction
causality should be seen consistently, across time/studies/populations
Define: nonspuriousness
the association between the two variables can’t be spurious
Define: causal mechanism
the reasons that the independent variable affects the dependent, a process, they link/create a pathway/model between the two variables
Define: hypothesis
specific expectations of relationships between variables, assumed to be trends not universality because group trends don’t determine individual outcomes