Identifying types of association Flashcards
Causal vs correlational
causal: one variable directly or indirectly influences the other variable(s)
correlational: changes in one variable are accompanied by changes in the other(s)
Correlational design
if covariation exists: determine direction, magnitudes, form of relationship
nonexperimental bc no manipulation
Variables in experimental research
- independent variable (values set by the experimenter)
- dependent variable (depends on the independent variable)
- extraneous variable (affects investigated behavior but is not of interest, can lead to mistakes if not tightly controlled)
- quasi-independent variable (a correlational variable that resembles an independent variable (e.g. age)
How to minimize the effects of extraneous variable
hold constant or randomize and distribute over treatments
Developmental designs
non-experimental and experimental, either cross-sectional(several participants from each of a number of age groups), longitudinal(a single group of participants with similar age is followed over time) or cohort-sequential (both combined)
Descriptive design
observing and describing the behavior
Quasi-experiments
no manipulation, cases that already show varying variable instead of causing the variable to vary (e.g. participants that already have depression)
Placebo-effect
response to medication that has no effect because participant thinks it does
Confounding
variable varies along with the independent variable; can not be distinguished
Mediation
mediating variable C, A does not cause B directly but A causes C and C causes B
Moderation
A influences B ONLY if C (specifies cases for which the causation is valid/ existing)
Common response
lurking variable C causes variation
Causation
A causes B to vary
Simpson’s paradox
association that holds for all of several groups can reverse direction when the data is combined into one single group because of lurking variables
