Experimental logic and Variables 1.2 Flashcards
True Experiments
Test causal relationships, compare equivalent groups
Quasi-experiments
Confidence is causality varies, typically compare nonequivalent groups, measure a DV.
Correlational research
Not a test of causal relationships, Measure two variables and how they relate to each other
Variable types
•Independent variable (IV)
- Value is determined and manipulated by the researcher
- Can create equivalent groups by random assignment to ppts to different levels
•Subject variable
-Variability determined by researcher but cannot be manipulated in the sense of
random assignment
-Cannot create equivalent groups through random assignment
•Example: children vs. adults
•Dependent variable (DV)
- Value depends on the independent or subject variable
- Example: reading time
Heuristics for identifying variables in descriptions
Independent variable
•“Manipulated”, “compared”, “vs.”
•Description implies that this variable influencing another variable
•A characteristic that could be randomly assigned
Subject variable
•“Compared”, “vs.”,
•Description implies that this variable is influencing another variable
•Characteristic that could NOT be randomly assigned
Dependent variable
•“Measured”
•Description implies that this variable was influenced by another variable
What are constructs?
Conceptual or theoretical term for the variable. Ex: moral judgement, learning
What are operational definitions?
Concretely specify how the construct was manipulated or measured. There are many ways to operationalize the same construct and od’s can often measure multiple constructs.
Causal realtionship
In True experiments: Manipulate IV and measure how DV changes, compare eq. groups
Internal Validity
Pertains to the mechanics of experiments, is it the IV that causing change in DV, free of confounds
What are confounds?
Variables that correlation with the levels of the IV
•Provide an alternative causal explanations for the results
Levels
ex: construct: emotions
levels: happy x sad
