Research Methods Exam Flashcards
How do we generally investigate causal claims?
Experiments
What are the 3 criteria for establishing causation?
- Covariance
- Temporal Precedence
- Internal Validity
Explain Covariance
An association establishes that A causes B or B causes A
Explain Temporal Precedence
directionality, figuring out what came first; did A cause B? or did B cause A?
Explain Internal Validity
is there a third variable that is associated with both A and B independently that could interfere with causation?
What is an experiment?
the manipulation of one variable and the measurement of another
independent variables can have multiple…
conditions
what are the two types of varaibles?
- independent
- dependent
explain independent variables
manipulated (ex. note taking methods)
explain dependent variables
measured (ex. academic performance)
what is a control group?
receives no treatment; placebo group
what is a treatment group?
receives the treatment
what things should you keep in mind when choosing variables and methodology?
- replicability
- generalizability
- ability to make causal claims
- outside interfering varaibles
why do we need experiments?
allows us to draw conclusions about causation
what is the easiest criteria to establish?
covariance
what happens when results are explained by systemic differences (confounds)?
we cannot infer causation
what are confounds?
alternative explanation for the change in the dependent variable
what are the two types of confounds?
- design confounds
- selection effects
what are design confounds?
mistakes when designing the experiment
example of a confound within this claim: alcohol use increases your risk of lung cancer
individuals who use alcohol may be more likely to also smoke
confound: smoking
what is a selection effect
errors in the selection or participation of participants
unsystematic variability is not the same as…
confound
whats a secondary way that selection bias occurs?
when people volunteer for a study
how do you prevent confounds?
make sure researchers treat participants the same and try to make sure there aren’t parts of the experiment that vary systematically
how do you prevent selection effects?
random assignment
what is a way to make sure groups are equal?
matched groups
why can dealing with confounds be difficult?
time
measuring more variables
resources
more complicated study designs
lots of variables
missed variables