Lecture: Methods Flashcards
Hypothesis
Question turned into prediction
Case study
intensive study of one or few individuals
What are the advantages of case studies?
important insights
rare conditions
can disconfirm a hypothesis
What are the limitations of case studies?
Hard to generalize
Hard to rule out alternatives/factors (no control condition)
Correlational study
Analyze the relationship (association) between two systematically measured variables
reported with r (1 to -1)
What are the strengths of correlational study?
suggestive about relationships can lead to more research
can make predictions
If X and Y are highly correlated, if X goes up, can predict Y
What are the limitations of a correlational study?
Correlation does not equal causation
Directionality- does not know what variable affects which
Third variable problem (which can cause both variables)
Longitudinal study
Follows the same subjects over time
What are the strengths of longitudinal study?
Address directionality issue
Address the third variable by measuring it and trying to control it
Experiment
One control group (at least 2 groups)
Random assignment (randomized controlled trial)
control: no treatment
experimental: treatment
Random assignment
Everyone has an equal chance of being in any condition
One person’s assignment doesn’t affect anyone else
can be “between-subjects” design
can be “within-subject” (everyone is in their own control) - problem can be that people might figure out the study order effects
Inferential statistics
Infer about population from a sample
What does it mean for a study to be statistically significant?
p-value: effect due to chance
p < 0.05 is not due to chance
Thus, we want a low p-value
The significance does not equal importance
What is p-hacking?
P-hacking using any means (other than fraud) to achieve statistical significance
How can you combat p-hacking?
Preregister the study so you can’t change the hypothesis later
