scientific method Flashcards
meta-science
- how we know what we know
- how do we learn what we don’t know
confirmation bias
- the mother of all biases
- ignore what agrees with us and accept what does agree with us
limits of:
- personal experience
- testimonials/anecdotes from others
- human judgement
- opinions of ‘smart’ people
scientific method
best protection against sloppy thinking and human judgement
hallmarks of scientific method
- objective observation and logically necessary conclusions
- parsimonious explanations
- replication
- skepticism
- careful designs
- falsifiability
- open-mindedness
theory
explanation that integrates principles and organizes and predicts behaviour or events
hypothesis
testable prediction to enable us to keep, reject, or revise the theory
construct
the concept being measured
operational definition
the method used to measure/define the construct
replication
same finding w/ diff. participants in diff. situations or when using diff. meausures
random sampling
- when each member of a population has equal chance of inclusion in a sample
- also called an unbiased sample
descriptive studies
- purpose: careful and accurate description
- ex. amount of sleep, depression
correlational studies
- purpose: evaluating relationships
- ex. is amount of sleep related to depression
r = 0.00
no association
r = 0.20
small association
r = 0.40
moderate association
r = 0.60
large association
illusory correlation
the perception of a relationship where no relationship exists
experimental studies
- purpose: exploring cause and effect
- ex. does amount of sleep cause depression?
double-blind randomized trial
research participants and the researcher are both unaware of the participant’s condition
mode
most frequently occurring score in a distribution
mean
average of values in a distribution
median
the middle score in a rank-ordered distribution
range
the difference between the highest and lowest scores in a distribution
standard deviation
a computed measure of how much the values vary around the mean
when is an observed difference reliable?
- larger samples
- larger mean difference (or correlational relationship)
- less variability
- when statistically, the observed difference in the sample is unlikely to happen due to chance (p-value)
