Methods Flashcards
Operational Definitions
How the researcher defines the variables in the experiment so that the variables are measurable
Sample
subset of a population
Stratified random sampling
Each subgroup of the population is randomly sampled in proportion to its size
Counterbalancing
Fixes a problem for within-subjects design (like testing effect).
Switches the order of exposure of levels
Nonequivalent group design
the control and experimental groups are not necessarily similar because random assignment was not used
Ex: education research
What confound does double-blinding deal with?
Experimenter bias: experimenters might inadvertently treat groups of subjects differently due to their own expectations
Demand Characteristics
any cues that suggest to subjects what the researcher expects from them
Deception might help remedy this
External validity
How generalizable the results of the experiment are to the rest of the population
Types of Descriptive Statistics
Frequency distribution Central tendency: mean, median, mode Variability/standard deviation Distribution/percentiles/z-scores Corrlations
z-score
standardizes across different distributions - how many SD your score is from the mean
** If every score in a distribution is converted to a z-score, the mean of the distribution = 0, and the SD=1.
score - mean
___________
SD
T-score
mean = 50, SD = 10
Test score interpretation
Skewed Distribution
Positive = more lower numbers (mean pushed to the left) Negative = more higher numbers (mean pushed to the right)
Alpha level
usually .05, .01, .001
criterion of significance
The probability of making a Type I error
Type I Error
Incorrectly rejecting the null hypothesis, your research hypothesis is actually wrong
Type II Error
Incorrectly rejecting your research hypothesis when it is actually right and accepting the null