Field CH1 & CH2 Flashcards
what is a variable?
anything that can be measured & can differ across entities or across time
predictor variable = independent
outcome variable = dependent
binary level of measurement
two distinct types of things
nominal level of measurement
more than 2 types of things; if a number, the number has no mathematic meaning
ordinal level of measurement
more than 2 types of things; the categories have logical order
interval
equal intervals of scale must represent equal differences in properties beign measured
ratio
equal intervals of scale must represent equal differences in properties being measured + the ratios must make sense
validity
whether an instrument measures what it is supposed to measure
reliability
whether an instrument can be interpreted consistently across different situations
criterion validity
whether you can establish that an instrument measures what it claims to, through a comparison of objective criteria
concurrent-validity
when data are recorded simultaneously using the new instrument & exiting criteria
predictive validity
when data from new instruments is used to predict future observations
content validitiy
the degree to which individual items represent the construct being measured
correlational/cross-sectional research
the degree to which individual items represent the construct being measured
- provides natural view as no researcher interference
- tells us nothing about the causal influences of variables
experimental research
one variable is manipulated to see its effects on another; both cause & outcome are variables
confounding variables (confounds)
extranous factors; a variable (that may or may not have been measured) other than the predictor variable
situations
treatment/conditions
unsystematic variation
small differences in performance due to unknown factors
skewness
the lack of symmetry
systematic variation
differences in performance created by specific experimental manipulation
kurtosis
pointyness (or lack thereof)
central tendency
where the center of a frequency distribution lies
model fit
the degree to which a statistical model represents teh data collected
parameters
estimated from the data (unlike variables) & are usually constnts believed to represent some fundamental truth about the variables & model
