z-score – how many standard deviations away from the mean normal model – unimodal and symmetric empirical rules 1SD = x%, 2SD = x%, 3SD = x% - 68%, 95%, 99.5% form – layout/underlying pattern relationship – strong means follows pattern, weak means scattered veriable of interest and explanatory/predictor variable – var of int = y var/response var, explan var = x var correlation coefficient – a measure of the strength of relationship between two variables lurking variable – a variable that is not among the explanatory or response variables in a study but that may influence the response variable parameters – statistics from a whole population statistics – mead/med/mode for a sample line of best fit – the line for which the sum of the squared residuals is smallest good regressions model means – when plot residuals (x vs y or predicted y vs y) there is no shape equal variance assumption – a good model will have the spread of the residuals consistent throuhgout and no shape high leverage – a pts x-val is far from the mean x influential point if – omitting it changes the line of best fit's slope a lot outlier – pts y val is far from mean y population and sample means/SDs symbols – pop: μ and σ sample: x bar, S sub x bar trial – each time an experiment is tried biased sample – A sample that does not fairly represent the population (over or runder reps part of pop) stratified sampling – pop divided into two homogeneous groups before sampling cluster sampling – split pop into groups that ea fully rep the entire pop before sampling systematic sampling – select some starting point and then select every nth element in the population nonresponse bias – bias introduced to a sample when a large fraction of those sampled fails to respond conveniance sampling bias – choosing easy subjects (ex: first 20 ppl you see) response bias – anything in a survey design that influences responses (intimidation, wording of q, etc) voluntary response is a – bias observational study – observes individuals and measures variables of interest but does not attempt to influence the responses experiment – random assignment of treatments to subjects that establishes a cause and effct relationship between the vars