1.5IQR rule
Click the card to flip 👆
1 / 26
Terms in this set (26)
undercoveragemiss a group because you cannot reach themnon-responseindividuals don't respond by choicebiaspoor questions/leading questionslurking variableseffects the explanatory and response variableconfounding variableeffects just the response variable, but is related to the explanatoryexperiments must have these three elements1. control 2. randomization 3. replicationsblockingreducing the effects of known attributes of the subjects that cannot be changed (stratified random sampling)matched pairs designeach person is a block and gets both treatments (accounts for individual variability)P(A∪B)P(A∪B) = P(A) + P(B) *** if mutually exclusive, otherwise subtract overlapP(A∩B)P(A)*P(B) *** if independentP(A|B)P(both)/P(B)events are independent when...P(A|B) = P(A)binomial mean and standard deviationµ = np, σ = √(npq)geometric mean and standard deviationµ = 1/p, σ = √(q/p^2)normal approximation can be used whennp>10 & nq>10varianceσ^2