sampling—process of selecting observations

quota sampling—based on knowledge of population characteristics being sampled (led to failure in 1948 presidential election prediction)

heterogeneity or variability in population requires sampling

sample is "representative" if the aggregate characteristics of sample approximate same characteristics of population

--requires that all members have equal chance of being selected

--more representative of population than other types of sampling

--probability theory allows estimates of the accuracy or representativeness of sample

element—unit about which information is collected

universe—theoretical and hypothetical aggregation of all elments

population—theoretically specified aggregation of survey elements

survey population—aggregation of elements from which the survey is actually selected

sampling unit—element or set of elements considered for selection in some stage of sampling

sampling frame—actual list of sampling units from which the sample is selected

--sampling frames often define the survey population

variable—set of attributes that contain variation

parameter—summary description of a given variable in a population

statistic—summary description of a given variable in a survey sample

sampling error—

confidence levels—

sampling distribution—distribution of all potential samples that can be drawn from a population

**sample will be distributed around the population parameter in a known way

standard error—indicates the extent to which sample estimates will be distributed around the population parameter (similar to standard deviation)

--can be translated into probabilities that sample mean falls with certain distance from true population mean

--is a function of the population parameter and the sample size—as sample size increases, standard error decreases

Simple random sample—assign a number to each element in list—then go to random number table and select numbers

Systematic sample—every kth element is chosen for inclusion—should start at random place

--sampling interval—distance between elements selected

Stratified sampling—divide population into homogenous subsets and select the appropriate number from each subset

Cluster sampling—used when impossible or impracticle to comple an exhaustive list of the elements composing the target population

--sample from a list first, then sample within the items chosen from the list

--efficient but introduces another layer of sampling error

--goal is to maximize the number of clusters selected while decreasing the number of elements within each cluster

Disproportionate Weight Sampling—sample subpopulations disproportionately to insure sufficient numbers of cases from each

nonprobability sampling

--purposive or judgemental sampling

--quota sampling