Chapter 6 Flashcards
Sampling unit
Target population elements actually available to be used during the sampling process
Ex: jury duty
Sampling frame
All eligible sampling units
Ex: list of registered voters, credit card users, etc.
Difference between probability and nonprobability sampling
Probability - known chance of being selected
Nonprobability - no known chance
Advantages of probability sampling
More representative, ability to generalize
Disadvantages of probability sampling
Higher cost, greater expertise needed
Advantages of nonprobability sampling
Target particular groups, inexpensive
Disadvantages of nonprobability sampling
Based on convenience, researcher won’t be able to calculate margin of error, sampling bias
Simple random sampling
Occurs when every sampling unit has a known/equal chance of being selected
ADV: easily understood, generalized, produces unbiased estimates, valid representation
DIS: difficulty of obtaining a complete/accurate listing of the target population elements
Systematic random sampling
Requires the target population be ordered in some way
ADV: frequently used as an easy way to draw a sample with randomness, availability of lists and the shorter time to draw a sample makes this an attractive method
DIS: possibility of hidden patterns in the list of names which creates bias, The number of units must be known, identifying number of units can be difficult
Stratified random sampling
Separates the target population into strata and selects samples from each strata
ADV: assures representativeness, provides and opportunity to make comparisons between strata, allows the ability to make estimates for the target population, greater precision and less error
DIS: determining the basis for stratifying, inclusion of irrelevant strata will waste time and money
Cluster sampling
Divides sampling units into mutually exclusive and collectively exhaustive subpopulations called clusters–units are selected from either random sampling or a census within a defined cluster
ADV: cost-effective, ease of implementation
DIS: clusters are often homogeneous (less precise sample estimates), questionable appropriateness of the designated cluster factor
Convenience sampling
Draws samples at the convenience of the researcher
ADV: enables many respondents to be interviewed in a short time period
DIS: can be risky to develop constructs/scales, data is not generalizable, and representativeness cannot be measured because sampling error estimates cannot be calculated
Judgement (purposive) sampling
Respondents are selected because the researcher believes they meet the requirements of the study
ADV: if judgement is correct, the sample will be better than one generated by convenience
DIS: representativeness cannot be measured
Quota sampling
Involves the selection of prospective participants according the prespecified quotas for: demographic characteristics, specific attitudes, or behaviors
ADV: sample contains subgroups in the proportions desired by researchers and reduces selection bias by field workers
DIS: success of study is dependent on subjective decisions made by researchers, generalization is questionable
Snowball (referral) sampling
A set of respondents are chosen and then they help the researcher identify additional respondents
ADV: identifies respondents who are of small, uniquely, hard-to-reach target populations
DIS: allows bias to enter, generalizability is limited
