mktg 465 final 06 sampling Flashcards
any complete group with a specific characteristic
population
complete group of individuals or things from whom information is needed to meet the objectives of the study
marketing research population
a list of elements of the population from which the sample may be drawn
sample frame
inclusion/exclusion of a subset of individuals without/with unique characteristics of interest
sampling frame error
single element or groups of elements subject to selection in the sample
sample units
if budget is small (sample or census)
sample
if sampling errors are low (sample or census)
sample
if population variance is large (sample or census)
census
if population size is large (sample or census)
sample
if the nature of measurement is nondestructive (sample or census)
census
if the time available is long (sample or census)
census
if sampling errors are high (sample or census)
census
if budget is large (sample or census)
census
if time available is short (sample or census)
sample
if population variance is small(sample or census)
sample
if population size is small (sample or census)
census
if the nature of measurement is destructive (sample or census)
sample
samples in which every element of the population has a known, non-zero probability of selection
probability sampling
what are the advantages of probability sampling?
- sample error can be computed
- the results can be projected to the whole population
what are the disadvantages of probability sampling?
expense and time required to design and execute
what are the 4 probability sampling methods?
- simple random sampling
- systematic sampling
- stratified sampling
- cluster sampling
random sampling from mutually exclusive and exhaustive subgroups
stratified sampling
what are the problems with stratified sampling?
- information to stratify not readily available
- proportional/disproportional sampling
random selection from all elements in the population
simple random sampling
what are the issues with simple random sampling
- construction of a representative sampling frame
- disproportionate subgroups
every nth element (skip interval) from the list following a random starting point
systematic sampling
what is the issue with systematic sampling?
- covers the entire sample frame
- possible bias from hidden patterns
random selection of a mutually exclusive group or cluster of elements
cluster sampling
what is the only method without names involved
cluster sampling
selection of specific elements from the population in a non-random manor
non-probability sampling
what are the 2 advantages to non-probability sampling?
- cost
- time
what are the 2 disadvantages to non-probability sampling?
- sampling error cannot be computed
- the degree to which the sample is representative of the population is unknown
what are the 4 types of non-probability sampling methods?
convenience sampling, judgement sampling, quota sampling, snowball sampling
samples in which selection of additional respondents is based on referrals from the initial respondents
snowball sampling
what are the issues with snowball sampling?
low-incidence or rare population, subject to the willingness of respondents to provide referrals
samples in which quotas are established for mutually exclusive population subgroups
quota sampling
what is the issue with quota sampling?
favors those most easily found
sampling the people that are most available
convenience sampling
what is the issue with convenience sampling?
potential for selection bias
selection criteria based on judgement that the element is representative of the population under study
judgement sampling
what is the issue with judgement sampling?
subject to the personal judgement of the researcher
the formula for sample size for a proportion? when looking for how many people or need to be surveyed
n= (z^2(pq))/e^2
for sample size for a proportion, what is n?
sample size
for sample size for a proportion, what is z?
confidence level, always 1.96
for sample size for a proportion, what is e?
maximum allowance for error
for sample size for a proportion, what is pq?
estimated proportion with specified characteristic, always (0.5*0.5)
what is the formula for adjusting based on sample size?
nā= nN/(N+n-1)
what is nā for population and sample size
revised sample size
what is n for population and sample size
original sample size
what is N for population and sample size
population size
