によって Matthew Pickard 17年前.
373
Ch2_Sampling
によって Matthew Pickard 17年前.
373
もっと見る
provide data/descriptions
differences
similarities
keep dropout rates low
gradient of similarity
contexts
times
places
people
dimension along which your study context can be related to other potential contexts to which you might wish to generalize
problems
impossible to sample across all times you want to generalize to
not able to draw fair or representative sample
don't know what part of population you want to generalize to
4. generalize results back to the population
3. conduct research on the sample
2. draw representative sample
1. identify population you would like to generalize to
Subtopic
phone book
statistic
sample unit
measurement
response
mean
sampling distribution
standard error
sampling error
the spread of the averages around the average of the averages
standard deviation
3 stdev = 99%
2 stdev =95%
1 stdev = 65%
sampling hard to find populations
sampling for diversity
nonproportional
proportional
sampling experts
sampling most frequent/typical case
may be more precise than simple random sampling
easy to do
5. Take every Kth unit
4. randomly select integer between 1 and K
3. calculate interval size (K = N/n)
2. decide on sample size (n)
1. number units in population
3. measure all units withing sampled clusters
2. randomly sample clusters
1. divide population into clusters (usually geographically)
different
disproportionate stratifed random sample
same
proportionate stratified random sample
more statistical precision (when subgroups are homogenous)
able to represent key subgroups in population, especially minority groups
homogenous subgroup
sample
