maths stats pros and cons Flashcards
simple random
every member of the population is likely to be chosen
systematic sampling
find sample n from population of size N by taking one of the first K members of the population at random the selecting every K the member after that where k= N/n
Stratifying
when you want distinct groups to be represented in your sample , split the population into these distinct groups and then sample within each in proportion to its size
opportunity
use population that you have access to
quota
when you have distinct groups to be represented in your sample decide how many members of each group you wish to sample in advance and use opportunity sampling until you have large enough sample
cluster
split population into clusters that you expect to be similar to each other then take a sample from each of these clusters
census
dis-
time-consuming
cannot be used when testing process destroys data
a large quantity of data
‘ad
accurate
sample
disa
not as accurate
not as large sample to make a reliable statement
ad
less time consuming
systematic pros and cons
quick and easy
suitable for large sample
sampling frame is needed
biases can arise if frames not used
stratified pros and cons
reflects population structure
guarantees proportional
must be two distinct groups
selection of each stratum can have the same problem as random
opportunity
easy ‘
cheap
unlikely to represent actual population
highly dependant
census
every member in the population
sample
subset of population
sample frame
subset of population
bias
does not represent population
