SAMPLING AND THE SOURCES OF BIAS

SAMPLING

explanatory analysis of the sample

generalize and conclude about sample

making an INFERENCE

for inference to be valid

sample needs to be REPRESENTATIVE of entire population

SOURCES OF SAMPLING BIAS

CONVENIENCE SAMPLE

NON-RESPONSE

VOLUNTARY RESPONSE BIAS

individuals who are easily accessible

they are more likely to be included in the sample

if only a (non-random) fraction of the random sample people respond to a survey such that the sample is no longer representative of the populatiom

e.g. certain groups of people with social status are less likely to respond to the survey

when the sample consists of people who volunteer to respond because they have strong opinions on the issue

difference between voluntary response bias and non-response bias

non-response

there is a random sample that is surveyed, but people who choose to respond are not representative of the sample

voluntary response

there is no initial random sample

SAMPLING METHODS

STRATIFIED SAMPLING

SIMPLE RANDOM SAMPLING (SRS)

CLUSTER SAMPLING

MULTISTAGE SAMPLING

randomly select cases from population, such that each case is equally likely to be selected

first divide the population into homogenous groups

STRATA

then randomly sample from within each startum

e.g. randomly drawing names from a hat

want to make sure both genders are equally represented in a study

divide the population first into males and females, and then randomly sample from within each group

divide the population into CLUSTERS, randomly sample a few clusters, and then sample all observation within these clusters

clusters are heterogeneous within themselves (x strata and stratified sampling)

each cluster is similar to another, such that we can get away with just sampling from a few of the clusters

add another step to cluster sampling

we divide the population into clusters, randomly sample a few clusters, and then we randomly sample observations from within these clusters