Population
The entire set of individuals or other entities to which study findings are to be generalized
Sample
A subset of a population used to study the population
Sampling units
the units of measurement on which information is being collected
Sample Generalizability
Can findings based on a sample be applied to the larger population from which the sample was drawn?
Cross-population generalizability
Can findings from a study of one population be applied to other populations?
Target population
the population to which the researcher would like to generalize study findings
Census
information obtained about all available members of a population
Representative sample
a sample that closely resembles the population from which it was selected
Sampling Error
Sampling error is the difference between the results we find in our sample and the results in the actual population
Sampling Methods
Probability sampling methods
Nonprobability Sampling Methods
Probability sampling methods
rely on random or chance selection
we know the odds of sampling someone/something from population
random sampling, cluster sampling, stratified random sampling
Nonprobability Sampling Methods
- probability that someone/something will be selected is unknown
convenience sampling, quota sampling, purposive sampling, snowball sampling
Simple Random Sampling
Cases are only selected on the basis of change
Systematic Random Sampling
method in which every nth element is randomly selected from a list
Cluster Sampling
samples are selected in two or more stages
Stage 1: random sample of naturally occurring clusters or groups
Stage 2: random sample of elements within each selected cluster is drawn
Probability Sampling Methods
Stratified random sampling- to ensure that various groups are included in your sample
Nonprobability Sampling Methods
Quota Sampling
Purposive Sampling
Quota Sampling
make sure sampling represents certain characteristics in the population
not random and cannot know all elements
Purposive Sampling
Studying a subset of a limited group
if you want to know only about parents of children with Down Syndrome or some other uncommon condition
Studying "key informants" who are particularly knowledgeable about an issue
but still try to get a range of viewpoi
Hidden populations
sensitive nature of behaviors
Difficult to distinguish
Snowball sampling
participants are selected and then asked to identify other participants
Useful for hard to reach or hard to identify interconnected populations
Targeted Sampling
using outreach techniques to attract a sample of people in hidden population
Time-space sampling
Figure out when and where hidden population members gather and randomly sample those places/times
Respondent driven sampling
select seeds for initial data collection
each seed gets a certain number of coupons to give to other members of the population in interest
3 requirements for identifying causal effects
1. Empirical Association between variable
2. Time order of the variables
3. spuriousness/nonspuriousness of the association