Module 0 Flashcards
What is a Census and why is not usually collected
A census is a SAMPLE that includes EVERYONE (the entire population)
-it is time consuming
-expensive
under-coverage (may not actually include everyone)
what are “sample statistics”
summaries found from data in a sample
what are the two types of “inferences” that you can make with sample statistics
(just name the two)
- Population Inference
- Casual Inference
What is a Population Inference
where the results from a. sample can be generalized to an ENTIRE population
What is a Casual Inference
making conclusions from the results of two different variables or treatment groups
what is Random Sampling
randomly selecting individuals in samples from the population (non-random leads to raised results)
What is Simple random sampling
randomly picking people, the population has the same chance of being selected
What is Stratified random sampling
the sample is first divided into different similar groups called STRATA
then you take a simple random sample fro each stata
what is Cluster random sample
splitting the population into similar groups (clusters) then selecting a few clusters at random as samples
what is Systematic random sample
selecting people of a population at a regular interval
ex.every 10th person that walks by
what is Selection Bias (undercoverage)
when a portion of the population has a smaller representation in the sample or isn’t even sampled at all. They usually differ from the rest of the population
ex. an opinion poll conducted by telephone will miss out the households without home phones
What is Response Bias
anything in a survey design that influences the responses
ex. respondents may lie if they don’t want to admit anything
Voluntary Response Bias
when individuals can choose to participate in the sample
What is Nonresponse Bias
When a large portion of those sampled fail to respond
What is Random Allocation and what inference can you make with this
where you choose individuals for treatment groups by applying a random mechanism
ex. flipping a coin
causal inference