Class Notes (1,100,000)
US (490,000)
UMD (9,000)
GVPT (200)
Matis (4)
Lecture 2

GVPT 100 Lecture 2: Sampling 101


Department
Government and Politics
Course Code
GVPT 100
Professor
Matis
Lecture
2

This preview shows half of the first page. to view the full 1 pages of the document.
Sampling 101: Why is it so important?
Extra Credit: Find a newspaper article that talks about politics but makes one of the errors
we talked about and write 2-3 pages about the error.
Sampling
oExpensive or impossible to gather data for the whole population
In these cases we gather a sample of the population
Why Sample
oCollecting information on each population member is:
Time consuming
Expensive
Logistically difficult
oSample data can be better:
Quick to capture mobile populations
Sample and Population
oTwo key concepts
Population- the universe of subjects the researcher wants to describe
Sample- a number of cases, subjects or observations drawn from a
population to conduct research
oPopulation parameter – the actual value in the population
Example: the percentage of Americans who approve of President Obama
oSample Statistic – the estimate of the population parameter
Example: the percentage of Americans in the Gallup Poll sample which
approve of President Obama
oInferential Statistics – a set of procedures for deciding how closely a relationship
we observe in a sample corresponds to the unobserved relationship in the
population from which the sample was drawn (Pollack 2012, pg.122)
Key Terms:
oRandom sample: every member of the population has an equal chance of being
chosen for the sample
oSampling frame: method for defining the population the research wants to study
oSelection bias (systematic error): some members of the population are more likely
to be included in the sample than others
oResponse Bias (systematic error): some members of the population are more
likely to respond than others
Error and Sample
oSelection/sampling bias
oResponse bias – some individuals are more likely to be measured than others
oRandom sampling error – the extent to which a sample statistic differs by chance
from the population parameter
Other methods than pure random sampling:
oStratified, cluster, and purposive sampling
You're Reading a Preview

Unlock to view full version