STA 13 Lecture Notes - Lecture 2: Simple Random Sample, Stratified Sampling, Cluster Sampling
Document Summary
Sampling methods being in the sample (other books define this differently) Random sampling: every member of the population has the same chance of. Simple random sample (srs): random sample plus subjects are sampled. Convenience sampling: get sample any old way that is easy. Stratified sampling: divide into groups (strata) with similar characteristics. Ex: 5,9,15 - sample all in those rows. Don"t manipulate conditions or try to influence response. 2 variables could be correlated, but the correlation isn"t the same as causation. Collect data on moms who smoke/don"t smoke. Find lower birth weight for mothers who smoke. This is called confounding - lurking variables . Confounding is when the presence, absence, or magnitude of a causal relationship between 2 variables is obscured (blocked) by 1 or more other variables. Other variables are called confounding variables ( lurking ) In general we want to know if a to b. A = lives in fl or ak.
