This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Public Health 6450 – Fall 2011 Andrew Mugglin and Lynn Eberly Division of Biostatistics School of Public Health University of Minnesota [email protected] Part 04 Review Overview of the Study Cycle Study Design Where are we going? Previously: • Probability distributions • Quantitative Discrete (Binomial, Multinomial) • Quantitative Continuous (Normal) Current topics: • Overview of the study cycle • Specifics: sampling and study design Mugglin and Eberly PubH 6450 Fall 2011 Part 04 2 / 48 Review Overview of the Study Cycle Study Design The study cycle: Focus on the corners Mugglin and Eberly PubH 6450 Fall 2011 Part 04 3 / 48 Review Overview of the Study Cycle Study Design Another View of the Study Cycle Tru e State o f N atu re Pa ra m e te rs: μ , σ 2 , β , ρ , θ , p , … O b se rvatio n s S a m p le S ta tistics: Mugglin and Eberly PubH 6450 Fall 2011 Part 04 4 / 48 Review Overview of the Study Cycle Study Design The corners: Population → sample → statistics → parameters Definitions • A population is ‘all members of a class.’ We measure some characteristic (call it X ) within the population of interest. • A sample is a subset of this population in which we measure the characteristic, x 1 ,..., x n . Mugglin and Eberly PubH 6450 Fall 2011 Part 04 5 / 48 Review Overview of the Study Cycle Study Design The corners: Population → sample → statistics → parameters Definitions • A population is ‘all members of a class.’ We measure some characteristic (call it X ) within the population of interest. • A sample is a subset of this population in which we measure the characteristic, x 1 ,..., x n . • Statistics (sometimes called sample statistics) measure properties of this characteristic’s distribution in the sample, e.g., sample mean ¯ x , sample variance s 2 . • Parameters represent properties of this characteristic’s distribution in the population (population mean of X , population variance in X ). Mugglin and Eberly PubH 6450 Fall 2011 Part 04 5 / 48 Review Overview of the Study Cycle Study Design Simple example • Population : Women with primary progressive multiple sclerosis, about whom we want to characterize their Kurtzke Disability Status Scale (a continuous scale measure of disability), X • Sample : n women ages 3555 years who are being treated for primary progressive multiple sclerosis at the University’s Multiple Sclerosis Center, for whom we measured their disability score, x 1 ,..., x n . • Statistics : sample mean disability score ¯ x , sample variance in disability score s 2 . • Parameters : population mean disability score ( E [ X ], E for expected disability score), population variance in disability score ( Var [ X ])....
View
Full
Document
 Fall '10
 AndyMugglin

Click to edit the document details