Class Notes (1,100,000)
US (440,000)
UC-Irvine (10,000)
PUBHLTH (600)
Lecture 9

PUBHLTH 101 Lecture Notes - Lecture 9: Linear Regression, Fallacy, Cohort Study


Department
Public Health
Course Code
PUBHLTH 101
Professor
David Timberlake
Lecture
9

This preview shows half of the first page. to view the full 2 pages of the document.
LEC 9
ECOLOGICAL STUDIES
An ecologic or aggregate study focuses on the comparison of groups rather than
individuals
A study in which units of analysis are populations or groups of people than individuals
Aggregated measures
Summary of individual-level variable
o The data may be possible to measure in individuals
o Ex: mean of dietary intake of meat in a country
o Mean consumption of alcohol in a city
o Median household income
o Incidence of a disease; proportion (%) w/ high diploma etc.
Types of Variables in Ecological Studies
o Group level measures: some data may be convenient/easy to measure in
individuals
Ex) Env. Data such as air population level in country/background gamma
radiation in different areas
o Global measures (contextual): some global measures for attributes of groups or
places
NO individual data / analog is available
Statistical Analysis for Ecological Studies
Linear regression or other regression models
Correlation analysis
Types of Ecological Studies:
o
Completely ecological studies:
All variables (Exposed, outcome, covariates) are
ecological, so unit of analysis of the group
o
Partially ecological studies:
some variables might be in groups but some of them
might be in individuals
o
Multi-level studies:
combines data collected at more levels
Types of study designs for ecological studies:
o
Time-trend studies:
comparison of diseases rates over time in populations
You're Reading a Preview

Unlock to view full version