Santa Fe Institute Collaboration Platform

COMPLEX TIME: Adaptation, Aging, & Arrow of Time

Get Involved!
Contact: Caitlin Lorraine McShea, Program Manager, cmcshea@santafe.edu

Difference between revisions of "Population and the Environment: Analytical Demography and Applied Population Ethics/Modeling complex populations - statistical inference from demographic data"

From Complex Time
Line 2: Line 2:
 
|Start time=October 16, 2018 09:55:00 AM
 
|Start time=October 16, 2018 09:55:00 AM
 
|End time=October 16, 2018 10:40:00 AM
 
|End time=October 16, 2018 10:40:00 AM
 +
|Is presentation=Yes
 
|Presenter=PaulHooper
 
|Presenter=PaulHooper
 
|Pre-meeting notes=Summary statistics such as the mortality rate, birth rate, or Total Fertility Rate can be useful for understanding some basic characteristics of a population. Often, however, we’re interested in having a more detailed understanding of how demographic events—fertility, mortality, or morbidity—relate to individual characteristics or environmental conditions. One may want to ask questions such as: Does education predict fertility, controlling for wealth? Does infant mortality vary with proximity to clean water, accounting for household-level differences? Does the interval between births depend on a parent’s age, economic strategy, or social network? This session will introduce a number of statistical models that are useful for answering these kinds of questions. We will discuss generalized regression models (customizable to model yes/no outcomes, count data, or continuous variables) as well as survival analysis (also called event history or duration analysis). These models allow us to estimate demographic rates as a function of multiple predictor variables, control for confounding variables, and take into account individual- or group-level heterogeneity.
 
|Pre-meeting notes=Summary statistics such as the mortality rate, birth rate, or Total Fertility Rate can be useful for understanding some basic characteristics of a population. Often, however, we’re interested in having a more detailed understanding of how demographic events—fertility, mortality, or morbidity—relate to individual characteristics or environmental conditions. One may want to ask questions such as: Does education predict fertility, controlling for wealth? Does infant mortality vary with proximity to clean water, accounting for household-level differences? Does the interval between births depend on a parent’s age, economic strategy, or social network? This session will introduce a number of statistical models that are useful for answering these kinds of questions. We will discuss generalized regression models (customizable to model yes/no outcomes, count data, or continuous variables) as well as survival analysis (also called event history or duration analysis). These models allow us to estimate demographic rates as a function of multiple predictor variables, control for confounding variables, and take into account individual- or group-level heterogeneity.

Revision as of 21:51, January 20, 2019

October 16, 2018
9:55 am - 10:40 am

Presenter

Paul Hooper

Abstract

Summary statistics such as the mortality rate, birth rate, or Total Fertility Rate can be useful for understanding some basic characteristics of a population. Often, however, we’re interested in having a more detailed understanding of how demographic events—fertility, mortality, or morbidity—relate to individual characteristics or environmental conditions. One may want to ask questions such as: Does education predict fertility, controlling for wealth? Does infant mortality vary with proximity to clean water, accounting for household-level differences? Does the interval between births depend on a parent’s age, economic strategy, or social network? This session will introduce a number of statistical models that are useful for answering these kinds of questions. We will discuss generalized regression models (customizable to model yes/no outcomes, count data, or continuous variables) as well as survival analysis (also called event history or duration analysis). These models allow us to estimate demographic rates as a function of multiple predictor variables, control for confounding variables, and take into account individual- or group-level heterogeneity.

Presentation file(s)
Download Presentation (Delete)
Related files
First Birth Data For Course.csv (x)

Post-meeting Reflection

Paul Hooper () Link to the source page

It's clear that an extended version of this course should include treatment of inequality (and more generally the distribution of the benefits and costs of environmental impacts within societies) and conflict between and within states. The #1 highlight is of course the group of people assembled here.

Reference Material

The optional reading Kaplan, Hooper, Stieglitz & Gurven (2015) The Causal Relationship between Fertility and Infant Mortality: Prospective analyses of a population in transition provides worked examples of analyzing fertility data (using Cox proportional hazards to model time to next birth) and infant mortality data (using logistic regression).

Access the Emory CASAS Cancer Survival Analysis Suite here: http://bbisr.shinyapps.winship.emory.edu/CASAS/

Title Author name Source name Year Citation count From Scopus. Refreshed every 5 days. Page views Related file
Mortality experience of Tsimane Amerindians of Bolivia: Regional variation and temporal trends Michael Gurven, Hillard Kaplan, Alfredo Zelada Supa American Journal of Human Biology 2007 157 6
The Causal Relationship between Fertility and Infant Mortality: Prospective analyses of a population in transition Hillard Kaplan, Paul L. Hooper, Jonathan Stieglitz, Michael Gurven Population in the Human Sciences: Concepts, Models, Evidence 2015 10 23
Data analysis using regression and multilevel/hierarchical models 0 7
Multilevel Analysis 0 5
On mixed-effect Cox models, sparse matrices, and modeling data from large pedigrees 0 6