Increasing the Lifespan of Software for Demographic Analysis

Egor Kotov, Max Planck Institute for Demographic Research
Headshot of Esther Denecke
Esther Denecke, Max Planck Institute for Demographic Research
Figure 1. Reproducibility struggles. Left: Three laptops showing different software configurations, symbolizing challenges in reproducing results. Right: A single laptop displaying multiple virtual containers with distinct software setups, highlighting improved reproducibility.
Figure 1. Reproducibility struggles. Left: failing to reproduce the exact software setup on multiple computers. Right: running countless software setups on a single laptop with the help of containers. Image source: Microsoft Image Creator powered by DALL-E 3. Generated with AI ∙ 25 October 2023

Many researchers face challenges with computational reproducibility. For instance, running analysis code written just a year earlier can be problematic. Even if it worked flawlessly and gave the expected results earlier, it might fail due to errors now (see Fig. 1). These issues are typically due to the use of newer versions of analysis software…. Read More

Expanding the Lifespan of Software for Demographic Analysis with Containers: An Application of Spatial Sampling

Egor Kotov, Max Planck Institute for Demographic Research
Headshot of Esther Denecke
Esther Denecke, Max Planck Institute for Demographic Research
Figure 1. Reproducibility struggles. Left: Three laptops showing different software configurations, symbolizing challenges in reproducing results. Right: A single laptop displaying multiple virtual containers with distinct software setups, highlighting improved reproducibility.
Figure 1. Reproducibility struggles. Left: failing to reproduce the exact software setup on multiple computers. Right: running countless software setups on a single laptop with the help of containers. Image source: Microsoft Image Creator powered by DALL-E 3. Generated with AI ∙ 25 October 2023

Introduction Software, such as specific R packages, evolve over time, which may prevent older analysis code from working as expected. For example, default values for arguments in a function can change. Therefore, for computational reproducibility, knowing which specific R and package versions were used to run the analysis is crucial. One popular solution in R… Read More

Unlocking Population Estimation Using Readily Available Data

Headshot of Mathew Hauer
Mathew Hauer, Florida State University
Figure 2: Figure 2 shows a graph depicting the correlation between the Censal Ratio Population estimate of 2019 (based on 2010 data) and the actual US Census Bureau’s 2019 population estimate of 2019. The x-axis shows a range of the 2019 censal ratio population estimate, ranging from 0 to 800,000. The y-axis shows a range of the 2019 population estimate and also ranges from 0 to 800,000. The graph demonstrates how the Censal Ratio Method produces a more closely accurate set of estimations than the symptomatic indicator alone. 
Figure 2: Comparison of Census Population Estimates and Censal Ratio Population Estimates across Alabama counties in 2019, where the Censal Ratio Estimates are produced using voter registration data in 2010 and 2019. Source: Alabama Elections Database & Census Bureau.

Population estimation techniques rely on past population data, number of births, deaths, and migration. While various techniques have been used to accurately produce population estimates, the gold standard has been the cohort-component method. However, this method is limited by the fact that some populations may lack the appropriate indicators (e.g., births, deaths, or other changes… Read More

Unlocking Population Estimation Using Readily Available Data: Applying the Simplified Censal Ratio Method

Headshot of Mathew Hauer
Mathew Hauer, Florida State University
Figure 2: Figure 2 shows a graph depicting the correlation between the Censal Ratio Population estimate of 2019 (based on 2010 data) and the actual US Census Bureau’s 2019 population estimate of 2019. The x-axis shows a range of the 2019 censal ratio population estimate, ranging from 0 to 800,000. The y-axis shows a range of the 2019 population estimate and also ranges from 0 to 800,000. The graph demonstrates how the Censal Ratio Method produces a more closely accurate set of estimations than the symptomatic indicator alone. 
Figure 2: Comparison of Census Population Estimates and Censal Ratio Population Estimates across Alabama counties in 2019, where the Censal Ratio Estimates are produced using voter registration data in 2010 and 2019. Source: Alabama Elections Database & Census Bureau.

Population estimation is generally a straightforward process: any population must result from a past population number plus the births minus the deaths plus the net migration. This cohort-component method is often considered the ‘gold standard’ for population estimation (Gerland, 2014).  However, the components of change (births, deaths, migrants) used to forecast a future population are… Read More

Knowing and understanding change: Methods insights using historical pandemic data

Headshot of Taylor van Doren
Taylor van Doren, Sitka Sound Science Center
Two plots of mortality rates for six major causes of death in the United States from 1900-1998: accidents, cancer, heart disease, influenza and pneumonia, stroke, and tuberculosis. The left image is age-adjusted mortality rates for the time period, showing a notable large increase and then decrease of heart disease and other various trends for the other causes. The right image is the log transformations of these mortality rates, showing a notable dramatic decrease in tuberculosis mortality, specifically.
The age-adjusted mortality rates (left) and logarithmic transformation of age-adjusted mortality rates (right) for six major causes of death in the United States from 1900-1998.

Pandemic diseases, like COVID-19, have far-reaching effects that are difficult to identify or predict during the course of the pandemic itself. Case numbers and mortality due to pandemic diseases ebb and flow, but even after time has passed, it is useful to know (1) when the waves occurred, (2) if the pandemic disease had effects… Read More

Estimating time points of significant change in cause-specific mortality: Joinpoint regression in R

Headshot of Taylor van Doren
Taylor van Doren, Sitka Sound Science Center
Two plots of mortality rates for six major causes of death in the United States from 1900-1998: accidents, cancer, heart disease, influenza and pneumonia, stroke, and tuberculosis. The left image is age-adjusted mortality rates for the time period, showing a notable large increase and then decrease of heart disease and other various trends for the other causes. The right image is the log transformations of these mortality rates, showing a notable dramatic decrease in tuberculosis mortality, specifically.
The age-adjusted mortality rates (left) and logarithmic transformation of age-adjusted mortality rates (right) for six major causes of death in the United States from 1900-1998.

INTRODUCTION Those engaged in demographic research are often interested in how and why the vital demographic processes (fertility, mortality, and migration) change in response to certain ecological, cultural, or behavioral stimuli. Today, in the midst of a global pandemic event, epidemiologists and demographers may be interested in the ability to identify points over time during… Read More