This text was posted on Demotrends by Maarten Bijlsma on September 1, 2015. The original can be found here.

The fields of epidemiology and demography are closely aligned. Even demographers interested in fertility or migration, not just mortality, can learn a great deal from epidemiology. As a recent study has argued, epidemiology is currently undergoing a methodological revolution, and this is likely to affect demography as well. The epidemiological revolution is, in fact, a causal inference revolution. In this post, I describe G-computation, a technique which is used by scientists employing a causal inference approach. G-computation allows users to infer population-level effects from individual-level effect estimates, and can therefore be of great value to demographers.

Continue reading →

This text was posted on Demotrends by Maarten Bijlsma on September 4, 2013. The original can be found here.

Last year, Yang & Land released their book on age-period-cohort (APC) analysis. A large section of the December issue of Demography is dedicated to age-period-cohort analysis. And just a week or two ago a paper by Bell & Jones on the Yang & Land APC model was published in Demographic Research. APC analysis appears to be a hot topic these days!

Continue reading →

This text was posted on Demotrends by Maarten Bijlsma on September 4, 2013. The original can be found here.

Most researchers are familiar with the difference between bias and precision. However, not everyone knows that we can allow for a little bit of bias in order to get big gains in precision, and when it can be beneficial to do so.  In this post I detail the why and how.

Continue reading →

This text was posted on Demotrends by Maarten Bijlsma on April 23, 2013. The original can be found here.

The Lexis diagram belongs in the demographer’s toolbox like a mechanic’s wrench. It is the essential tool for visualizing lifelines and events as they occur along the intersection of age, (time) period and (birth) cohort. This means it is of great use to scientists doing longitudinal, time-to-event and even time series analyses. As such, it can also be of use to sociologists, epidemiologists and statisticians, but many experts in those disciplines seem unaware of it! Unfortunately, demographers may be to blame.

Continue reading →