September 14, 2026 | News | Overview

Recommended Reading

Here is an overview of the latest reading recommendations for papers published by scientists at the MPIDR.

Comparison of Microsimulated and Register-based Kinship Data

Paper first published: July 30, 2026

Two-part infographic with stacked area charts showing the average number of different types of kin by birth cohort, top from microsimulation, bottom from Swedish registers.

In their recent paper, Liliana P. Calderón-Bernal and colleagues examine how closely estimates of kinship based on microsimulation match those derived from Swedish population registers. The researchers compared empirical estimates of kinship for a large number of kinship types, derived from an earlier register-based study for Sweden by co-author Martin Kolk, with estimates of similar kin quantities derived from a simulated population of Swedish individuals. These were generated using the SOCSIM microsimulation program.

‘We ran a demographic microsimulation of the Swedish population from 1751 to 2017 to obtain a complete register of all individuals ever alive during the simulated period, including their vital events and kinship relationships,' Calderón-Bernal explains.

Through the same approach used by Kolk et al. in the 'Swedish Kinship Universe, the team estimated the synthetic kinship networks. Using the SOCSIM results, they then traced the kinship networks (children, parents, siblings, grandchildren, grandparents, aunts and uncles, nieces and nephews, and cousins) of the synthetic individuals alive at the end of 2017 based on parent-child links and compared them with the register-based figures.

'The figures for relatives derived from the SOCSIM microsimulation and the registers are largely similar, particularly with regard to the average number of living relatives. However, the microsimulation yields slightly lower values for most types of kin in cohorts not affected by intergenerational parent-child truncation in the registers,’ says Calderón-Bernal.

But microsimulation can fill in gaps: Despite the undercount and under-dispersion in the microsimulation for recent cohorts, a fully tracked synthetic population better accounts for some relatives of early cohorts likely omitted from the registers due to missing parent–child information, survival bias, or truncation of kinship ties due to migration.

Keywords

Original Publication

Calderón-Bernal, L. P.; Alburez-Gutierrez, D.; Kolk, M.; Zagheni, E.:

Assessing the agreement between microsimulated and register-based kin counts in Sweden

Population Studies, 1–20. (2026)

Family Complexity Influences Health in Mid-Adulthood

Paper first published: September 17, 2025

A child watches worried parents arguing in a brightly lit room.

Researchers from the Max Planck Institute for Demographic Research (MPIDR), the University of Helsinki, Tilburg University, the Centre for Research on Social Inequalities (CRIS), the University of Oxford and the Einstein Center for Population Diversity (ECPD) demonstrated for the first time that complex family life histories impact physical and mental health in middle adulthood. Notable gender-specific differences emerged: Women's physical health deteriorates significantly following a separation involving children, whereas men's mental and physical health are more adversely affected by any type of separation.

The study used data from the UK's "Understanding Society" survey to analyze the family histories and health of 3,407 women and 2,638 men aged 55. The researchers examined the participants’ family histories between the ages of 18 and 55, including their relationship status and whether they had become parents. By measuring this 'family complexity' — that is, the number and unpredictability of family transitions, such as separations, changes in partners, changes in relationship status or becoming a single parent — the researchers were able to show that the cumulative stress of family changes has long-term health consequences.

A key finding is that: Particularly for women mental health reflects current family status, while physical health depends heavily on the long-term accumulation of family stressors. “Our results show that health inequalities arise not only from current family status, but also from the entire life course,” explains Philipp Dierker.

Original Publication

Dierker, P., M. Kühn and Z. Van Winkle (2026): "The Role of Family Complexity in Mental and Physical Health in Mid-Adulthood." Journal of Marriage and Family (JMF) DOI: 10.1111/jomf.70019

Keywords

Analysing Biases in Genealogies

Paper published: December 5, 2025

Diagrams show age-dependent fertility and mortality rates from 1900-1905 with simulation results.

Researchers at the Max Planck Institute for Demographic Research (MPIDR) investigated the accuracy of data from ascending genealogies for estimating fertility and mortality rates. Liliana P. Calderón-Bernal et al. run SOCSIM demographic microsimulation, using data from the Human Fertility Collection (HFC), the Human Fertility Database (HFD) and the Human Mortality Database (HMD), all for Sweden from 1751 to 2022, to simulate complete records of individuals including their kinship relationships. Based on these simulations, the researchers reconstructed virtual family trees and tested how three common biases in genealogies affect the estimates.

Certain families are missing altogether, because the lineage got extinct.
Only direct ancestors are considered, which results in missing data on collateral kin, particularly regarding children and young adults who died early.
Certain individuals are missing because they had no descendants or are omitted from the records.

The results:

When only direct ancestors are considered, the average number of children is significantly underestimated (by about 42%), and average life expectancy is significantly overestimated (by about 33%).
When the descendants of ancestors are included, fertility estimates become higher, and life expectancy estimates improve — the overestimation drops to about 1.8%.

Although genealogical data from online databases is useful and easily accessible, it is incomplete. A comparison of "fully recorded" and "distorted" family trees derived from a simulation shows that retrospective reconstruction and the type of ancestors included in a genealogy can lead to misestimation of the number of children people had and their lifespan. Including more relatives in the family tree makes the information more accurate.

This type of analysis is relevant given the emergence of new data sources resulting from the digitization and crowdsourcing of historical records, such as large online genealogy databases. Although these databases show promise for historical and kinship demographic research, they suffer from coverage and representativeness issues.

Original Publication

Liliana P. Calderón-Bernal, Diego Alburez-Gutierrez, Emilio Zagheni: Analysing Biases in Genealogies Using Demographic Microsimulation in European Journal of Population (2025); DOI: 10.1007/s10680-025-09756-4

Keywords

Genealogies, Microsimulation, Biases, Historical demography, Kinship