Josef Brüderl, Ansgar Hudde, Marita Jacob
Sociological Science December 4, 2025
10.15195/v12.a34
Abstract
In life course research, it is common practice to analyze the effects of life events on outcomes. This is usually done by estimating “impact functions.” To date, most studies have estimated yearly impact functions. However, Hudde and Jacob (2023) (hereafter H&J) pointed out that most panel data sets include information on the month of events. Consequently, they proposed exploiting this information by estimating monthly impact functions. In this adversarial collaboration, we address two issues regarding H&J’s work. First, H&J did not provide sufficient guidance on how to estimate monthly impact functions. We will provide a step-by-step description of how to do so. Second, the procedure H&J proposed for smoothing monthly estimates produces confidence intervals (CIs) that are likely too narrow. This can lead to misleading conclusions. Therefore, we suggest using more appropriate bootstrapped CIs.
In life course research, it is common practice to analyze the effects of life events on outcomes. This is usually done by estimating “impact functions.” To date, most studies have estimated yearly impact functions. However, Hudde and Jacob (2023) (hereafter H&J) pointed out that most panel data sets include information on the month of events. Consequently, they proposed exploiting this information by estimating monthly impact functions. In this adversarial collaboration, we address two issues regarding H&J’s work. First, H&J did not provide sufficient guidance on how to estimate monthly impact functions. We will provide a step-by-step description of how to do so. Second, the procedure H&J proposed for smoothing monthly estimates produces confidence intervals (CIs) that are likely too narrow. This can lead to misleading conclusions. Therefore, we suggest using more appropriate bootstrapped CIs.
 | This work is licensed under a Creative Commons Attribution 4.0 International License. |
Reproducibility Package: Stata replication code is available on the Open Science Framework (OSF), https://osf.io/kx9ne/ (file: “Monthly Impact Functions-Replication File.zip”). The replication file includes the prepared pairfam data that we used for all of our analyses. If you would like to reproduce our data preparation (also included in the replication file), you can order the pairfam data at https://www.pairfam.de/en/data/data-access/.
- Citation: Brüderl, Josef, Ansgar Hudde, Marita Jacob. 2025. “What You Need to Know When Estimating Monthly Impact Functions: Comment on Hudde and Jacob, “There’s More in the Data!”” Sociological Science 12: 862-870.
- Received: May 16, 2025
- Accepted: August 31, 2025
- Editors: Arnout van de Rijt, Kristian B. Karlson
- DOI: 10.15195/v12.a34
No reactions yet.