Björn Siepe
@bsiepe.bsky.social
PhD Student in Psychological Methods (University of Marburg)
Interested in time series, simulation studies & open science
https://bsiepe.github.io
Interested in time series, simulation studies & open science
https://bsiepe.github.io
Further details:
▶️Search and filter datasets on openesmdata.org
▶️Auto-generate R/Python code
▶️(Meta-)data are stored on Zenodo with DOIs
▶️Metadata and software on GitHub enable community contributions
▶️Contribution guidelines allow further database extensions so that openESM can continue to grow
▶️Search and filter datasets on openesmdata.org
▶️Auto-generate R/Python code
▶️(Meta-)data are stored on Zenodo with DOIs
▶️Metadata and software on GitHub enable community contributions
▶️Contribution guidelines allow further database extensions so that openESM can continue to grow
October 22, 2025 at 7:34 PM
Further details:
▶️Search and filter datasets on openesmdata.org
▶️Auto-generate R/Python code
▶️(Meta-)data are stored on Zenodo with DOIs
▶️Metadata and software on GitHub enable community contributions
▶️Contribution guidelines allow further database extensions so that openESM can continue to grow
▶️Search and filter datasets on openesmdata.org
▶️Auto-generate R/Python code
▶️(Meta-)data are stored on Zenodo with DOIs
▶️Metadata and software on GitHub enable community contributions
▶️Contribution guidelines allow further database extensions so that openESM can continue to grow
▶️Our example analysis shows how to use openESM: We estimated within-person correlations between positive and negative affect across 39 datasets (>500K observations)
▶️We find a robust negative correlation (−0.49 [-0.54, -0.42]) and outline ideas for future research building on this
▶️We find a robust negative correlation (−0.49 [-0.54, -0.42]) and outline ideas for future research building on this
October 22, 2025 at 7:34 PM
▶️Our example analysis shows how to use openESM: We estimated within-person correlations between positive and negative affect across 39 datasets (>500K observations)
▶️We find a robust negative correlation (−0.49 [-0.54, -0.42]) and outline ideas for future research building on this
▶️We find a robust negative correlation (−0.49 [-0.54, -0.42]) and outline ideas for future research building on this
▶️In our introduction paper, we outline why large-scale analyses are important for substantive, design, and statistical research
▶️To make such research easier, we provide rich metadata for each dataset, plus dedicated R and Python packages to easily access and handle the data
▶️To make such research easier, we provide rich metadata for each dataset, plus dedicated R and Python packages to easily access and handle the data
October 22, 2025 at 7:34 PM
▶️In our introduction paper, we outline why large-scale analyses are important for substantive, design, and statistical research
▶️To make such research easier, we provide rich metadata for each dataset, plus dedicated R and Python packages to easily access and handle the data
▶️To make such research easier, we provide rich metadata for each dataset, plus dedicated R and Python packages to easily access and handle the data
We built the openESM database:
▶️60 openly available experience sampling datasets (16K+ participants, 740K+ obs.) in one place
▶️Harmonized (meta-)data, fully open-source software
▶️Filter & search all data, simply download via R/Python
Find out more:
🌐 openesmdata.org
📝 doi.org/10.31234/osf...
▶️60 openly available experience sampling datasets (16K+ participants, 740K+ obs.) in one place
▶️Harmonized (meta-)data, fully open-source software
▶️Filter & search all data, simply download via R/Python
Find out more:
🌐 openesmdata.org
📝 doi.org/10.31234/osf...
October 22, 2025 at 7:34 PM
We built the openESM database:
▶️60 openly available experience sampling datasets (16K+ participants, 740K+ obs.) in one place
▶️Harmonized (meta-)data, fully open-source software
▶️Filter & search all data, simply download via R/Python
Find out more:
🌐 openesmdata.org
📝 doi.org/10.31234/osf...
▶️60 openly available experience sampling datasets (16K+ participants, 740K+ obs.) in one place
▶️Harmonized (meta-)data, fully open-source software
▶️Filter & search all data, simply download via R/Python
Find out more:
🌐 openesmdata.org
📝 doi.org/10.31234/osf...
We highlight some advantages of a Bayesian multilevel approach:
▶️We can estimate the probability of a node being the most central node
▶️For some individuals, there may be a clearly "most central" node, whereas this is very unclear for others
▶️This information is crucial for practical applications
▶️We can estimate the probability of a node being the most central node
▶️For some individuals, there may be a clearly "most central" node, whereas this is very unclear for others
▶️This information is crucial for practical applications
April 28, 2025 at 8:31 AM
We highlight some advantages of a Bayesian multilevel approach:
▶️We can estimate the probability of a node being the most central node
▶️For some individuals, there may be a clearly "most central" node, whereas this is very unclear for others
▶️This information is crucial for practical applications
▶️We can estimate the probability of a node being the most central node
▶️For some individuals, there may be a clearly "most central" node, whereas this is very unclear for others
▶️This information is crucial for practical applications
Simulation study w/ different estimation methods shows:
▶️Treatment selection: Identifying the most central node is often difficult
▶️Outcome prediction: Using centrality for outcome prediction requires large n & t & large effects
▶️Using a one-step model such as BmlVAR can help under some conditions
▶️Treatment selection: Identifying the most central node is often difficult
▶️Outcome prediction: Using centrality for outcome prediction requires large n & t & large effects
▶️Using a one-step model such as BmlVAR can help under some conditions
April 28, 2025 at 8:31 AM
Simulation study w/ different estimation methods shows:
▶️Treatment selection: Identifying the most central node is often difficult
▶️Outcome prediction: Using centrality for outcome prediction requires large n & t & large effects
▶️Using a one-step model such as BmlVAR can help under some conditions
▶️Treatment selection: Identifying the most central node is often difficult
▶️Outcome prediction: Using centrality for outcome prediction requires large n & t & large effects
▶️Using a one-step model such as BmlVAR can help under some conditions
▶️ We introduce a Bayesian approach (BmlVAR) that estimates a multilevel VAR model and between-person network feature regression in one step
▶️ This accounts for uncertainty that is usually ignored when first estimating individual networks and then using centrality point estimates in a second step
▶️ This accounts for uncertainty that is usually ignored when first estimating individual networks and then using centrality point estimates in a second step
April 28, 2025 at 8:31 AM
▶️ We introduce a Bayesian approach (BmlVAR) that estimates a multilevel VAR model and between-person network feature regression in one step
▶️ This accounts for uncertainty that is usually ignored when first estimating individual networks and then using centrality point estimates in a second step
▶️ This accounts for uncertainty that is usually ignored when first estimating individual networks and then using centrality point estimates in a second step
Can we use features of dynamic networks (e.g. centrality) to improve treatment selection and outcome prediction?
New preprint on the topic: We highlight the role of uncertainty & introduce a Bayesian multilevel approach for uncertainty quantification of network features 🧵
osf.io/preprints/ps...
New preprint on the topic: We highlight the role of uncertainty & introduce a Bayesian multilevel approach for uncertainty quantification of network features 🧵
osf.io/preprints/ps...
April 28, 2025 at 8:31 AM
Can we use features of dynamic networks (e.g. centrality) to improve treatment selection and outcome prediction?
New preprint on the topic: We highlight the role of uncertainty & introduce a Bayesian multilevel approach for uncertainty quantification of network features 🧵
osf.io/preprints/ps...
New preprint on the topic: We highlight the role of uncertainty & introduce a Bayesian multilevel approach for uncertainty quantification of network features 🧵
osf.io/preprints/ps...
With all the new people on here (hi!), a delayed summary of our paper on Bayesian estimation and comparison of n=1 gVAR/network models, recently published in PsychMethods: How can we understand whether differences in these popular, but highly parameterized models reflect more than just noise? 🧵
October 29, 2024 at 12:30 PM
With all the new people on here (hi!), a delayed summary of our paper on Bayesian estimation and comparison of n=1 gVAR/network models, recently published in PsychMethods: How can we understand whether differences in these popular, but highly parameterized models reflect more than just noise? 🧵
We then classify different types of missingness based on the stages when it occurs and highlight them on examples from our review.
We overview the different approaches to handling missingness and their pros and cons, and illustrate them in a case study.
We overview the different approaches to handling missingness and their pros and cons, and illustrate them in a case study.
October 1, 2024 at 6:56 AM
We then classify different types of missingness based on the stages when it occurs and highlight them on examples from our review.
We overview the different approaches to handling missingness and their pros and cons, and illustrate them in a case study.
We overview the different approaches to handling missingness and their pros and cons, and illustrate them in a case study.
We reviewed almost 500 simulation studies across 4 journals and explored how missingness is reported. Most articles do not discuss missingness, and overall reporting could be improved. Also, the code-sharing practices are suboptimal
October 1, 2024 at 6:54 AM
We reviewed almost 500 simulation studies across 4 journals and explored how missingness is reported. Most articles do not discuss missingness, and overall reporting could be improved. Also, the code-sharing practices are suboptimal
Have you run #simulation studies and encountered issues with missing results, failure, or non-convergence? In a new preprint by S. Pawel, F. Bartoš, me & A. Lohmann, we overview different approaches to address these issues and review current practices.
arxiv.org/abs/2409.18527 #stats
arxiv.org/abs/2409.18527 #stats
October 1, 2024 at 6:52 AM
Have you run #simulation studies and encountered issues with missing results, failure, or non-convergence? In a new preprint by S. Pawel, F. Bartoš, me & A. Lohmann, we overview different approaches to address these issues and review current practices.
arxiv.org/abs/2409.18527 #stats
arxiv.org/abs/2409.18527 #stats
Haha, amazing! Here's a screenshot of a slide (I don't think I ever showed it in the end) from a 2022 presentation about the project
July 5, 2024 at 11:36 AM
Haha, amazing! Here's a screenshot of a slide (I don't think I ever showed it in the end) from a 2022 presentation about the project
New preprint w/ the WARN-D team (incl. @eikofried.bsky.social @rayyantutunji.bsky.social, @aljoscharimpler.bsky.social & others): We explain our exploration of EMA items in data of ~600 individuals. We investigate distributions/changes over time/context/interindividual differences & more
January 26, 2024 at 10:22 AM
New preprint w/ the WARN-D team (incl. @eikofried.bsky.social @rayyantutunji.bsky.social, @aljoscharimpler.bsky.social & others): We explain our exploration of EMA items in data of ~600 individuals. We investigate distributions/changes over time/context/interindividual differences & more
We believe that the time is ripe for a shift in methodological research toward more rigor and transparency in simulation studies and that our ADEMP-PreReg template can help to achieve this goal. /end
October 31, 2023 at 2:49 PM
We believe that the time is ripe for a shift in methodological research toward more rigor and transparency in simulation studies and that our ADEMP-PreReg template can help to achieve this goal. /end
We provide Monte Carlo standard error calculations and sample size planning formulas for the most common performance measures and showcase the ADEMP methodology in an example simulation study. /5
October 31, 2023 at 2:48 PM
We provide Monte Carlo standard error calculations and sample size planning formulas for the most common performance measures and showcase the ADEMP methodology in an example simulation study. /5
We also summarize general recommendations to improve simulation methodology. /4
October 31, 2023 at 2:48 PM
We also summarize general recommendations to improve simulation methodology. /4
The template helps with design & reporting of sim studies by prompting questions about Aims, Design, Estimands, Performance Measures, Reporting, and Computational Aspects. As such, it's helpful for both sim study novices seeking guidance & experts wanting a standardized blueprint /3
October 31, 2023 at 2:47 PM
The template helps with design & reporting of sim studies by prompting questions about Aims, Design, Estimands, Performance Measures, Reporting, and Computational Aspects. As such, it's helpful for both sim study novices seeking guidance & experts wanting a standardized blueprint /3
We overview the ADEMP framework from Morris et al. (2019) adapted to simulation studies in psychology. In our review, we find that the majority of studies do not justify the sample size, do not report Monte Carlo uncertainty, and do not provide code or computational details. /2
October 31, 2023 at 2:46 PM
We overview the ADEMP framework from Morris et al. (2019) adapted to simulation studies in psychology. In our review, we find that the majority of studies do not justify the sample size, do not report Monte Carlo uncertainty, and do not provide code or computational details. /2
We reviewed 100 psych. simulation studies & find room for improvement in planning/reporting. As a remedy, we (František Bartoš, @timpmorris.bsky.social Anne-Laure Boulesteix, @danielheck.bsky.social & Samuel Pawel) present ADEMP-PreReg, a simulation study preregistration & reporting template 🧵/1
October 31, 2023 at 2:46 PM
We reviewed 100 psych. simulation studies & find room for improvement in planning/reporting. As a remedy, we (František Bartoš, @timpmorris.bsky.social Anne-Laure Boulesteix, @danielheck.bsky.social & Samuel Pawel) present ADEMP-PreReg, a simulation study preregistration & reporting template 🧵/1