Dominik Vollbracht

Statistical modeling · Measurement and surveys · Automated analysis and reporting pipelines M.Sc. in Psychology · Doctoral thesis submitted, defense in February 2027 · Available from April 2027

Karlsruhe, Germany dominik@vollbracht.email github.com/donvollb linkedin.com/in/dominik-vollbracht

CV as PDF Deutsche Fassung

Profile

In my doctoral thesis on the statistical modeling of survey data, I studied how response formats shape the data. I used multilevel models for longitudinal data, dynamic structural equation models, and multidimensional item response models, and I extended the Beta Item Response Model to detect response styles (preprint and R package). Alongside my doctorate, I completed the qualification program of the DFG Research Training Group Statistical Modeling in Psychology (SMiP). At RPTU Kaiserslautern-Landau, I also redesigned the course evaluation system on the Landau campus (around 600 courses per semester), from automated registration to reproducible analyses and reports in R and Quarto. It is now in routine use.

Professional Experience

Oct 2021 – present

Research Associate RPTU Kaiserslautern-Landau (University of Koblenz-Landau until 2022)

Research Methods and Evaluation Department of Psychology · Research, teaching and methodological consulting

  • Doctoral thesis: Comparing Visual Analogue and Likert-Type Scales: Psychometric Properties and Response Styles.
    • Designed and ran an experience sampling study with random assignment to the two formats and analyzed it with multilevel and dynamic structural equation models (MSEM, DSEM)
    • Measurement invariance held across formats and reliability was largely comparable (published in Behavior Research Methods)
    • Extended the Beta Item Response Model to account for response styles (BIRM-RS), tested it in a simulation study (Stan, HPC cluster), and applied it to empirical data (preprint, under review)
    • Compared response styles across formats (BIRM-RS for visual analogue (slider) scales, multidimensional partial credit model for Likert-type scales): extreme responding largely generalized across formats (preprint, under review)
  • Methodological consulting:
    • Advised research projects of other groups, within and beyond RPTU
    • Co-authored several resulting papers, including in clinical and cognitive psychology
  • International collaborations: University of Basel; Centre for Educational Measurement (CEMO), University of Oslo
  • Teaching and supervision:
    • Taught courses in the bachelor’s and master’s programs
    • Supervised 13 theses (4 as primary supervisor, 9 as second supervisor)
    • Coordinated student assistants

Oct 2021 – Dec 2025

Research Associate, Course Evaluation (DigiKompASS Project) RPTU Kaiserslautern-Landau · Methodological Center (Zentrum für Methoden, Diagnostik und Evaluation)

University-wide project on digital teaching. Led the redesign of the course evaluation system on the Landau campus, a campus-wide feedback survey with reports for lecturers and departments (around 600 courses and 6,000–8,000 responses per semester), from concept to routine use.

  • Concept and questionnaire:
    • Conducted a needs analysis among students
    • Extended the core questionnaire as a common benchmark across departments, with optional, purely formative modules for different teaching formats that keep results comparable
    • Tested the factor structure and measurement invariance across disciplines
  • Rollout: piloted in stages and rolled out to all departments, in coordination with the data protection officer and the department committees
  • Automation:
    • Built a web scraper in R to collect course data from the campus management system, which has no export
    • Generated personalized LimeSurvey registrations with validation and automatic confirmations
    • Cut registration from several workdays to a few hours per semester
    • Lecturers can now tailor the survey to their course
  • Analysis and reporting:
    • Rebuilt the analysis and reporting pipeline as the tested R package setanalysis (see Software)
    • Aggregated reports for departments and, for cohort surveys, sometimes more than 100 tailored reports per round
    • In routine use and now maintained by colleagues
  • LLM analysis of open-ended responses: guided the development of a pipeline with open-source language models, piloted it and checked it against two manual content analyses
  • External clients:
    • Occasional paid evaluations commissioned by companies, some using the LLM pipeline
    • Coordinated and delivered the reports
  • Further evaluations: evaluated events such as a summer school; advised other subprojects on how to evaluate their work

Mar 2025 – Apr 2025

Visiting Researcher University of Basel · Statistics & Data Science, Faculty of Psychology (Prof. Mirka Henninger)

Collaborated on response style models; gave a colloquium talk

Dec 2019 – Sep 2021

Student Research Assistant University of Koblenz-Landau · Methodological Center

  • Revised the R scripts for analyses and reports of course evaluations and university-wide cohort surveys (the basis of the later R package setanalysis)
  • Supported the registration process, including the import into EvaSys
  • Set up account management and provided user support for the university’s LimeSurvey platform
  • Evaluated a career development program against its stated goals largely independently and wrote the evaluation report

Jun 2017 – Feb 2018

Student Research Assistant University of Koblenz-Landau · Department of Psychology

Ran research studies; analyzed video in Blender (face tracking for emotion recognition)

Selected Software

R package

setanalysis · with set-template

Turns survey exports into many personalized reports from a single Quarto template (EvaSys import, one function per question type, rule-based control of report content). In routine use at RPTU; snapshot tests, pkgdown, CI with GitHub Actions. Lead developer; a student assistant made substantial contributions under my guidance

R package

birmrssim

Parameter recovery simulations for the BIRM-RS: four Stan models, parallel execution, citable via doi.org/10.5281/zenodo.20305322

Web apps

Interactive model visualizations

Shiny apps that run in the browser via shinylive, without a server, and explain new models to non-specialists:

Skills

Statistical modeling

  • Psychometrics questionnaire development, exploratory and confirmatory factor analysis, IRT (including bounded continuous data, e.g., from visual analogue scales), and measurement invariance: checking that scores mean the same across groups, departments, or response formats before comparing them
  • Structural equation modeling path analysis, latent growth curve and change models, latent state and latent state-trait models
  • Multilevel models and multilevel SEM (MSEM) nested data, e.g., employees in teams or repeated measures
  • Intensive longitudinal data ambulatory assessment (experience sampling), autoregressive models and dynamic SEM (DSEM) for processes in daily life
  • Bayesian statistics hierarchical models in Stan and JAGS
  • Experimental design and power analysis experiments and simulation studies on the robustness of methods
  • Latent class and profile analysis e.g., segmenting respondents by response patterns
  • Regression models linear and generalized linear models, including logistic and multinomial regression

Further methods

Basics of machine learning, multinomial processing tree (MPT) models, and stochastic process models

Programming

R – package development with tests (testthat), documentation (pkgdown), and CI (GitHub Actions)

Git and GitHub Shiny and shinylive SQL and Python basics (pandas, NumPy)

Reporting

Quarto (parameterized reports, custom templates and extensions, presentations, websites, and reproducible manuscripts, including my doctoral thesis), R Markdown, Typst, LaTeX, HTML, and CSS

Tools

HPC cluster (Slurm) Stan JAGS Mplus LimeSurvey (incl. API) EvaSys

AI and LLMs

Open-source LLMs via Ollama (local and server-based) in analysis pipelines, validation of LLM output against human judgments · AI-assisted development with Claude Code

Languages

German (native) English (fluent)

Education

expected 2027

PhD in Psychology RPTU Kaiserslautern-Landau · Advisor: Prof. Tanja Lischetzke

  • Doctoral thesis on psychometric properties and response styles of Likert-type and visual analogue scales
  • Submitted, defense in February 2027

2021

Master of Science in Psychology, with honors University of Koblenz-Landau

Master’s thesis on measurement invariance of course evaluations across disciplines

2020

Bachelor of Science in Psychology University of Koblenz-Landau

2013 – 2014

Year-long internship at a special education school Karl-Preising-Schule, Bad Arolsen

2013

Abitur (German university entrance qualification) Christian-Rauch-Schule, Bad Arolsen

Additional Training

2023 – 2026

Qualification program of the DFG Research Training Group Statistical Modeling in Psychology Universities of Freiburg, Heidelberg, Mannheim, Tübingen, and RPTU · 51 program days

Courses on advanced statistical modeling, machine learning, scientific Python, simulation studies, LLMs as a research tool, and data visualization

Teaching

Courses

  • Multivariate Methods (bachelor’s)
  • Advanced Multivariate Methods (master’s)
  • Evaluation: Intensive Longitudinal Data (master’s)
  • Psychometrics (master’s)
  • Redesigned the first three with colleagues

Workshops

As an external lecturer for doctoral students and postdocs: multilevel models for longitudinal data (University of Duisburg-Essen, two days; Humboldt-Universität zu Berlin, one day) · multilevel models and multilevel SEM (Humboldt-Universität zu Berlin)

Other

  • Tutor for the empirical research practicum (bachelor’s, weekly sessions)
  • Mentor for first-year students
  • Tutoring in math, chemistry, biology, and English, Studienkreis Landau (2015–2018)

Awards

2022 – 2025

Five teaching awards, Department of Psychology, RPTU, based on student course evaluations

2024

Young Researcher Award, Multilevel Conference, Utrecht

2023, 2025

Two Open Science Awards in Gold, Department of Psychology, RPTU, for preregistration, open data, and reproducible analyses

Appendix: Publications and Talks

ORCID: 0000-0003-1040-7851

Peer-Reviewed Journal Articles

  • Vollbracht, D., Ottenstein, C., Ecker, S., & Lischetzke, T. (2026). Slider versus Likert scales: Psychometric properties in ambulatory assessment. Behavior Research Methods, 58(4), 97. doi.org/10.3758/s13428-026-02992-4
  • Ottenstein, C., Ecker, S., Vollbracht, D., & Lischetzke, T. (2026). Multiverse of data sets regarding indirect measures of emotional clarity: An examination of the degrees of freedom in analyzing response times to emotion items. Collabra: Psychology, 12(1), 162291. doi.org/10.1525/collabra.162291
  • Ecker, S., Ottenstein, C., Vollbracht, D., & Lischetzke, T. (2026). Does the procedure matter? Applying a multiverse analysis approach to negative emotion differentiation. Collabra: Psychology, 12(1), 159939. doi.org/10.1525/collabra.159939
  • Knorr, A., Wegmann, E., Vollbracht, D., Brand, M., & Antons, S. (2026). Predicting moments of impaired control over addictive behaviors: Relevance of craving and inhibitory control measured in laboratory and ambulatory settings. Journal of Behavioral Addictions. doi.org/10.1556/2006.2025.00102
  • Schmid, A. M., Thomas, T. A., Oelker, A., Vollbracht, D., Meurer, N., Mallon, L., Diers, M., Dominick, N., Kampa, M., Wolf, O. T., Brand, M., Müller, A., & Steins-Loeber, S. (2026). Daily stress and gaming – an ambulatory assessment study in individuals with pathological, risky, and non-problematic gaming behavior. Acta Psychologica, 263, 106321. doi.org/10.1016/j.actpsy.2026.106321
  • Thomas, T. A., Schmid, A. M., Vollbracht, D., Oelker, A., Wolf, O. T., Brand, M., Steins-Loeber, S., & Müller, A. (2026). Pavlovian-to-instrumental transfer effect after devaluation as indicator of habitual behavior together with stress responsivity predicts daily use of buying-shopping platforms. Acta Psychologica, 264, 106453. doi.org/10.1016/j.actpsy.2026.106453

Preprints (Under Review)

  • Vollbracht, D., Müller, M., Lischetzke, T., & Henninger, M. (2026). Extreme response style generalizes across Likert-type and visual analogue scales. PsyArXiv. doi.org/10.31234/osf.io/zevb7_v1
  • Vollbracht, D., Lischetzke, T., & Henninger, M. (2026). Detecting response styles in visual analogue scales. PsyArXiv. doi.org/10.31234/osf.io/xteku_v2
  • Ulitzsch, E., Vollbracht, D., Zhang, L., Jover Martínez, A., Domingue, B., Haslbeck, J. M. B., Lischetzke, T., & Vogelsmeier, L. V. D. E. (2026). Likert versus visual analogue scales in ecological momentary assessment: Initial evidence that Likert scales come with reduced capacity to sustain attentiveness. PsyArXiv. doi.org/10.31234/osf.io/p8fg6_v1

Conference Presentations (First Author)

  • Vollbracht, D., Lischetzke, T., & Henninger, M. (2025, November). Detecting response styles in slider scales [Talk]. SMiP Retreat Fall 2025, Ludwigshafen, Germany.
  • Vollbracht, D., Lischetzke, T., & Henninger, M. (2025, September). Detecting response styles in slider scales [Talk]. Conference of the DGPs Methods and Evaluation Section 2025, Berlin, Germany.
  • Vollbracht, D., Ottenstein, C., Ecker, S., & Lischetzke, T. (2025, May). Slider vs. Likert scales: Psychometric properties in ambulatory assessment [Talk]. Society for Ambulatory Assessment Conference 2025, Leuven, Belgium.
  • Vollbracht, D., Ottenstein, C., Ecker, S., & Lischetzke, T. (2024, September). Slider scales vs. radio buttons: A comparison of psychometric properties in experience sampling methods [Talk]. 53rd DGPs Congress / 15th ÖGP Conference, Vienna, Austria.
  • Vollbracht, D., Ottenstein, C., Ecker, S., & Lischetzke, T. (2024, March). Slider scales vs. radio buttons: A comparison of psychometric properties in experience sampling methods [Talk]. Multilevel Conference 2024 (Young Researcher Award), Utrecht, The Netherlands.
  • Vollbracht, D., Ottenstein, C., Ecker, S., & Lischetzke, T. (2023, June). Slider scales or radio buttons? Comparing the psychometric properties of slider scales with traditional radio button scales in ambulatory assessment [Poster]. Society for Ambulatory Assessment Conference 2023, Amsterdam, The Netherlands.
  • Vollbracht, D., Ottenstein, C., Ecker, S., & Lischetzke, T. (2023, June). Slider scales or radio buttons? Comparing the psychometric properties of slider scales with traditional radio button scales in ambulatory assessment [Poster]. SMiP Summer School 2023, Mannheim, Germany.

Peer Review

Ad hoc reviewer for Collabra: Psychology