Quantitative Data Science Methods – Psychometrics, Econometrics and Machine Learning (QDS)

Overview

This research-oriented master’s programme is designed to deepen your methodological and technical expertise in quantitative data science, with a clear emphasis on methods development and scientific work. It prepares graduates to conduct independent research, contribute to advances in the field, and continue into PhD studies. The curriculum also equips students for responsible leadership roles by fostering independent judgment, critical thinking, and decision-making in research contexts. Teaching is closely linked to active research — the programme benefits from strong collaboration between research institutes inside and outside the university and covers state-of-the-art applications.

The degree is broad and interdisciplinary, centring on three core domains — psychometrics, econometrics and machine learning — while allowing students to specialise further in areas of interest. You will not only learn to apply established techniques but also to critically evaluate their assumptions and limitations and to develop new methods. The programme is delivered in English and combines coursework, ethics training, and a substantial research project to bridge theoretical knowledge and practical, research-led experience.

Core components and requirements

• Foundations

  • General statistical and technical modules to level up differing background knowledge.
  • Personalized module combinations (e.g., statistics/probability theory or programming) are available; completion is recommended within the first two semesters.

• Psychometrics

  • Methods include (semi)parametric latent variable modelling, item modelling, dynamic longitudinal models, Bayesian statistics, knowledge space theory, and decision-making models.
  • Emphasis on identifying problematic assumptions and understanding method limitations.

• Econometrics

  • Introduction to quantitative econometric techniques with flexibility in topics.
  • Methods such as time series analysis and machine learning taught for applications like microeconometrics and financial markets.

• Machine Learning

  • Core concepts such as data literacy, deep learning, and statistical/probabilistic machine learning.

• Data Ethics

  • Seminars and supplementary events to develop the ability to reflect on the ethical and responsible use of data in decision-making and applications.

• Project Seminar

  • An individual (or group) research project that deepens theoretical and practical knowledge in any of the core disciplines.
  • Project topics can align with optional specialisations and may be carried out in collaboration with the university’s research groups.

This four-semester, interdisciplinary Master's curriculum is organised into four thematic areas: an interdisciplinary Foundations component (QDS‑FO) plus three core domains—Psychometrics and Mathematical Psychology (QDS‑PS), Econometrics (QDS‑EC) and Machine Learning (QDS‑ML). The programme requires students to earn a minimum number of ECTS credits in each of the three core domains across three semesters to preserve its interdisciplinary character while allowing focused depth in one area.

Students can choose to specialise in one of the three core areas. Specialisation is staged through taught modules, a project seminar, and the Master’s thesis: the taught modules for a chosen specialisation can comprise up to 30 ECTS, the project seminar may develop a research topic aligned with that specialisation, and the Master’s thesis offers further opportunity for in‑depth work in the chosen domain. Alternatively, a broader, interdisciplinary route is available (for example, a distribution of about 21 ECTS in each core area) with interdisciplinary topics for the project and thesis.

Overall, the structure is designed so graduates acquire solid interdisciplinary methodological foundations (QDS‑FO) and the option to develop advanced competencies and research experience in psychometrics, econometrics or machine learning through coursework and supervised research components.

Requirements at a glance

• Programme length: 4 semesters
• Core structure: QDS‑FO (Foundations) + three core areas QDS‑PS, QDS‑EC, QDS‑ML
• Interdisciplinarity rule: a minimum number of ECTS must be earned in each of the three core areas, distributed across three semesters
• Specialisation path (optional): taught modules covering up to 30 ECTS in one core area
• Research project (project seminar): can be based on the chosen specialisation
• Master’s thesis: can further specialise in one core area or take an interdisciplinary approach
• Broad path alternative: e.g., ~21 ECTS in each core area with interdisciplinary project and thesis

For full details and downloadable documents, see the Methods Center website: https://uni-tuebingen.de/de/160267

Graduates are prepared for research-focused careers and for progression to PhD programmes thanks to the programme's emphasis on methods development and independent scientific work. The degree is suitable for roles that require advanced quantitative skills, such as methodological research positions in academia and research institutes.

Beyond academia, the strong methodological training (in psychometrics, econometrics and machine learning), practical project experience and ethical reflection equip graduates for specialist data-science, analytics and modelling roles in industry, finance, public sector organisations and applied research centres.