HOME >> HCI User Studies Toolkit

Statistical Decision Tree

The Statistical Decision Tree is a free interactive web application for students, teachers, and researchers who need orientation when choosing a statistical test or model. It is designed for HCI research and adjacent empirical work, especially during study planning, method teaching, and first-pass analysis decisions. The app guides users through key design questions such as outcome type, number and type of predictors, independent versus paired data, repeated measures, and parametric versus nonparametric routes. It then shows suitable methods together with assumption checks, effect-size notes, follow-up recommendations, and example code in R and Python. The tool is meant as orientation support for students and novice researchers. It does not replace statistical training, supervision or study-specific method consultation.

Swipe horizontally to inspect the full decision tree on smaller screens.

Current question

Result

The recommended test or model appears after the last required question has been answered.

This application requires JavaScript to render the interactive decision tree.

What This Statistical Decision Tree Helps With

Interactive statistical test selection

This page helps users choose a statistical test or model for common study designs in HCI, UX research, psychology-adjacent experiments, and other empirical user studies. It supports decisions such as t-tests, ANOVA, repeated-measures ANOVA, linear regression, mixed models, chi-square tests, Fisher's exact test, logistic regression, count models, MANOVA, PERMANOVA, ART models, GEE, and GLMM-related routes.

Designed for study planning and first-pass analysis

The application is intended for students, teachers, and researchers who need a practical entry point for statistical test selection. It combines study-design questions, assumption checks, interpretation notes, effect-size guidance, post-hoc options, and short code examples in R and Python. The emphasis is on method orientation, not automated statistical advice.

Useful for R and Python workflows

The decision tree includes executable example code for many routes, covering core workflows in R and Python. This makes the page useful not only for finding a test, but also for understanding how typical analyses are implemented in teaching, prototyping, and early reproducible-analysis setups.

FAQ

When should I use this tool?

Use it when you need orientation for choosing a statistical test, model, or follow-up procedure from a study design. Typical cases include selecting between parametric and nonparametric tests, deciding between ANOVA and mixed models, comparing repeated measures, or identifying suitable models for binary, count, ordinal, or multivariate outcomes.

Does the page replace statistical consultation?

No. The page is a didactic reference and orientation aid. It is especially helpful for teaching, self-study, and early planning, but it does not replace statistical supervision, method consultation, or design-specific modeling decisions.

Which disciplines is the decision tree for?

The main target domain is HCI and user-study research, but the logic is also useful for many empirical projects in usability, UX, interactive systems, communication, applied psychology, and neighboring behavioral or design research contexts.

Source And Citation

Statistical Decision Tree Repository

Source code on GitHub for the current browser-based Statistical Decision Tree and its interactive decision support logic.

GitHub

Citation And Disclaimer

Prof. Dr. Valentin Schwind. Hochschule der Medien Stuttgart. No liability for external links, correctness, completeness and up-to-dateness of any content. Site visits might result in storing anonymized data (date, time, page viewed). Utilization at the user's own risk. Data can be stored on the computers to facilitate the user's website access. Contribute here.

Find/cite the publication of the toolkit here:


                                Valentin Schwind, Stefan Resch, and Jessica Sehrt. 2023. The HCI User Studies Toolkit: Supporting Study Designing and Planning for Undergraduates and Novice Researchers in Human-Computer Interaction. In Extended Abstracts of the 2023 CHI Conference on Human Factors in Computing Systems (CHI EA '23), April 23-28, 2023, Hamburg, Germany. ACM, New York, NY, USA, 7 pages.
                                https://doi.org/10.1145/3544549.3585890