A complete statistical analysis for your thesis
in minutes
With eduSTAT, you’ll get a structured analysis,
appropriate statistical methods, and a
preformulated statistics section for your thesis.
Please note: app currently onlyavailable in German language!
Please note: the app is currently only available in German!
This FFG-funded statistical software is part of a comprehensive project to improve the quality of scientific research.
The implemented methods were validated using test power analysis. About the validations.
Many students’ nightmare when writing their thesis: statistics
Selecting tests, checking assumptions, interpreting p-values. Just one methodological error is enough, and your paper gets rejected.
With eduSTAT, you get a structured analysis of your data, access to appropriate statistical methods, and a scientifically formulated methodology and results section fora successful thesis.
You are currently viewing a placeholder content from YouTube. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
More Information
Are the statistics for your thesis stressing you out?
Your survey is ready, or you’re still working on your research proposal, but you know that without solid methodology, your project won’t be approved. When you get to the point of selecting test subjects, defining eligibility criteria, or determining sample size, however, what started out as a good idea can quickly turn into a methodological obstacle course.
If you choose the wrong test,
assumptions are violated,
or you don't fully document your results...
…then the consequences might be:
Criticism of your methodology section
Required corrections
A lower grade
That’s why many students try to get by using tutorials, online forums, or AI and LLMs to cram statistics at the last minute. But especially when dealing with sensitive data or more complex concepts, it can quickly lead to misunderstandings. Your professor rarely has time for in-depth one-on-one guidance. So you have no choice but to learn to use statistics software…
In short: Most existing statistics applications are powerful computational tools.
For your thesis, however, you need a tool that guides you through your analysis in a structured way and helps you provide correct scientific documentation.
That’s exactly what you get with eduSTAT.
Please note: app currently onlyavailable in German language!
eduSTAT: Scientific statistical methodology support for your thesis
Scientific presentation of results
The results are not only presented as statistical figures, but also explained in scientific language, described methodologically, and embedded in a structured results section, including charts and figure captions.
Intuitive workflow from start to finished result
From importing your data to the completed analysis, only a few clicks (approximately 3–15 steps) are usually required to reach a structured final result.
Automatic selection of suitable statistical tests
Based on your variables, the software automatically selects appropriate methods, such as the Mann–Whitney test, G-test, Welch’s test, or suitable correlation analyses.
Analysis even for nonsignificant results
If your results are not statistically significant, the analysis does not simply stop. The software examines possible influencing factors and confounding variables, helping you understand why the outcome turned out the way it did.
Transparent justification for test selection
eduSTAT explains why a particular method was selected, enabling you to understand and justify the decision from a scientific perspective.
Local processing in your browser
All calculations are performed directly on your device. No data is transferred to external servers.
Automatic integration of references to relevant literature
Suitable references are provided and correctly cited for the methods, tests, and procedures used.
Direct processing of your excel data
Upload your spreadsheet and start your analysis immediately.
's Scientific Analysis of the Results
Intuitive operation all the way to a finished result
Automatic selection of appropriate
statistical tests
Analysis Even When Results Are Not Significant
Clear rationale
for the test election
Local Processing of "
" in the Browser
Automatic inclusion of relevant bibliographic references fr
Direct Processing of Your
Excel Data
After the assessment, you’ll receive…
…visually presented results,
including automatically generated graphics, charts, and tables
Bar chart
Pie chart
Histogram
Scatter plots
Box-and-whisker plots
Contingency tables
…technical explanations of your results
…a structured section on methodology and results, including references
...everything you need for the statistics section
of your thesis.
This is what the completed statistics
section of your thesis looks like
Here is an example of a fully generated Word document, including:
-
Charts
-
Methodology
-
References
-
Presentation of results
This lets you see exactly how your analysis will be structured and presented in scientific language.
Download the sample statistics section as a Word document
Here’s how it works: Four steps to
your completed analysis
Import your data
Upload your data file (e.g. Excel or CSV) directly into the application.
Select and define variables
Select the variables you want to analyse. If necessary, you can define groups or comparison variables. eduSTAT4 automatically recognises the structure of your data and determines the most suitable statistical methods.
Run the analysis automatically
The software performs the appropriate statistical tests, examines relevant relationships, and generates statistical parameters, visualisations, methodological explanations, and a structured results section.
Download your results
Export your complete analysis as a Word document, including methodology, results tables, charts, figure captions, and references. You can integrate the results directly into your thesis and continue editing them as needed.
Try it today with no obligation
Try eduSTAT with your own dataset and see for yourself how well-organised your statistical analysis can be.
Please note: app currently onlyavailable in German language!
About the developer
Dr. rer. nat. Rudolf Golubich
Developer and provider
About the developer
As a physicist with a Ph.D. (Vienna University of Technology) specialising in theoretical particle physics, I am experienced in highly complex data analysis.
In addition to my research, I have been working for many years as a freelance data scientist, developer, and tutor in statistics and data analysis. During that time, I have advised numerous students on their quantitative theses.
Time and again, I have seen that statistical analysis is a major challenge for many people. With eduSTAT, I want to help you tackle this challenge!
eduSTAT is based on a process of clearly defined statistical analysis. Unlike generative AI, eduSTAT operates on the basis of a deterministic, rule-based system so it can deliver reproducible results.
The development of eduSTAT was funded by the Österreichische Forschungsförderungsgesellschaft (FFG).
The goal of this funding is to implement an innovative, methodologically sound approach to improve the quality of scholarly work.
Development in Austria
eduSTAT was developed entirely in Austria.
Dr. rer. nat. Rudolf Golubich, a developer and researcher based in Austria, was responsible for the conceptual design, scientific development, and technical implementation.
Please note: app currently onlyavailable in German language!
Frequently asked questions about eduSTAT
eduSTAT is a browser-based statistics program developed specifically for students and researchers. The application automates statistical analysis – from data processing and the selection of appropriate tests to the scientifically structured presentation of results. All calculations are performed locally in the browser. No data is transferred to a server. eduSTAT helps users conduct quantitative analyses that are structured, transparent, and methodologically sound.
The statistical calculations and mathematical operations make use of the math.js library. This library enables precise numerical calculations and the implementation of statistical functions directly in the browser. The accuracy of the calculations was validated by comparing them with Wolfram Mathematica, both through target/actual comparisons and through power analyses: → See the validations and test data
G-test
Likelihood-ratio test for nominal data
- p-value (significance test)
- G-statistic
- Minimum sample size for significance
Mann-Whitney U test
Nonparametric test for independent samples
- p-value (significance test)
- U-statistic and z-value
- Hodges-Lehmann estimator with 95% confidence interval
- Biserial rank correlation (effect size)
- Minimum sample size for test power
Wilcoxon signed-rank test
Nonparametric test for paired samples
- p-value (significance test)
- Test statistics (tp-tm)
- Hodges-Lehmann-estimator with 95% confidence interval
- Biserial rank correlation (effect size)
- Minimum sample size (1.4826 × MAD/HL)
Welch’s test
T-test for unequal variances
- p-value (significance test)
- t-statistic with adjusted degrees of freedom
- Group difference with 95% confidence interval
- Minimum sample size for test power
- Minimum sample size for significance
Pearson correlation
Product-moment correlation for linear relationships
- p-value (significance test)
- Correlation coefficient with 95% confidence interval
- t-statistic
- Effect size classification
- Minimum sample size for significance
Spearman’s correlation
Rank correlation for monotonic relationships
- p-value (significance test)
- Correlation coefficient with 95% confidence interval
- t-statistic
- Effect size classification
- Minimum sample size for significance
Anderson-Darling test
Normality test for small to medium samples
- p-value (significance test)
- Anderson-Darling statistic
- Test power
- Minimum sample size for 80% test power
To process Excel and CSV files (among other file types), eduSTAT uses Sheet.js Community Edition, a powerful JavaScript library for spreadsheets. This enables fully client-side data processing without the need to transmit data to a server. Exporting to Word documents is based on docx.js, which enables the generation of docx documents directly in the browser.
The visualisations are created using the community version of Plotly.js, an open-source library for statistics diagrams. This library enables the creation of high-quality charts directly in the browser without relying on a server.
Yes. eduSTAT processes all data locally in the browser. No data is transferred to external servers or stored in the cloud. The data remains entirely on the user’s device.
No. eduSTAT automates the execution and documentation of statistical analyses. Users remain responsible for the content of their scientific work. The software supports structured implementation, but it does not replace an understanding of methodology.
Yes. The underlying methodology has been published in a scientific journal:
This publication can be cited for methodology.
eduSTAT relies on established open-source libraries. Among others, it uses:
•plotly.js – Data visualisation under the MIT License• SheetJS/jszip – Document processing under the Apache License 2.0• math.js – Mathematical calculations under the Apache License 2.0• docx.js – Word document generation under the MIT License• FileSaver.js – File export under the MIT License
eduSTAT was developed with great care, and the methods implemented were validated against reference systems and through power analyses. The calculations are based on established statistical methods. See the validations.
eduSTAT assists with the structured execution and documentation of the analysis. However, as with any software, a paper’s scientific evaluation cannot be 100% guaranteed.
Users retain responsibility for the content of the research question and its interpretation.
No, extensive prior knowledge of statistics is not required. eduSTAT guides you through the analysis in a structured way and explains why certain tests are used. Of course, you should have a basic understanding of your research question. However, the statistical process is clearly explained.
Yes. eduSTAT was developed specifically for theses using small to medium-sized datasets.
The output is structured so that the methodology and results sections can be integrated directly into academic papers. The underlying methodology has been published and is citable.
Nonsignificant results are not uncommon in science. eduSTAT doesn’t simply stop the analysis. Instead, it helps you identify possible influencing factors and objectively interpret the result. This allows you to document even nonsignificant findings with scientific accuracy.