Start in three steps
About five minutes, from nothing to a result you can check.
-
Take a CSV and its dictionary
Pick a study from the table below. Download the CSV, and open its dictionary before the data — it tells you what each variable means, how it is coded, and where the missing values sit.
-
Import it into jamovi
Open jamovi Desktop, choose
File › Open › Browse, and select the CSV. Check the measurement levels and group ordering against the dictionary before running anything. -
Check yourself against the record
Run the workflow named for that study, then compare your output with the matching file in
expected-results/. If a number differs, one of you is wrong — and you can find out which.
The twelve datasets
Each one exercises a different design. Chapter numbers refer to the companion book, but nothing here requires it — ANALYSIS_MATRIX.md names the workflow and the result fields for every file.
| Dataset | What it teaches | jamovi workflow | Ch. |
|---|---|---|---|
| study_01_foundations_sleep_stress.csvdictionaryverified results | Foundations — sleep hours and stress in a student sample | Fresh CSV import, descriptives, histogram, boxplot | 1–7 |
| study_02_mindfulness_paired.csvdictionaryverified results | A paired pre/post design | Paired-samples t test, bootstrap demonstration, power planning | 8–11, 17 |
| study_03_feedback_two_groups.csvdictionaryverified results | Two independent groups | Welch independent-samples t test, effect size | 16, 18 |
| study_04_practice_spacing_anova.csvdictionaryverified results | Three or more groups | One-way ANOVA with post hoc comparisons | 19 |
| study_05_sleep_strategy_factorial.csvdictionaryverified results | Two crossed factors and an interaction | Factorial ANOVA, marginal means, interaction plot | 20 |
| study_06_mood_repeated.csvdictionaryverified results | Repeated measurements on the same people | Repeated-measures ANOVA, sphericity check | 21 |
| study_07_belonging_survey_regression.csvdictionaryverified results | Survey data, association and prediction | Simple and multiple regression, residual plots | 12–15 |
| study_08_help_seeking_categorical.csvdictionaryverified results | Counts and categories | Chi-square test, contingency table | 22 |
| study_09_skewed_wellbeing_nonparametric.csvdictionaryverified results | Skewed distributions, when the usual test will not do | Kruskal-Wallis, Mann-Whitney U | 23 |
| study_10_developmental_emotion_recognition.csvdictionaryverified results | Age groups and a covariate | ANOVA and ANCOVA | 24 |
| study_11_single_case_habit_tracking.csvdictionaryverified results | A single-case series across phases | Phase analysis, NAP, visual inspection | 25 |
| study_12_reporting_handoff_capstone.csvdictionaryverified results | Capstone — auditing and reporting a result | Audit, export, APA summary verification | 26–27 |
Browse the folders directly: data, dictionaries, expected-results, figures — or read the full analysis matrix.
What is in the repository
Everything below is public, free, and requires no account.
Twelve CSV datasets
Deterministic synthetic data, one file per teaching example, in data/.
A dictionary for every file
Machine-readable: variable meanings, coding, allowed values, and where the missing values are placed.
Analysis matrix
ANALYSIS_MATRIX.md maps each dataset to its jamovi workflow and the exact output fields that answer its question.
Verified result records
expected-results/ holds every reported statistic, recomputed independently from the source files.
Six figures
Generated from the same data at fixed seeds, in figures/generated/, so you can regenerate them yourself.
Checksums and errata
SHA256SUMS confirms the file you have is the file that was published. ERRATA.md is where corrections go.
Free psychology statistics practice materials
A complete problems-and-solutions resource for practice, teaching, adaptation, and reuse.
Eleven units, 232 problems, and 232 fully worked solutions
Psychology Statistics Practice Materials with Jamovi: Model Choice, Worked Solutions, and Scientific Evidence begins with research questions and design, then develops model choice, description, uncertainty, confidence intervals, hypothesis testing, effect sizes, power, correlation, regression, t tests, and one-way ANOVA.
Download the print-ready PDF or the editable DOCX. Both are licensed CC BY 4.0, so instructors and learners may share and adapt the resource with appropriate attribution. All scenarios are fictional and all teaching data are synthetic.
Version 1.1 has the permanent DOI 10.5281/zenodo.22262048.
Volume 2 — beyond group means
The four topics a first course reaches last, with the same structure and the same license.
Four units, 96 problems, and 96 fully worked solutions
Psychology Statistics Practice Materials with Jamovi, Volume 2: Categorical, Rank-Based, Quasi-Experimental, and Single-Case Evidence continues the unit numbering of Volume 1 at Unit 12. It covers categorical outcomes and association, rank-based comparisons and robust thinking, quasi-experimental comparison and covariate adjustment, and single-case designs and nonoverlap.
Its organising claim is that none of these is a substitute for a method in Volume 1. Each changes the question, not only the procedure — a categorical outcome has no mean, a rank method does not estimate a difference in the outcome’s units, adjusting for a consequence of the grouping variable removes part of the effect, and forty-two days of one person are one case.
Every teaching variant in the problems is a filter on one of four public datasets, so a reader can reproduce it in jamovi Desktop with a row filter and no new file. Both formats are licensed CC BY 4.0. All scenarios are fictional and all teaching data are synthetic.
Version 1.0 has the permanent DOI 10.5281/zenodo.22286929. Unit 13 matches jamovi’s epsilon-squared definition: 0.302 for the complete Study 09 file and 0.325 for its deletion variant, while retaining 0.291 and 0.314 under the separate label bias-adjusted rank effect size.
Together the two volumes cover fifteen units, 328 problems, and 328 fully worked solutions.
Use it in your own teaching
The licence is the point, not a footnote.
Take any of this into your own course, without asking
The datasets, dictionaries, figures, errata, and documentation are released under the Creative Commons Attribution 4.0 International licence. Short utility scripts are MIT. You may copy, adapt, and redistribute any of it — in worksheets, assignments, lecture slides, or your own published materials, including commercially — provided you give credit.
Every file is deterministic and synthetic. There are no real participant records, so nothing here raises a privacy question when it goes into a classroom, a public assignment, or a shared repository. Results computed from these files are teaching examples, not findings about real people.
A request, not a condition: if there is a design you keep wanting clean teaching data for and cannot find, say so. Adding one is a day's work, and guessing what people need is worse than being told.
Corrections and reader support
Check a verified correction, report a reproducible problem, or discuss the public teaching materials.
Check known corrections
Read the live errata log before reporting a possible book or companion-file error.
Report a reproducible problem
Use a structured Issue Form for an erratum, file-specific question or defect, or jamovi output mismatch.
Join a bounded discussion
Use Discussions for the public datasets, open practice materials, published workflows, and teaching feedback. Read the community scope first.
Please do not post private data, homework, or requests for individualized model selection or statistical consulting.
About
These files were built by Nicholas Elliott Karlson, PhD, a statistician and educator, and published by NEKpress Research. His work focuses on applied statistical reasoning, research design, reproducibility, and workflows in jamovi, Python, and R that make quantitative results easier to inspect, explain, and defend.
The data was built for a book, and does not require it
Psychology Research Methods and Statistics by Design with Jamovi takes each topic from a research question through a jamovi workflow to an APA-style result and the limitation that keeps it honest. Twenty-seven chapters. The datasets on this page are its companion — but they stand alone, and the analysis matrix names every workflow independently of the text.