mirtverse: A Friendly Workflow and Reporting Layer for 'mirt'
mirtverse 0.3.0 adds an AI-in-measurement layer, an open textbook, seven more teaching labs, tools for instructors and an agent that analyzes test data with Claude. The paper describing the package is out as a preprint on PsyArXiv (https://doi.org/10.31234/osf.io/j9ya7_v1). The full list is in the changelog . Highlights AI in measurement, with psychometric checks. train_scorer() and agreement_report() train and judge an automated scoring engine. item_model() , generate_isomorphs() and calibrate_families() generate items from templates and check that each family is equally difficult. item_features() and predict_difficulty() predict difficulty
Resource information
- Format
- zip
- Original repository
- 23980162
- DOI
- 10.5281/zenodo.23980162
- Credit basis
- Creators in Zenodo record metadata
- Access
- Open files available from the original Zenodo deposit
- Last checked
- Nov 5, 2026
- License
- gpl-3.0-or-later
- Last record update
- Nov 5, 2026
About this resource
mirtverse 0.3.0 adds an AI-in-measurement layer, an open textbook, seven more teaching labs, tools for instructors and an agent that analyzes test data with Claude. The paper describing the package is out as a preprint on PsyArXiv (https://doi.org/10.31234/osf.io/j9ya7_v1). The full list is in the changelog . Highlights AI in measurement, with psychometric checks. train_scorer() and agreement_report() train and judge an automated scoring engine. item_model() , generate_isomorphs() and calibrate_families() generate items from templates and check that each family is equally difficult. item_features() and predict_difficulty() predict difficulty from item text. ai_audit() checks an AI component in one call (agreement, differential scoring, drift and DIF), and validity_argument() turns the audit into a Kane-style validity argument. The experimental simulate_llm_respondents() and compare_synthetic() test language-model "respondents" against real data. Runs without an API key. mv_llm_replay() answers every language-model call from recorded responses, so lessons and examples run offline and give the same result every time. mv_llm_record() records a live session. AI Hub. A new section of the website for psychometricians learning AI: A learning path, a table that translates 40+ machine learning terms into psychometric ideas, a reading list and a one-day workshop kit ( copy_workshop() ). Open textbook. Item Response Theory with mirt and mirtverse (https://www.okanbulut.com/mirtverse/book/, DOI 10.5281/zenodo.23885605): 18 chapters, from classical item analysis to the validity of AI scoring, every example in mirt and mirtverse. Teaching. Labs 4 to 10: Linking two forms, designing an adaptive test, scoring open-ended responses, automated scoring, item text and difficulty, automatic item generation, and auditing an AI-scored test. An instructor toolkit ( new_course() , new_lab() , swap_lab_data() , check_material() ) builds labs on any dataset, and mv_mcp_server() lets an AI assistant edit course materials safely. For theses and papers. cite_methods() drafts an APA Methods paragraph with every number taken from the fitted model, and the new "Thesis kit" article walks from item responses to that paragraph. The mirtverse agent. A Claude Code plugin with five skills and an MCP server, mv_agent_server() , that fits models, checks items, tests DIF, writes the report and saves every step as one R script. Website. A new home page, a "Start here" menu by audience, a gallery, three narrated video demos, a "Which IRT model?" flowchart and poster, and Turkish materials. Install install.packages("mirtverse", repos = c("https://okanbulut.r-universe.dev", "https://cloud.r-project.org")) Or install this release from GitHub: remotes::install_github("")