Patricia Martinkova
I am a senior researcher, the head of the Department of Statistical Modelling at the Institute of Computer Science of the Czech Academy of Sciences, and the leader of the Computational Psychometrics Group. I am also an Associate Professor at the Faculty of Education at Charles University, where I initiated the Center for Educational Measurement and Psychometrics with the 2018–2020 PRIMUS project. I am Fulbright alumna and 2013-2015 visiting research scholar with the University of Washington and an affiliate at the UW Center for Statistics and the Social Sciences and at the UW Department of Statistics.
My research focuses on designing advanced models and estimators to extract detailed, granular insights from ratings and measurements across diverse domains such as education, psychology, and health. I utilize simulation studies and real-world data examples to showcase the superior properties and advantages of these newly developed methods and estimators compared to those traditionally employed in practice.
I am a big fan of reproducible research and freely available software. Part of my research is exploited in the interactive online application ShinyItemAnalysis, which is now being used in many research fields worldwide.
My teaching interests cover various topics of applied statistics and psychometrics. I teach courses NMST570 Statistical Methods in Psychometrics and NMST571 Seminar in Psychometrics. I am interested in incorporating active learning techniques into teaching.
Recent news
The book "Computational Aspects of Psychometric Methods. With R" is now published with Chapman and Hall/CRC. Copies may be ordered online. See Publications for other recent manuscripts.
The EduCoDe project will be funded by the Czech Science Foundation for period 2025 - 2027.
Our team was the local organizer of the IMPS 2024 conference, with more than 570 registrants from 40 countries. Thanks to all of you who participated!
Project "Research of Excellence on Digital Technologies and Wellbeing" co-funded by EU for period 2024 - 2028. Our team focuses on developing, optimizing and applying techniques for statistical modelling of complex data on wellbeing.
New R package SIAtools: 'ShinyItemAnalysis' Modules Development Toolkit now on CRAN.
ShinyItemAnalysis version 1.5.1 was released!
The SIA add-on modules are better discoverable! See the
full changelog.
SIAmodules version 0.1.1 now on CRAN!
See the
changelog.
Manuscript "Assessing quality of selection procedures: Lower bound of false positive rate as a function of inter-rater reliability" has been published in British Journal of Mathematical and Statistical Psychology.
Manuscript "Is video games' effect on attitudes universal? Results from an empirical study comparing video games' impact on the attitude change of players with different backgrounds" published in Journal of Computer Assisted Learning.
Manuscript "Item Difficulty Prediction Using Item Text Features: Comparison of Predictive Performance across Machine-Learning Algorithms" has been published in Mathematics.
Review of "Supervised Machine Learning for Text Analysis in R"
has been published in Journal of the American Statistical Society.
Manuscript "Combining Item Purification and Multiple Comparison Adjustment Methods in Detection of Differential Item Functioning" has been published in Multivariate Behavioral Research.
ShinyItemAnalysis version 1.5.0 was released!
The app is now extendable via add-on modules! See the
full changelog here.
A collection of ShinyItemAnalysis modules were releases in a R package
SIAmodules. They are no longer hosted at a standalone webpage, but dynamicaly recognized and
added in ShinyItemAnalysis.
Manuscript "Assessing inter-rater reliability with heterogeneous variance components models: Flexible approach accounting for contextual variables" has been published in Journal of Educational and Behavioral Statistics.
COMIRESTROKE study protocol article has been printed in Frontiers in Neurology.
New preprint on applicant selection has been added to arXiv.
I will give a spotlight talk on Computational aspects of reliability estimation at the IMPS conference in Bologna.
I presented at Metascience 2021 conference. See the slides of my talk on range-retstricted IRR in grant proposal peer review. Here is the accompanying ShinyItemAnalysis module.
New article Can video games change attitudes towards history? Results from a laboratory experiment measuring short- and long-term effects published in J Comput Assist Learn.
I will be presenting at IMPS 2021. Talks on reliability and on DIF detection are available for registered participants.
I will be presenting at useR!2021 on Teaching computational aspects of psychometrics with R and Shiny. See the slides.
Article When zero may not be zero: A cautionary note on the use of inter-rater reliability in evaluating grant peer review published in the Journal of the Royal Statistical Society Series A.
An article on the difNLR package was published in The R Journal.
Article Is academic tracking related to gains in learning competence? Using propensity score matching and differential item change functioning analysis for better understanding of tracking implications is available in Learning and Instruction.
Full-day workshop Introduction to Differential Item Functioning (DIF) analysis with R and ShinyItemAnalysis will be held at AEA-Europe conference in Lisabon on Nov 13, 2019. Registrations are open!
Article ShinyItemAnalysis for Teaching Psychometrics and to Enforce Routine Analysis of Educational Tests has been published in The R Journal.
International workshop on Psychometric Computing Psychoco 2019 will take place on February 21 - 22 in Prague, Czech Republic.
Article Disparities in ratings of internal and external applicants: A case for model-based inter-rater reliability has been published in PLOS ONE.
Article Checking equity: Why DIF analysis should be a routine part of developing conceptual assessments by Martinkova et. al, has been published in CBE - Lifesciences Education.
© 2024 Patricia Martinkova / Template design adapted from Andreas Viklund