WK83: Item Response Theory

Clinical measurement of latent constructs (e.g. quality of life, pain, depression) often requires use of questionnaires. Measurement of such constructs using questionnaires relies on statistical measurement models. The Item Response Theory (IRT) provides statistical models which link the latent construct score of a patient to the questionnaire responses of the patient. IRT models are the current golden standard in this context, compared to the Classical Test Theory (CTT) model where the latent scores are linked to responses by sum-scores.

Date:
23, 24, 27 January 2020 Tuition fee: € 950,-
City:
Amsterdam
Course coordinator:
Marianne Avetisyan
Language:
English
Learning method:
Lectures and Computerpractical
Examination:
Written examination (facultative)
Examination dates:
See schedule at 'Tentamens'
Number of EC:
2
  • Course Description

    Clinical measurement of latent constructs (e.g. quality of life, pain, depression) often requires use of questionnaires. Measurement of such constructs using questionnaires relies on statistical measurement models. The Item Response Theory (IRT) provides statistical models which link the latent construct score of a patient to the questionnaire responses of the patient. IRT models are the current golden standard in this context, compared to the Classical Test Theory (CTT) model where the latent scores are linked to responses by sum-scores.

    An important practical advantage of the IRT based measurement instruments is the flexibility in the clinical use. For instance, creating short form questionnaires tailored for specific target groups requires little effort. Further, a computerized adaptive test (CAT) can be developed which selects the most informative questions for each individual during the administration, based on the previous responses the individual. The IRT framework has advantages in clinical research as well. For instance, IRT based construct scores are demonstrated to be preferable to the alternative (i.e. using CTT and sum-scores) in the analysis of clinical trials (Ard, Galasko, & Edland, 2013; Gorter, Fox, & Twisk, 2015).

    Theoretical and practical strengths of IRT culminated in two large scale projects in medical measurement field over the last decades: PROMIS (http://www.healthmeasures.net/explore-measurement-systems/promis) and NeuroQoL (http://www.healthmeasures.net/explore-measurement-systems/neuro-qol). In the same period, articles suggesting the use of the IRT framework in the psychological and psychiatric assessment (Fraley, Waller, & Brennan, 2000) and books focusing on familiarizing the clinical measurement researchers with IRT methods were published (Ayala, 2009; Embretson & Reise, 2013; Reise & Revicki, 2014).

    This course is an introductory course on IRT with a focus on its use in development and validation of patient reported outcome measures. We will cover the underlying principles of IRT and the practical issues researchers may face while using IRT in medical measurement context. Our priority will be to do this in an accessible, interactive and intuitive way. First, we will start with the conceptual differences between IRT and Classical Test Theory (CTT). Then, we will study the theoretical pillars of IRT (e.g. the concept of latent trait, statistical modeling of items and responses, assumptions of IRT). Once the ground is set, we will move on to the practical issues in applications of IRT, such as checking the model assumptions, evaluating the model fit and estimating reliability of measurements. Finally, we will focus on advanced issues such as differential item functioning, the principles behind item banking and computerized adaptive testing.

    The course consists of alternating series of lectures and practicals. Lectures will introduce the theoretical and practical issues in IRT, whereas practicals will focus on applying this understanding on example datasets and problems.


  • Faculty


    Marianna Avetisyan, PhD, course coordinator
    Department of Epidemiology & Biostatistics. Amsterdam UMC, location VUmc and Dutch-Flemish PROMIS Group.

    Caroline B. Terwee, PhD Assistant professor of Clinimetrics
    Department of Epidemiology & Biostatistics. Amsterdam UMC, location VUmc and Dutch-Flemish PROMIS Group.

    Dr. Caroline Terwee is coordinator of the knowledge center measurement instruments of the VU University Medical Center. Her expertise is on the development and evaluation of measurement instrument and related methodological issues, with a special interest in systematic reviews of measurement instruments. She (co)-authored more than 70 international peer reviewed papers on measurement issues and is co-author of the book “Measurement in Medicine”.




  • Learning objectives

    An active participant of this course will develop an understanding of the IRT in the context of medical measurement. Practically, the participant will be able to evaluate a scientific article which uses IRT and perform IRT analyses.

    These main goals are considered to be sum of the following sub goals:

    1.The participant should be able to explain the concept of latent trait and how it is utilized in building statistical models of items and responses,

    2.The participant should be able to explain the basic assumptions of IRT and evaluate their implications in medical measurement context, as well as check these assumptions and judge their impact on analysis,

    3.The participant should be able to explain the properties of different IRT models and judge their usefulness in practical situations,

    4.The participant should be able to estimate item parameters of IRT models and interpret the parameters produced by different IRT models,

    5.The participant should be able to explain the differential item functioning and its implications in medical measurement context, as well as test whether the measurement invariance holds for an instrument,

    6.The participant should be able to explain the test information concept, calculate reliability of an IRT based instrument and use it in item banking and computerized adaptive test production.


  • Course material

    On the first day of the course, a course reader will be handed out. The course reader will contain (1) presentations used in lectures, (2) detailed theoretical explanations, (3) examples of relevant literature and (4) a guide for practicals. The course materials will also be available on Canvas, the digital learning environment of VUmc.


  • Course Requirements

    Attendants are expected to have at least basic knowledge of epidemiological and statistical methods. We will be using R during practicals; hence familiarity with the basic R operations is required. We will provide example R files to illustrate the level of familiarity we expect, as well as self-study resources for you to develop this familiarity.

    For the practicals, you will need your own laptop.


  • Target group

    The course is designed for healthcare practitioners and researchers who are active in medical, allied health, psychological, or behavioral research and who deal with the development, evaluation, and interpretation of health measurements using IRT.


  • Exam and accreditation


    Exam

    Participants who take this course as part of the Master Epidemiology always complete the course with an exam. Other participants can choose if they want to complete the course with an exam.
    The exam will be in English. Only when you pass the exam you get a certificate showing the credits (study points/EC).
    The examination dates can be found on the website of EpidM.

    Anyone who wants to participate in the examination should apply at least four weeks before the exam to register via the website: https://www.epidm.nl/nl/tentamens/

    An example examination will be available on Canvas, the digital learning environment of Vumc.

    Please note that use of books (digital or hard copy) is not allowed during the examination.

    Accreditation points

    Only for Dutch students!
    If you wish to be considered for accreditation points connected to this course, you must sign the attendance list on the last day of the course.

    To qualify for the accreditation points, you must have been present throughout the course.



This website uses cookies to provide you with the best user experience. By using this site, you accept the use of cookies. More information Hide this notification
You are using an outdated browser, for optimal use we suggest that you update your browser.