Colloquium, Trainings, and Events


  • Image for Some Recent Developments in the Item Response Theory Modeling of Forced-Choice Data

    Some Recent Developments in the Item Response Theory Modeling of Forced-Choice Data

    Last updated: March 12, 2019, 2:02 p.m.

    Speaker: Dr. Jimmy de la Torre. Faculty of Education, The University of Hong Kong

    Date: March 7, 2019, 2 p.m.

    Venue: UP Stat Lecture Hall 1

    Abstract: Likert format is the most popular item type for the measurement of non-cognitive constructs (e.g., attitude, personality). However, it is well known that responses to this format can be affected by response biases such as social desirability and faking. To address the bias-related issues in Likert format items, alternative item formats (i.e., forced-choice items) have been proposed. Traditional scoring of forced-choice items produces ipsative scores, which, some researchers claim, are appropriate for intra-individual, but not inter-individual comparisons. In this presentation, several item response theory models, which can be used in conjunction with different forced-choice formats, are presented to extract scores from ipsative instruments that can be used for inter-individual comparisons. In addition to the model formulation, model estimation and related issues are discussed. Simulation studies and a real data example are provided to examine the practical viability of the different models. The presentation concludes with a discussion of some unresolved issues and potential research directions pertaining to the analysis of forced-choice data.

    Kindly pre-register with Ms. Nancy Angala (npangala@up.edu.ph).

Highlights