Graduate Studies, UNL

 

Dissertations and Doctoral Documents from University of Nebraska-Lincoln, 2023–

First Advisor

James Bovaird

Degree Name

Doctor of Philosophy (Ph.D.)

Committee Members

Jiangang Xia, Jordan Wheeler, Rafael De Ayala

Department

Educational Psychology

Date of this Version

2025

Document Type

Dissertation

Citation

A dissertation presented to the faculty of the Graduate College of the University of Nebraska in partial fulfillment of requirements for the degree Doctor of Philosophy (Ph.D.)

Major: Educational Psychology

Under the supervision of Professor

Lincoln, Nebraska, December 2025

Comments

Copyright 2025, the author. Used by permission

Abstract

Responding to the high demand for new test items, model-based item development has emerged as a promising approach for efficient item creation. Item models within Assessment Engineering (AE) offer a systematic framework that effectively generates a large volume of new items. In the development of multiple-choice questions, well-constructed distractors are essential for ensuring item quality, although creating them requires significant time and resources. A distractor model is a sub-model within an item model, producing correct answers and distractors aligned with all potential items to be generated. Various types of distractor models have different characteristics and offer efficient methods for generating distractors depending on the nature of items and tests.

This study focuses on the features and usage of various types of distractor models within the AE framework, exploring their characteristics through empirical data. It aims to assess how different types of distractor models affect psychometric properties such as item difficulty and discrimination in real testing environments. By employing item statistics alongside examinees’ verbal responses from the think-aloud method, this study showed that items with certain types of distractor models are more difficult than those with other distractor models when other conditions of these items are consistent. The findings of this study provide empirical insights into the characteristics and effectiveness of these distractor models, offering practical guidance for item development using these models in both formative and summative assessment.

Advisor: James Bovaird

Download

Share

COinS