The process of generating challenging and appropriate distractors for multiple-choice questions is a complex and time-consuming task. Existing methods for an automated generation have limitations in proposing challenging distractors, or they fail to effectively filter out incorrect choices that closely resemble the correct answer, share synonymous…

Multi-angle question answering models have recently been proposed that promise to perform related tasks like question generation. However, performance on related tasks has not been thoroughly studied. We investigate a leading model called Macaw on the task of multiple choice question generation and evaluate its performance on three angles that…

We investigate encoder-decoder GRU networks with attention mechanism for solving a diverse array of elementary math problems with mathematical symbolic structures. We quantitatively measure performances of recurrent models on a given question type using a test set of unseen problems with a binary scoring and partial credit system. From our…

Multiple choice questions are traditionally expensive to produce. Recent advances in large language models (LLMs) have led to fine-tuned LLMs that generate questions competitive with human-authored questions. However, the relative capabilities of ChatGPT-family models have not yet been established for this task. We present a carefully-controlled…

One of the main goals of science education is to make students gain science process skills. Thus, it is significant to measure whether students gain those skills or not. For this purpose, various tests have been produced and used in various studies. This study aims to examine science process skills tests which have been used in the theses produced…

As implementation of the "Next Generation Science Standards" moves forward, there is a need for new assessments that can measure students' integrated three-dimensional science learning. The National Research Council has suggested that these assessments be multicomponent tasks that utilize a combination of item formats including…

Learning experience designers consistently balance the trade-off between open and close-ended activities. The growth and scalability of Computer Based Learning Platforms (CBLPs) have only magnified the importance of these design trade-offs. CBLPs often utilize close-ended activities (i.e. Multiple-Choice Questions [MCQs]) due to feasibility…

This study describes the role that Rasch measurement played in the development of assessments aligned to the "Next Generation Science Standards," tasks that require students to use the three dimensions of science practices, disciplinary core ideas and cross-cutting concepts to make sense of energy-related phenomena. A set of 27…

Technological advancements, particularly in artificial intelligence, significantly transform our society and work practices. Computational thinking (CT) has emerged as a crucial 21st-century skill, enabling individuals to solve problems more effectively through an automation-oriented perspective and fundamental concepts of computer science. To…

Tests of statistical significance often play a decisive role in establishing the empirical warrant of evidence-based research in education. The results from pattern-based assessment items, as introduced in this paper, are categorical and multimodal and do not immediately support the use of measures of central tendency as typically related to…

In this study we evaluated the stability of item parameters and student scores, using the pre-equated (pre-pandemic) parameters from Spring 2019 and post-equated (post-pandemic) parameters from Spring 2021 in two calibration and equating designs related to item parameter treatment: re-estimating all anchor parameters (Design 1) and holding the…

Many national science frameworks (e.g., Next Generation Science Standards) argue that developing mathematical modeling competencies is critical for students' deep understanding of science. However, science teachers may be unprepared to assess these competencies. We are addressing this need by developing virtual lab performance assessments that…

Nowadays, Malaysian ESL students struggle to read English texts, particularly narrative texts. Lacking in utilising and applying the suitable reading strategies are among factors that hinder reading comprehension. This study aimed to investigate the effectiveness of Question-Answer Relationship (QAR) strategy on ESL pupils' narrative reading and…

Item response theory (IRT) is a popular method to infer student abilities and item difficulties from observed test responses. However, IRT struggles with two challenges: How to map items to skills if multiple skills are present? And how to infer the ability of new students that have not been part of the training data? Inspired by recent advances…

The "Next Generation Science Standards" calls for new assessments that measure students' integrated three-dimensional science learning. The National Research Council has suggested that these assessments utilize a combination of item formats including constructed-response and multiple-choice. In this study, students were randomly assigned…