Educational Measurement-Issues and Practice最新文献

The template-based automated item-generation (TAIG) approach that involves template creation, item generation, item selection, field-testing, and evaluation has more steps than the traditional item development method. Consequentially, there is more margin for error in this process, and any template errors can be cascaded to the generated items. Therefore, it is essential to eliminate the source of errors and ensure the quality of the template so items can be problem-free. The article introduces a process to reduce template errors at the early stage of template development, minimize the impact of template errors on generated items, and increase the survival rates of generated items. The article also discusses a statistical method to establish confidence in the quality of the template by systematically examining the quality of the generated items. The proposed method can reduce the review process for some items generated from a template.

I study student characteristics and academic performance at the University of California, where consideration of an applicant's ethnicity has been banned since 1996 and SAT scores were used in admitting students to the university until fall 2021. I show the following: (1) SAT scores were more important than high school grades in predicting first-year university GPA; (2) the use of SAT scores alone or with high school grades in determining admission is biased in favor of admitting underrepresented minorities and students who are socioeconomically disadvantaged; (3) SAT scores are more important and high school grades are less important in predicting GPA for underrepresented minorities and/or those students from low-income families than they are for those students who are white and/or from high-income families; and (4) the University of California found ways to admit a significant number of underrepresented minorities despite many of them having low SAT scores.

Computerized adaptive testing (CAT) has gained deserved popularity in the administration of educational and professional assessments, but continues to face test security challenges. To ensure sustained quality assurance and testing integrity, it is imperative to establish and maintain multiple stable item pools that are consistent in terms of psychometric characteristics and content specifications. This study introduces the Honeycomb Pool Assembly (HPA) framework, an innovative solution for the construction of multiple parallel item pools for CAT that maximizes item utilization in the item bank. The HPA framework comprises two stages—cell assembly and pool assembly—and uses a mixed integer programming modeling approach. An empirical study demonstrated HPA's effectiveness in creating a large number of parallel pools using a real-world high-stakes CAT assessment item bank. The HPA framework offers several advantages, including (a) simultaneous creation of multiple parallel pools, (b) simplification of item pool maintenance, and (c) flexibility in establishing statistical and operational constraints. Moreover, it can help testing organizations efficiently manage and monitor the health of their item banks. Thus, the HPA framework is expected to be a valuable tool for testing professionals and organizations to address test security challenges and maintain the integrity of high-stakes CAT assessments.

The recent surge of machine learning (ML) has impacted many disciplines, including educational and psychological measurement (hereafter shortened as measurement). The measurement literature has seen rapid growth in applications of ML to solve measurement problems. However, as we emphasize in this article, it is imperative to critically examine the potential risks associated with involving ML in measurement. The MxML project aims to explore the relationship between measurement and ML, so as to identify and address the risks and better harness the power of ML to serve measurement missions. This paper describes the first study of the MxML project, in which we summarize the state of the field of applications, extensions, and discussions about ML in measurement contexts with a systematic review of the recent 10 years’ literature. We provide a snapshot of the literature in (1) areas of measurement where ML is discussed, (2) types of articles (e.g., applications, conceptual, etc.), (3) ML methods discussed, and (4) potential risks associated with involving ML in measurement, which result from the differences between what measurement tasks need versus what ML techniques can provide.

Categorical growth models describe examinee growth in terms of performance-level category transitions, which implies that some percentage of examinees will be misclassified. This paper introduces a new procedure for estimating the classification accuracy of categorical growth models, based on Rudner's classification accuracy index for item response theory–based assessments. Results of a simulation study are presented to provide evidence for the accuracy and validity of the approach. Also, an empirical example is presented to demonstrate the approach using data from the Indiana Student Performance Readiness and Observation of Understanding Tool growth model, which classifies examinees into growth categories used by the Office of Special Education Programs to monitor the progress of preschool children who receive special education services.

In scientific literacy, knowledge integration (KI) is a scaffolding-based theory to assist students' scientific inquiry learning. To drive students to be self-directed, many courses have been developed based on KI framework. However, few efforts have been made to evaluate the outcome of students' learning under KI instruction. Moreover, finer-grained information has been pursued to better understand students' learning and how it progresses over time. In this article, a normative procedure of building and choosing cognitive diagnosis models (CDMs) and attribute hierarchies was formulated under KI theory. We examined the utility of CDMs for evaluating students' knowledge status in KI learning. The results of the data analysis confirmed an intuitive assumption of the hierarchical structure of KI components. Furthermore, analysis of pre- and posttests using a higher-order, hidden Markov model tracked students' skill acquisition while integrating knowledge. Results showed that students make significant progress after using the web-based inquiry science environment (WISE) platform.

Remote proctoring, or monitoring test takers through internet-based, video-recording software, has become critical for maintaining test security on high-stakes assessments. The main role of remote proctors is to make judgments about test takers' behaviors and decide whether these behaviors constitute rule violations. Variability in proctor decision making, or the degree to which humans/proctors make different decisions about the same test-taking behaviors, can be problematic for both test takers and test users (e.g., universities). In this paper, we measure variability in proctor decision making over time on a high-stakes English language proficiency test. Our results show that (1) proctors systematically differ in their decision making and (2) these differences are trait-like (i.e., ranging from lenient to strict), but (3) systematic variability in decisions can be reduced. Based on these findings, we recommend that test security providers conduct regular measurements of proctors’ judgments and take actions to reduce variability in proctor decision making.

The purpose of this study is to introduce and evaluate a method for generating reading comprehension items using template-based automatic item generation. To begin, we describe a new model for generating reading comprehension items called the text analysis cognitive model assessing inferential skills across different reading passages. Next, the text analysis cognitive model is used to generate reading comprehension items where examinees are required to read a passage and identify the irrelevant sentence. The sentences for the generated passages were created using OpenAI GPT-3.5. Finally, the quality of the generated items was evaluated. The generated items were reviewed by three subject-matter experts. The generated items were also administered to a sample of 1,607 Grade-8 students. The correct options for the generated items produced a similar level of difficulty and yielded strong discrimination power while the incorrect options served as effective distractors. Implications of augmented intelligence for item development are discussed.

English language proficiency (ELP) assessment scores are used by states to make high-stakes decisions related to linguistic support in instruction and assessment for English learner (EL) students and for EL student reclassification. Changes to both academic content standards and ELP academic standards within the last decade have resulted in increased academic rigor and language demands. In this study, we explored the association between EL student performance over time on content (English language arts and mathematics) and ELP assessments, generally finding evidence of positive associations. Modeling the simultaneous association between changes over time in both content and ELP assessment performance contributes empirical evidence about the role of language in ELA and mathematics development and provides contextual information to serve as validity evidence for score inferences for EL students.