Learning to interpret data in context is an important educational outcome. To assess students' attainment of this outcome, it is necessary to examine the interplay between their contextual and statistical reasoning. We describe a research method designed to do so. The method draws upon Toulmin's (1958, 2003) model of argumentation for the first…

To provide personalized support on educational platforms, it is crucial to model the evolution of students' knowledge states. Knowledge tracing is one of the most popular technologies for this purpose, and deep learning-based methods have achieved state-of-the-art performance. Compared to classical models, such as Bayesian knowledge tracing, which…

The National Center on Educational Outcomes (NCEO) held a virtual meeting of an Interim Assessment Advisory Panel on February 16 and 17, 2021, to tap into the panel members' collective knowledge about using interim assessments to support valid interpretations of what students with disabilities know and can do. The panel represented…

Education systems are increasingly creating rich, longitudinal data sets with frequent, and even real-time, data updates of many student measures, including daily attendance, homework submissions, and exam scores. These data sets provide an opportunity for district and school staff members to move beyond an indicators-based approach and instead…

This dissertation reports the results of three studies and a pilot study. The first study was a Monte Carlo validation study that examined the accuracy of a new visual inspection method, the semi-interquartile range method. Results of the study indicated that this method had lower levels of power than a previously validated method, the…

The accountability movement in education has caused school administrators and teachers to think differently about how they report, interpret, and use student assessment data. For example, legislative measures such as No Child Left Behind require school officials to report how all students are progressing toward established standards typically…

This dissertation comprises three papers that propose, discuss, and illustrate models to make improved inferences about research questions regarding student achievement in education. Addressing the types of questions common in educational research today requires three different "extensions" to traditional educational assessment: (1)…

This paper introduces five commonly used approaches to analyzing multiple-choice test data. They are classical test theory, factor analysis, cluster analysis, item response theory, and model analysis. Brief descriptions of the goals and algorithms of these approaches are provided, together with examples illustrating their applications in physics…

Students with severe/profound mental retardation performed functional tasks under training or probe conditions. Analysis of nine students' records indicated that students do not react to probe conditions as if they were extinction conditions; training data do not consistently reflect performance under probe conditions; and data collected under…

Explains how many of the narratives told by children are mapped onto frameworks carried in long-term memory. Assessment methods that speech-language pathologists can use to determine whether a child can use narrative frameworks and how to explain the assessment results are discussed. (Author/CR)

A recent conference on statistics education recommended that more emphasis be placed on the interpretation of research (IOR). Ways for developing and assessing IOR and providing a systematic framework for creating and selecting instructional materials for the independent assessment of specific IOR concepts are the focus of this paper. The…

This final report describes the activities and outcomes of a project designed to develop and establish a psychoeducational assessment model to enhance the ability of educators, psychologists, and learning specialists to design and conduct meaningful evaluations of students who are deaf-blind. By focusing on psychoeducational assessment, this…

Officials with the National Center for Education Statistics asked the National Research Center's Board on Testing and Assessment to convene a workshop to assist them with their decisions about reporting results for accommodating test takers. The day-long workshop brought together four panels of representatives from state assessment offices,…

Discusses which kinds of data educators should respect and which they should reject. Asserts that most state accountability tests fail to produce the kinds of data that will improve teaching and learning. Teachers can get the data they need from their own instructionally useful classroom assessments. (WFA)

This paper gives an overview of the methods of moderation, or quality assurance and quality control, as they may be more widely known, that are used to enhance the quality of student assessment. The discussion is based on the educational systems of the United Kingdom but is applicable to assessment in other countries. Quality in assessment is seen…