We propose a novel model, Wide & Deep Item Response Theory (Wide & Deep IRT), to predict the correctness of students' responses to questions using historical clickstream data. This model combines the strengths of conventional Item Response Theory (IRT) models and Wide & Deep Learning for Recommender Systems. By leveraging clickstream…

The vibrant demand for academic excellence in the twenty-first century has brought diverse determinants of students' outcome into play. However, few studies have validated the instruments and examined the mediating effect between exogenous and endogenous variables of the student outcome model. This study, therefore, investigates the psychometric…

This paper briefly discusses four measurement challenges of data science or "big data" in educational assessments that are enabled by technology: 1. Dealing with change over time via time-based data. 2. How a digital performance space's relationships interact with learner actions, communications and products. 3. How layers of…

Accountability systems that measure student learning rather than student achievement have the potential to more accurately evaluate school quality. However, one methodological concern has remained surprisingly absent from discussions of value-added modeling. Standardized assessments that exhibit either positive or negative correlations between…

Increasing college participation rates, and diversity in student population, is posing a challenge to colleges in their attempts to facilitate learners achieve their full academic potential. Learning analytics is an evolving discipline with capability for educational data analysis that could enable better understanding of learning process, and…

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)…

Effective teaching has long been an issue of national concern, but in recent years focus on the effectiveness of programs to produce high-quality teachers has sharpened. Long-standing achievement gaps persist despite large-scale legislative changes at the federal and state levels, and American students continue to show poorer performance on…

Value-added modeling is an alternative approach to test-based accountability systems based on the proportions of students scoring at or above pre-determined proficiency levels. Value-added modeling techniques provide opportunities to estimate an individual teacher's effect on student learning, while allowing for the possibility to control for the…

As currently practiced, value-added assessment relies on a strong assumption about the scales used to measure student achievement, namely that these are interval scales, with equal-sized gains at all points on the scale representing the same increment of learning. Many of the metrics in which test results are expressed do not have this property…

This study explores the latent structure of scores on the Learning and Study Strategies Inventory (LASSI) and analyzes the relationship between this structure and students' academic performance. Two independent samples of college freshmen (n = 527) and seniors (n = 429) completed the LASSI. Data analysis of the first sample revealed acceptable…

A detailed description of five latent structure models of achievement measurement is presented. The first project paper, by David L. McArthur, analyzes the history of mental testing to show how conventional item analysis procedures were developed, and how dissatisfaction with them has led to fragmentation. The range of distinct conceptual and…

The conceptual framework for the instructionally sensitive assessment that guided the analyses of data from the Second International Mathematics Study (SIMS) are described. The intellectual rationale for the project effort and the current draft version of the guiding conception of "instructionally sensitive psychometrics" are discussed,…

The purpose of this study was to examine, describe, evaluate, and compare the rating behavior of faculty consultants who scored essays written for the Advanced Placement English Literature and Composition (AP® ELC) Exam. Data from the 1999 AP ELC Exam were analyzed using FACETS (Linacre, 1998) and SAS. The faculty consultants were not all…