This quantitative study investigates the predictive power of machine learning (ML) models on degree completion among adult learners in higher education, emphasizing the enhancement of data-driven decision-making (DDDM). By analyzing three ML models - Random Forest, Gradient-Boosting machine (GBM), and CART Decision Tree - within a not-for-profit,…

Machine learning (ML) techniques have been successfully applied to a wide array of applications. This dissertation aims to take application data handling into account when developing ML-based solutions for real-world problems through a holistic framework. To demonstrate the generality of our framework, we consider two real-world applications: spam…

Data-based decision making (DBDM) is a process of using student data to inform instructional decisions and intensify intervention for students whose data indicate inadequate academic and behavioral progress. Data teams, an important structure for DBDM, are a collaborative group of school faculty who meet to systematically analyze student data,…

This study investigates how 11- to 12-year-old students construct data-based decision trees using data cards for classification purposes. We examine the students' heuristics and reasoning during this process. The research is based on an eight-week teaching unit during which students labeled data, built decision trees, and assessed them using test…

Evidence-based policymaking can transform the delivery of education services, restore public trust in schools, and improve outcomes for students. It can cut through the noise of political and cultural divisions and give decision makers clarity on how to prioritize the use of limited resources. And it can help build a shared understanding of where…

Based on our experience developing and delivering a highly successful data visualization course within a Master of Science in Business Analytics program, we present a taxonomy for data visualization courses and recommend content and pedagogical features for each type of data visualization course therein. We also discuss the interdependence between…

Higher education has faced increased pressure to prove its quality through 'economic efficiency' and 'value for money', thrusting institutions into what researchers call an 'audit culture'. This study explores whether and how the audit culture has impacted educational developers in Canadian post-secondary institutions. Results include a summary of…

State postsecondary education data systems are vital assets for policymakers, researchers, and the public. The State Higher Education Executive Officers Association's (SHEEO) Communities of Practice project builds upon SHEEO's efforts to measure the capacity and effective use of state postsecondary data systems and provides states with…

Implementing supported and customized employment in all their components is essential for supporting job seekers with disabilities to achieve their career goals. We asked 42 employment consultants in nine employment programs to respond daily to three quick questions about their employment support activities, for 6 months. Through monthly coaching,…

Learning analytics is an emerging trend in American community colleges brought about by technological advancements, institutional accountability, and external pressure on institutions to substantiate and improve learning. Learning analytics can potentially improve student engagement, retention, and success but also possess inherent ethical and…

This nonexperimental, survey-based online quantitative study on nonprofit managers' technical training measures the extent of the influence on big data technology use. The unified theory of acceptance and use of technology is a theoretical framework to determine whether business managers are trained to have know-how in using big data technology.…

The purpose of this project was based on the premise that equipping Christian educational leaders in a Christian school setting with comprehensive, holistic, and lifespan spiritual survey results would result in more intentional faith and learning integration into all of school life. The project consisted of six, one hour seminars where…

Biomedical data science education faces the challenge of preparing students for conducting rigorous research with increasingly complex and large datasets. At the same time, philosophers of science face the challenge of making their expertise accessible for scientists in such a way that it can improve everyday research practice. Here, we…

Actionability is a critical issue in learning analytics for driving impact in learning, bridging the gap between insights and improvement. This dissertation places actionability at the forefront, integrating it throughout the learning analytics process to fully leverage its potential. The study involves designing, developing, and implementing…

Improvement networks are a relatively recent phenomenon in US education that create interorganizational networks of educators working together to improve specific educational problems. A shared emphasis of these networks is the use of data to support the improvement process, but little is known about their data use in practice. This study takes an…