Conrad Borchers

Verified

Conrad verified their affiliation via an institutional email.

Learn more

Verified

Conrad verified their affiliation via an institutional email.

Learn more

• Master of Science
• PhD Student at Carnegie Mellon University

Large Language Models (LLMs) hold promise as dynamic instructional aids. Yet, it remains unclear whether LLMs can replicate the adaptivity of intelligent tutoring systems (ITS)--where student knowledge and pedagogical strategies are explicitly modeled. We propose a prompt variation framework to assess LLM-generated instructional moves' adaptivity a...

The maturity of scientific communities can be measured by their degree of alignment on a common conceptual vision to guide research efforts. Recently, scholars in the CSCL community have called for increased efforts to establish shared theoretical frameworks to accelerate progress in the field of CSCL. The purpose of this study is to investigate if...

Two major sets of U.S. content standards have changed since 2010: the Common Core State Standards (CCSS) and Next Generation Science Standards (NGSS). While the CCSS received widespread pushback, the NGSS were broadly uncontentious. Why? Drawing from policy learning and networked publics theories, we used mixed effects modeling to estimate how twee...

Thematic analysis (TA) is a method used to identify, examine, and present themes within data. TA is often a manual, multistep, and time-intensive process requiring collaboration among multiple researchers. TA's iterative subtasks, including coding data, identifying themes, and resolving inter-coder disagreements, are especially laborious for large...

Thematic analysis (TA) is a method used to identify, examine, and present themes within data. TA is often a manual, multistep, and time-intensive process requiring collaboration among multiple researchers. TA’s iterative subtasks, including coding data, identifying themes, and resolving inter-coder disagreements, are especially laborious for large...

Algorithmic bias continues to be a key concern of learning analytics. We study the statistical properties of the Absolute Between-ROC Area (ABROCA) metric. This fairness measure quantifies group-level differences in classifier performance through the absolute difference in ROC curves. ABROCA is particularly useful for detecting nuanced performance...

Public Internet Data Mining methods enable studying educational institutions' public-facing communication. Multiple online data sources can illuminate differences in how different audiences are addressed online, opening the door for critical inquiry into emerging issues of representation and targeted advertising. The present study presents a case s...

Intelligent tutoring systems (ITSs) are effective in helping students learn; further research could make them even more effective. Particularly desirable is research into how students learn with these systems, how these systems best support student learning, and what learning sciences principles are key in ITSs. CTAT+Tutorshop provides a full stack...

Large Language Models (LLMs) like GPT-4o can help automate text classification tasks at low cost and scale. However, there are major concerns about the validity and reliability of LLM outputs. By contrast, human coding is generally more reliable but expensive to procure at scale. In this study, we propose a hybrid solution to leverage the strengths...

Intelligent tutoring system (ITS) provides learners with step-by-step problem-solving support through scaffolding. Most ITSs have been developed in the USA and incorporate American instructional strategies. How do non-American students perceive and use ITS with different native problem-solving strategies? The present study compares Stoich Tutor, an...

Assessing learners in ill-defined domains, such as scenario-based human tutoring training, is an area of limited research. Equity training requires a nuanced understanding of context, but do contemporary large language models (LLMs) have a knowledge base that can navigate these nuances? Legacy transformer models like BERT, in contrast, have less re...

Algorithmic bias is a pressing concern in educational data mining (EDM), as it risks amplifying inequities in learning outcomes. The Area Between ROC Curves (ABROCA) metric is frequently used to measure discrepancies in model performance across demographic groups to quantify overall model fairness. However, its skewed distribution--especially when...

Learning performance data, such as correct or incorrect answers and problem-solving attempts in Intelligent Tutoring Systems (ITSs), facilitate the assessment of knowledge mastery and the delivery of effective instructions. However, these data tend to be highly sparse (80% $\sim$ 90% missing observations) in most real-world applications. This data...

The Doer Effect states that completing more active learning activities, like practice questions, is more strongly related to positive learning outcomes than passive learning activities, like reading, watching, or listening to course materials. Although broad, most evidence has emerged from practice with tutoring systems in Western, Industrialized,...

Caregivers (i.e., parents and members of a child's caring community) are underappreciated stakeholders in learning analytics. Although caregiver involvement can enhance student academic outcomes, many obstacles hinder involvement, most notably knowledge gaps with respect to modern school curricula. An emerging topic of interest in learning analytic...

Equity is a core concern of learning analytics. However, applications that teach and assess equity skills, particularly at scale are lacking, often due to barriers in evaluating language. Advances in generative AI via large language models (LLMs) are being used in a wide range of applications, with this present work assessing its use in the equity...

The role of multiple-choice questions (MCQs) as effective learning tools has been debated in past research. While MCQs are widely used due to their ease in grading, open response questions are increasingly used for instruction, given advances in large language models (LLMs) for automated grading. This study evaluates MCQs effectiveness relative to...

Personalized problem selection enhances student practice in tutoring systems. Prior research has focused on transparent problem selection that supports learner control but rarely engages learners in selecting practice materials. We explored how different levels of control (i.e., full AI control, shared control, and full learner control), combined w...

In supervised machine learning (SML) research, large training datasets are essential for valid results. However, obtaining primary data in learning analytics (LA) is challenging. Data augmentation can address this by expanding and diversifying data, though its use in LA remains underexplored. This paper systematically compares data augmentation tec...

The sharing of personally identifiable information (PII) through social media platforms poses known risks to children's online privacy and safety. While the risks of oversharing PII through a range of digital contexts are becoming better understood, limited research has documented the social media practices of educational institutions that have a f...

Algorithmic bias continues to be a key concern of learning analytics. We study the statistical properties of the Absolute Between-ROC Area (ABROCA) metric. This fairness measure quantifies group-level differences in classifier performance through the absolute difference in ROC curves. ABROCA is particularly useful for detecting nuanced performance...

Self-regulated learning (SRL) is essential for learning across various contexts and domains. While technology-based learning environments can support SRL, comparisons of SRL processes across learning platforms and domains are scarce. As most prior research has investigated SRL patterns across learner performance levels, methods are lacking to inves...

Learning performance data describe correct and incorrect answers or problem-solving attempts in adaptive learning, such as in intelligent tutoring systems (ITSs). Learning performance data tend to be highly sparse (80\%\(\sim\)90\% missing observations) in most real-world applications due to adaptive item selection. This data sparsity presents chal...

Manuscript available at: https://doi.org/10.1007/978-3-031-64315-6_2 Ethical issues matter for artificial intelligence in education (AIED). Simultaneously, there is a gap between fundamental ethical critiques of AIED research goals and research practices doing ethical good. This article discusses the divide between AIED ethics (i.e., critical soci...

Coursera-REC is a Large Language Model-based course recommendation system designed to enhance MOOC learning experiences by tailoring recommendations to user-specific goals and preferences. Utilizing Retrieval-Augmented Generation (RAG), it retrieves contextual data from a comprehensive knowledge base to offer clear, reasoned course suggestions. Thi...

Teacher reflection is essential for K-12 classrooms, including effective and personalized instruction. Multimodal Learning Analytics (MMLA), integrating data from digital and physical learning environments, could support teacher reflection. Classroom data collected from sensors and TEL environments are needed to produce such analytics. These novel...

Full-text: https://cborchers.com/pdf/2024-isls-caregiver-tool-design.pdf

Full-text: https://cborchers.com/pdf/2024-isls-goal-setting-tool-design.pdf

Educational data mining increasingly leverages enrollment data for higher education applications. However, these data describe final end-of-semester course selections, not the often complex enrollment activities leading up to a finalized schedule. Fine-grain transaction data of student waitlist, add, and drop actions during academic semester planni...

Peer tutoring can improve learning by prompting learners to reflect. To assess whether peer interactions are conducive to learning and provide peer tutoring support accordingly , what tutorial dialog types relate to student learning most? Advancements in collaborative learning analytics allow for merging machine learning-based dialog act classifica...

Curriculum Analytics (CA) studies curriculum structure and student data to ensure the quality of educational programs. One desirable property of courses within curricula is that they are not unexpectedly more difficult for students of different backgrounds. While prior work points to likely variations in course difficulty across student groups, rob...

Think-aloud protocols are a common method to study self-regulated learning (SRL) during learning by problem-solving. Previous studies have manually transcribed and coded students' verbalizations, labeling the presence or absence of SRL strategies and then examined these SRL codes in relation to learning. However, the coding process is difficult to...

To replicate and expand on previous results showing that student learning rates are regular under favorable learning conditions (Koedinger et al., 2023), we used a dataset of 426 students who engaged with a cognitive tutoring system throughout an academic year. We used the individual additive factors model (iAFM) to estimate student parameters: int...

Intelligent Tutoring Systems (ITSs) have significantly enhanced adult literacy training, a key factor for societal participation, employment opportunities, and lifelong learning. Our study investigates the application of advanced AI models, including Large Language Models (LLMs) like GPT-4, for predicting learning performance in adult literacy prog...

Learning performance data (e.g., quiz scores and attempts) is significant for understanding learner engagement and knowledge mastery level. However, the learning performance data collected from Intelligent Tutoring Systems (ITSs) often suffers from sparsity, impacting the accuracy of learner modeling and knowledge assessments. To address this, we i...

One-on-one tutoring is an effective instructional method for enhancing learning, yet its efficacy hinges on tutor competencies. Novice math tutors often prioritize content-specific guidance, neglecting aspects such as social-emotional learning. Social-emotional learning promotes equity and inclusion and nurtures relationships with students, which i...

In numerous studies, intelligent tutoring systems (ITSs) have proven effective in helping students learn mathematics. Prior work posits that their effectiveness derives from efficiently providing eventually-correct practice opportunities. Yet, there is little empirical evidence on how learning processes with ITSs compare to other forms of instructi...

Facebook is widely used and researched. However, though the data generated by educational technology tools and social media platforms other than Facebook have been used for research purposes, very little research has used Facebook posts as a data source—with most studies relying on self-report studies. While it has historically been impractical (or...

Research indicates that teachers play an active and important role in classrooms with AI tutors. Yet, our scientific understanding of the way teacher practices around AI tutors mediate student learning is far from complete. In this paper, we investigate spatiotemporal factors of student-teacher interactions by analyzing student engagement and learn...

Diagnosing orthopantomograms (OPTs: panoramic radiographs) is an essential skill dentistry students acquire during university training. While prior research described experts’ visual search behavior in radiology as global-to-focal for chest radiographs and mammography, generalizability to a hybrid search task in OPTs (i.e., searching for multiple,...

Course load analytics (CLA) inferred from LMS and enrollment features can offer a more accurate representation of course workload to students than credit hours and potentially aid in their course selection decisions. In this study, we produce and evaluate the first machine-learned predictions of student course load ratings and generalize our model...

Schools and school districts use social media for a variety of reasons, but alongside the benefits of schools' social media use come potential risks to students' privacy. Using a novel dataset of around 18 million Facebook posts by schools and districts in the United States, we explore the extent to which personally identifiable information of stud...

Public schools and districts use social media to share announcements and communicate with parents and the community, but alongside such uses run risks to students’ privacy. Using a novel data set of 18 million posts on Facebook by schools and school districts in the United States, we sought to establish how frequently photos of students were shared...

Credit hours traditionally quantify expected instructional time per week in a course, informing student course selection decisions and contributing to degree requirement satisfaction. In this study, we investigate determinants of course load beyond this metric, including from course assignment structure and LMS interactions. Collecting 596 course l...

The growing capability and availability of generative language models has enabled a wide range of new downstream tasks. Academic research has identified, quantified and mitigated biases present in language models but is rarely tailored to downstream tasks where wider impact on individuals and society can be felt. In this work, we leverage one popul...

Schools and school districts now use social media for a variety of reasons, but alongside the benefits that accompany these K-12 educational institutions' social media use come questions about privacy and safety. We use a dataset of around 18 million posts by United States schools and districts on Facebook to explore how these posts might risk the...

Schools and districts use social media to share announcements and build vibrant communities, but alongside such uses run risks to students’ privacy and safety. Using a novel data set of more than 17 million posts from 2010-2020 of schools and school districts in the United States on Facebook, we sought to establish how frequently photos of students...

For many schools and districts in the United States, Facebook has emerged as an important tool for sharing timely information, building a sense of community, highlighting staff and students, and many other purposes. However, neither researchers nor schools and districts have paid enough attention to how their Facebook use may pose a risk to the pri...

The extraction of sentiment from text requires many method-ological decisions to make inferences about mood, opinion, and engagement in informal learning contexts. This study compares sentiment software (SentiStrength, LIWC, tidy-text, VADER) on N = 1,382,493 tweets in the context of the Next Generation Science Standards reform (N = 546,267) and U....

As the use of social media increases in daily life, it has also increased for institutions in the field of education. While there may be benefits for schools to use this media outlet, the privacy of students within those schools may be at risk when their names and photos are shared on such a publicly accessible domain. In this study, we analyzed th...

System-wide educational reforms are difficult to implement in the United States, but despite the difficulties, reforms can be successful, particularly when they are associated with broad public support. This study reports on the nature of the public sentiment expressed about a nationwide science education reform effort, the Next Generation Science...

For many if not posts schools and districts around the United States, the use of Facebook has emerged as a novel communication practice that serves several purposes, including sharing timely information, building a sense of community, and highlighting staff and students. An element of these posts that neither researchers nor, we think, most schools...

As the use of social media increases in daily life, it has also increased for institutions in the field of education. While there may be benefits for schools to use this media outlet, the privacy of students within those schools may be at risk when their names and photos are shared on such a publicly accessible domain. In this study, we analyzed th...

While the Next Generation Science Standards (NGSS) are a long-standing and widespread standards-based educational reform effort, they have received less public attention, and no studies have explored the sentiment of the views of multiple stakeholders toward them. To establish how public sentiment about this reform might be similar to or different...

Teachers frequently use Twitter to engage in professional learning activities. A prominent example of teachers' use of Twitter for such purposes is evident within the #NGSSchat community, which encouraged synchronous (at the same time) conversations between teachers and other educational stakeholders regarding the Next Generation Science Standards...

Inquiry-based learning, auch „forschendes Lernen“ genannt, stellt klassische didaktische Unterrichtskonzepte auf den Kopf. Das Vorwissen und insbesondere die Neugier der Lernenden bestimmen den Lernprozess, wobei der Prozess zwischen der Fragestellung und der Schlussfolgerung im Mittelpunkt steht.