Juan D. Pinto
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Juan D. Pinto has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Luc Paquette
Curriculum and Instruction University of Illinois Urbana‐Champaign Champaign Illinois USA
6 shared publications - Qianhui Liu
University of Illinois Urbana-Champaign
3 shared publications - Rebecca Mary Quintana
University of Michigan
3 shared publications Aysa Xuemo Fan
University of Illinois Urbana-Champaign
2 shared publications- Aysa Xuemo Fan
University of Illinois at Urbana-Champaign
2 shared publications - Yingbin Zhang
Institute of Artificial Intelligence in Education, South China Normal University, Guangzhou, China
2 shared publications
Latest Publications
- 2 November 2024
- Published by Springer Nature
- p. 136-151
Using LLM-Based Filtering to Develop Reliable Coding Schemes for Rare Debugging Strategies
- 2 November 2024
- Published by Springer Nature
- p. 136-151
Using LLM-Based Filtering to Develop Reliable Coding Schemes for Rare Debugging Strategies
- 28 September 2024
- Published by Springer Nature
- p. 93-109
Applications of Explainable AI (XAI) in Education
- 28 September 2024
- Published by Springer Nature
- p. 93-109
Applications of Explainable AI (XAI) in Education
- Open Access
Deep Learning for Educational Data Science
With the ever-growing presence of deep artificial neural networks in every facet of modern life, a growing body of researchers in educational data science -- a field consisting of various interrelated research communities -- have turned their attention to leveraging these powerful algorithms within the domain of education. Use cases range from advanced knowledge tracing models that can leverage open-ended student essays or snippets of code to automatic affect and behavior detectors that can identify when a student is frustrated or aimlessly trying to solve problems unproductively -- and much more. This chapter provides a brief introduction to deep learning, describes some of its advantages and limitations, presents a survey of its many uses in education, and discusses how it may further come to shape the field of educational data science.
Open AccessDeep Learning for Educational Data Science
With the ever-growing presence of deep artificial neural networks in every facet of modern life, a growing body of researchers in educational data science -- a field consisting of various interrelated research communities -- have turned their attention to leveraging these powerful algorithms within the domain of education. Use cases range from advanced knowledge tracing models that can leverage open-ended student essays or snippets of code to automatic affect and behavior detectors that can identify when a student is frustrated or aimlessly trying to solve problems unproductively -- and much more. This chapter provides a brief introduction to deep learning, describes some of its advantages and limitations, presents a survey of its many uses in education, and discusses how it may further come to shape the field of educational data science.
- 22 October 2023
- Published by Springer Nature
- p. 125-139
Investigating the Relationship Between Programming Experience and Debugging Behaviors in an Introductory Computer Science Course
- 22 October 2023
- Published by Springer Nature
- p. 125-139
Investigating the Relationship Between Programming Experience and Debugging Behaviors in an Introductory Computer Science Course
- Open Access
- 3 October 2023
- Published by International Society of the Learning Sciences
Intelligent Tutors, Cultural Blind Spots: Implications of Underrepresentation in Adaptive Learning Research
As intelligent tutoring systems (ITS) continue to mature and spread to lower-and middle-income countries (LMICs), it is becoming clear that basing such tools around models of learning overwhelmingly studied in Western countries has left severe blind spots that need to be addressed. This paper aims to raise awareness of the disconnect between the collectivist value systems espoused by many cultures and the Western individualistic mentality of ITS developers and researchers, along with some implications.
- 3 October 2023
- Published by International Society of the Learning Sciences
Open AccessIntelligent Tutors, Cultural Blind Spots: Implications of Underrepresentation in Adaptive Learning Research
As intelligent tutoring systems (ITS) continue to mature and spread to lower-and middle-income countries (LMICs), it is becoming clear that basing such tools around models of learning overwhelmingly studied in Western countries has left severe blind spots that need to be addressed. This paper aims to raise awareness of the disconnect between the collectivist value systems espoused by many cultures and the Western individualistic mentality of ITS developers and researchers, along with some implications.
- Open Access
- 10 March 2023
- Published by Wiley in Computer Applications in Engineering Education
- Vol. 31 (4) , 1041-1058
Utilizing programming traces to explore and model the dimensions of novices' code‐writing skill
Studies have found that most novice programmers have low proficiency in writing code. However, it is unclear what subskills compose code writing and which subskills novice programmers struggle with. This study utilizes programming traces to identify latent subskills that constitute code writing so that teachers can offer specific instruction on the weak subskills. Data were collected from an undergraduate course teaching introductory computer science in Java. Six hundred and fourteen students made submissions to homework programming questions in a web‐based learning system. Based on the submission traces, we computed 11 features related to correctness and time students spent on their submissions. We conducted an exploratory factor analysis on two‐thirds of students selected randomly and identified four factors. The first factor, code style proficiency, was mainly related to code style errors. The second, syntactic proficiency, concerned compiler errors. The third is semantic proficiency, which concerns runtime and logic errors. The fourth, syntactic debugging proficiency, concerned the success rate and time required for fixing compiler and code style errors. A confirmatory factor analysis conducted on the remaining one‐third of the data supported the four‐factor structure. The factor model showed measurement invariance between the data set where the model was developed and two new datasets, one from the same sample but collected at a different time point and another from a different sample and context (onsite course vs. online course). The factors were related to prior programming abilities, programming language familiarity, and future exam performance. These associations provided validity evidence for the factor model.
- 10 March 2023
- Published by Wiley in Computer Applications in Engineering Education
- Vol. 31 (4) , 1041-1058
Open AccessUtilizing programming traces to explore and model the dimensions of novices' code‐writing skill
Studies have found that most novice programmers have low proficiency in writing code. However, it is unclear what subskills compose code writing and which subskills novice programmers struggle with. This study utilizes programming traces to identify latent subskills that constitute code writing so that teachers can offer specific instruction on the weak subskills. Data were collected from an undergraduate course teaching introductory computer science in Java. Six hundred and fourteen students made submissions to homework programming questions in a web‐based learning system. Based on the submission traces, we computed 11 features related to correctness and time students spent on their submissions. We conducted an exploratory factor analysis on two‐thirds of students selected randomly and identified four factors. The first factor, code style proficiency, was mainly related to code style errors. The second, syntactic proficiency, concerned compiler errors. The third is semantic proficiency, which concerns runtime and logic errors. The fourth, syntactic debugging proficiency, concerned the success rate and time required for fixing compiler and code style errors. A confirmatory factor analysis conducted on the remaining one‐third of the data supported the four‐factor structure. The factor model showed measurement invariance between the data set where the model was developed and two new datasets, one from the same sample but collected at a different time point and another from a different sample and context (onsite course vs. online course). The factors were related to prior programming abilities, programming language familiarity, and future exam performance. These associations provided validity evidence for the factor model.
- 22 October 2022
- Published by Springer Nature in Education and Information Technologies
- Vol. 28 (4) , 4673-4701
Combining latent profile analysis and programming traces to understand novices’ differences in debugging
- 22 October 2022
- Published by Springer Nature in Education and Information Technologies
- Vol. 28 (4) , 4673-4701
Combining latent profile analysis and programming traces to understand novices’ differences in debugging
- Open Access
- 1 December 2021
- Published by The Online Learning Consortium in Online Learning
- Vol. 25 (4) , 101-118
What We Learned When We Compared Discussion Posts from One MOOC Hosted on Two Platforms
We compared discussion posts from a data science ethics MOOC that was hosted on two platforms. We characterized one platform as “open” because learners can respond to discussion prompts while viewing and responding to others. We characterized the other platform as “locked” because learners must respond to a discussion prompt before they can view and respond to others. Our objective is to determine whether these platform differences are consequential and have the potential to impact learning. We analyzed direct responses to two discussion prompts from two modules located in modules two and six of an eight module course. We used conventional content analysis to derive codes directly from the data. Posts on the “open” platform were characterized by failure to completely address the prompt and showed evidence of persuasion tactics and reflective activity. Posts on the “locked” platform were characterized by an apparent intent to complete the task and an assertive tone. Posts on the “locked” platform also showed a diversity of ideas through the corpus of responses. Our findings show that MOOC platform interfaces can lead to qualitative differences in discussion posts in ways that have the potential to impact learning. Our study provides insight into how “open” and “locked” platform designs have the potential to shape ways that learners respond to discussion prompts in MOOCs. Our study offers guidance for instructors making decisions on MOOC platform choice and activities situated within a learning experience.We used conventional content analysis to derive codes directly from the data. Posts on the “open” platform were characterized by failure to completely address the prompt and showed evidence of persuasion tactics and reflective activity. Posts on the “locked” platform were characterized by an apparent intent to complete the task and an assertive tone. Posts on the “locked” platform also showed a diversity of ideas through the corpus of responses. Our findings show that MOOC platform interfaces can lead to qualitative differences in discussion posts in ways that have the potential to impact learning. Our study provides insight into how “open” and “locked” platform designs have the potential to shape ways that learners respond to discussion prompts in MOOCs. Our study offers guidance for instructors making decisions on MOOC platform choice and activities situated within a learning experience.
- 1 December 2021
- Published by The Online Learning Consortium in Online Learning
- Vol. 25 (4) , 101-118
Open AccessWhat We Learned When We Compared Discussion Posts from One MOOC Hosted on Two Platforms
We compared discussion posts from a data science ethics MOOC that was hosted on two platforms. We characterized one platform as “open” because learners can respond to discussion prompts while viewing and responding to others. We characterized the other platform as “locked” because learners must respond to a discussion prompt before they can view and respond to others. Our objective is to determine whether these platform differences are consequential and have the potential to impact learning. We analyzed direct responses to two discussion prompts from two modules located in modules two and six of an eight module course. We used conventional content analysis to derive codes directly from the data. Posts on the “open” platform were characterized by failure to completely address the prompt and showed evidence of persuasion tactics and reflective activity. Posts on the “locked” platform were characterized by an apparent intent to complete the task and an assertive tone. Posts on the “locked” platform also showed a diversity of ideas through the corpus of responses. Our findings show that MOOC platform interfaces can lead to qualitative differences in discussion posts in ways that have the potential to impact learning. Our study provides insight into how “open” and “locked” platform designs have the potential to shape ways that learners respond to discussion prompts in MOOCs. Our study offers guidance for instructors making decisions on MOOC platform choice and activities situated within a learning experience.We used conventional content analysis to derive codes directly from the data. Posts on the “open” platform were characterized by failure to completely address the prompt and showed evidence of persuasion tactics and reflective activity. Posts on the “locked” platform were characterized by an apparent intent to complete the task and an assertive tone. Posts on the “locked” platform also showed a diversity of ideas through the corpus of responses. Our findings show that MOOC platform interfaces can lead to qualitative differences in discussion posts in ways that have the potential to impact learning. Our study provides insight into how “open” and “locked” platform designs have the potential to shape ways that learners respond to discussion prompts in MOOCs. Our study offers guidance for instructors making decisions on MOOC platform choice and activities situated within a learning experience.