-
Unpacking Approaches to Learning and Teaching Machine Learning in K-12 Education: Transparency, Ethics, and Design Activities
Authors:
Luis Morales-Navarro,
Yasmin B. Kafai
Abstract:
In this conceptual paper, we review existing literature on artificial intelligence/machine learning (AI/ML) education to identify three approaches to how learning and teaching ML could be conceptualized. One of them, a data-driven approach, emphasizes providing young people with opportunities to create data sets, train, and test models. A second approach, learning algorithm-driven, prioritizes lea…
▽ More
In this conceptual paper, we review existing literature on artificial intelligence/machine learning (AI/ML) education to identify three approaches to how learning and teaching ML could be conceptualized. One of them, a data-driven approach, emphasizes providing young people with opportunities to create data sets, train, and test models. A second approach, learning algorithm-driven, prioritizes learning about how the learning algorithms or engines behind how ML models work. In addition, we identify efforts within a third approach that integrates the previous two. In our review, we focus on how the approaches: (1) glassbox and blackbox different aspects of ML, (2) build on learner interests and provide opportunities for designing applications, (3) integrate ethics and justice. In the discussion, we address the challenges and opportunities of current approaches and suggest future directions for the design of learning activities.
△ Less
Submitted 5 June, 2024;
originally announced June 2024.
-
Learning about Data, Algorithms, and Algorithmic Justice on TikTok in Personally Meaningful Ways
Authors:
Luis Morales-Navarro,
Yasmin B. Kafai,
Ha Nguyen,
Kayla DesPortes,
Ralph Vacca,
Camillia Matuk,
Megan Silander,
Anna Amato,
Peter Woods,
Francisco Castro,
Mia Shaw,
Selin Akgun,
Christine Greenhow,
Antero Garcia
Abstract:
TikTok, a popular short video sharing application, emerged as the dominant social media platform for young people, with a pronounced influence on how young women and people of color interact online. The application has become a global space for youth to connect with each other, offering not only entertainment but also opportunities to engage with artificial intelligence/machine learning (AI/ML)-dr…
▽ More
TikTok, a popular short video sharing application, emerged as the dominant social media platform for young people, with a pronounced influence on how young women and people of color interact online. The application has become a global space for youth to connect with each other, offering not only entertainment but also opportunities to engage with artificial intelligence/machine learning (AI/ML)-driven recommendations and create content using AI/M-powered tools, such as generative AI filters. This provides opportunities for youth to explore and question the inner workings of these systems, their implications, and even use them to advocate for causes they are passionate about. We present different perspectives on how youth may learn in personally meaningful ways when engaging with TikTok. We discuss how youth investigate how TikTok works (considering data and algorithms), take into account issues of ethics and algorithmic justice and use their understanding of the platform to advocate for change.
△ Less
Submitted 24 May, 2024;
originally announced May 2024.
-
Youth as Peer Auditors: Engaging Teenagers with Algorithm Auditing of Machine Learning Applications
Authors:
Luis Morales-Navarro,
Yasmin B. Kafai,
Vedya Konda,
Danaë Metaxa
Abstract:
As artificial intelligence/machine learning (AI/ML) applications become more pervasive in youth lives, supporting them to interact, design, and evaluate applications is crucial. This paper positions youth as auditors of their peers' ML-powered applications to better understand algorithmic systems' opaque inner workings and external impacts. In a two-week workshop, 13 youth (ages 14-15) designed an…
▽ More
As artificial intelligence/machine learning (AI/ML) applications become more pervasive in youth lives, supporting them to interact, design, and evaluate applications is crucial. This paper positions youth as auditors of their peers' ML-powered applications to better understand algorithmic systems' opaque inner workings and external impacts. In a two-week workshop, 13 youth (ages 14-15) designed and audited ML-powered applications. We analyzed pre/post clinical interviews in which youth were presented with auditing tasks. The analyses show that after the workshop all youth identified algorithmic biases and inferred dataset and model design issues. Youth also discussed algorithmic justice issues and ML model improvements. Furthermore, youth reflected that auditing provided them new perspectives on model functionality and ideas to improve their own models. This work contributes (1) a conceptualization of algorithm auditing for youth; and (2) empirical evidence of the potential benefits of auditing. We discuss potential uses of algorithm auditing in learning and child-computer interaction research.
△ Less
Submitted 16 April, 2024; v1 submitted 8 April, 2024;
originally announced April 2024.
-
Investigating Youths' Everyday Understanding of Machine Learning Applications: a Knowledge-in-Pieces Perspective
Authors:
Luis Morales-Navarro,
Yasmin B. Kafai
Abstract:
Despite recent calls for including artificial intelligence (AI) literacy in K-12 education, not enough attention has been paid to studying youths' everyday knowledge about machine learning (ML). Most research has examined how youths attribute intelligence to AI/ML systems. Other studies have centered on youths' theories and hypotheses about ML highlighting their misconceptions and how these may hi…
▽ More
Despite recent calls for including artificial intelligence (AI) literacy in K-12 education, not enough attention has been paid to studying youths' everyday knowledge about machine learning (ML). Most research has examined how youths attribute intelligence to AI/ML systems. Other studies have centered on youths' theories and hypotheses about ML highlighting their misconceptions and how these may hinder learning. However, research on conceptual change shows that youths may not have coherent theories about scientific phenomena and instead have knowledge pieces that can be productive for formal learning. We investigate teens' everyday understanding of ML through a knowledge-in-pieces perspective. Our analyses reveal that youths showed some understanding that ML applications learn from training data and that applications recognize patterns in input data and depending on these provide different outputs. We discuss how these findings expand our knowledge base and implications for the design of tools and activities to introduce youths to ML.
△ Less
Submitted 31 March, 2024;
originally announced April 2024.
-
Twenty Constructionist Things to Do with Artificial Intelligence and Machine Learning
Authors:
Yasmin Kafai,
Luis Morales-Navarro
Abstract:
In this paper, we build on the 1971 memo "Twenty Things to Do With a Computer" by Seymour Papert and Cynthia Solomon and propose twenty constructionist things to do with artificial intelligence and machine learning. Several proposals build on ideas developed in the original memo while others are new and address topics in science, mathematics, and the arts. In reviewing the big themes, we notice a…
▽ More
In this paper, we build on the 1971 memo "Twenty Things to Do With a Computer" by Seymour Papert and Cynthia Solomon and propose twenty constructionist things to do with artificial intelligence and machine learning. Several proposals build on ideas developed in the original memo while others are new and address topics in science, mathematics, and the arts. In reviewing the big themes, we notice a renewed interest in children's engagement not just for technical proficiency but also to cultivate a deeper understanding of their own cognitive processes. Furthermore, the ideas stress the importance of designing personally relevant AI/ML applications, moving beyond isolated models and off-the-shelf datasets disconnected from their interests. We also acknowledge the social aspects of data production involved in making AI/ML applications. Finally, we highlight the critical dimensions necessary to address potential harmful algorithmic biases and consequences of AI/ML applications.
△ Less
Submitted 17 April, 2024; v1 submitted 9 February, 2024;
originally announced February 2024.
-
Understanding Growth Mindset Practices in an Introductory Physical Computing Classroom: High School Students' Engagement with Debugging by Design Activities
Authors:
Luis Morales-Navarro,
Deborah A. Fields,
Yasmin B. Kafai
Abstract:
Background and Context: While debugging is recognized as an essential practice, for many students, encountering bugs can generate emotional responses such as fear and anxiety that can lead to disengagement and the avoidance of computer programming. Growth mindsets can support perseverance and learning in these situations, yet few studies have investigated how growth mindsets emerge in practice amo…
▽ More
Background and Context: While debugging is recognized as an essential practice, for many students, encountering bugs can generate emotional responses such as fear and anxiety that can lead to disengagement and the avoidance of computer programming. Growth mindsets can support perseverance and learning in these situations, yet few studies have investigated how growth mindsets emerge in practice amongst K-12 computing students facing physical computing debugging challenges. Objective: We seek to understand what (if any) growth mindset practices high school students exhibited when creating and exchanging buggy physical computing projects for their peers to solve during a Debugging by Design activity as part of their introductory computing course. Method: We focused on moment-to-moment microgenetic analysis of student interactions in designing and solving bugs for others to examine the practices students exhibited that demonstrated the development of a growth mindset and the contexts in which these practices emerged. Findings: We identified five emergent growth mindset practices: choosing challenges that lead to more learning, persisting after setbacks, giving and valuing praise for effort, approaching learning as constant improvement, and develo** comfort with failure. Students most often exhibited these practices in peer-to-peer interactions and while making buggy physical computing projects for their peers to solve. Implications: Our analysis contributes to a more holistic understanding of students' social, emotional, and motivational approaches to debugging physical computing projects through the characterization of growth mindset practices. The presented inventory of growth mindset practices may be helpful to further study growth mindset in action in other computing settings.
△ Less
Submitted 11 February, 2024; v1 submitted 2 February, 2024;
originally announced February 2024.
-
Failure Artifact Scenarios to Understand High School Students' Growth in Troubleshooting Physical Computing Projects
Authors:
L. Morales-Navarro,
D. A. Fields,
D. Barapatre,
Y. B. Kafai
Abstract:
Debugging physical computing projects provides a rich context to understand cross-disciplinary problem solving that integrates multiple domains of computing and engineering. Yet understanding and assessing students' learning of debugging remains a challenge, particularly in understudied areas such as physical computing, since finding and fixing hardware and software bugs is a deeply contextual pra…
▽ More
Debugging physical computing projects provides a rich context to understand cross-disciplinary problem solving that integrates multiple domains of computing and engineering. Yet understanding and assessing students' learning of debugging remains a challenge, particularly in understudied areas such as physical computing, since finding and fixing hardware and software bugs is a deeply contextual practice. In this paper we draw on the rich history of clinical interviews to develop and pilot "failure artifact scenarios" in order to study changes in students' approaches to debugging and troubleshooting electronic textiles (e-textiles). We applied this clinical interview protocol before and after an eight-week-long e-textiles unit. We analyzed pre/post clinical interviews from 18 students at four different schools. The analysis revealed that students improved in identifying bugs with greater specificity, and across domains, and in considering multiple causes for bugs. We discuss implications for develo** tools to assess students' debugging abilities through contextualized debugging scenarios in physical computing.
△ Less
Submitted 15 December, 2023; v1 submitted 28 November, 2023;
originally announced November 2023.
-
Not Just Training, Also Testing: High School Youths' Perspective-Taking through Peer Testing Machine Learning-Powered Applications
Authors:
L. Morales-Navarro,
M. Shah,
Y. B. Kafai
Abstract:
Most attention in K-12 artificial intelligence and machine learning (AI/ML) education has been given to having youths train models, with much less attention to the equally important testing of models when creating machine learning applications. Testing ML applications allows for the evaluation of models against predictions and can help creators of applications identify and address failure and edge…
▽ More
Most attention in K-12 artificial intelligence and machine learning (AI/ML) education has been given to having youths train models, with much less attention to the equally important testing of models when creating machine learning applications. Testing ML applications allows for the evaluation of models against predictions and can help creators of applications identify and address failure and edge cases that could negatively impact user experiences. We investigate how testing each other's projects supported youths to take perspective about functionality, performance, and potential issues in their own projects. We analyzed testing worksheets, audio and video recordings collected during a two week workshop in which 11 high school youths created physical computing projects that included (audio, pose, and image) ML classifiers. We found that through peer-testing youths reflected on the size of their training datasets, the diversity of their training data, the design of their classes and the contexts in which they produced training data. We discuss future directions for research on peer-testing in AI/ML education and current limitations for these kinds of activities.
△ Less
Submitted 14 December, 2023; v1 submitted 21 November, 2023;
originally announced November 2023.
-
Connecting Beliefs, Mindsets, Anxiety, and Self-Efficacy in Computer Science Learning: An Instrument for Capturing Secondary School Students' Self-Beliefs
Authors:
Luis Morales-Navarro,
Michael T. Giang,
Deborah A. Fields,
Yasmin B. Kafai
Abstract:
Background and Context: Few instruments exist to measure students' CS engagement and learning especially in areas where coding happens with creative, project-based learning and in regard to students' self-beliefs about computing. Objective: We introduce the CS Interests and Beliefs Inventory (CSIBI), an instrument designed for novice secondary students learning by designing projects (particularly…
▽ More
Background and Context: Few instruments exist to measure students' CS engagement and learning especially in areas where coding happens with creative, project-based learning and in regard to students' self-beliefs about computing. Objective: We introduce the CS Interests and Beliefs Inventory (CSIBI), an instrument designed for novice secondary students learning by designing projects (particularly with physical computing). The inventory contains subscales on beliefs on problem solving competency, fascination in design, value of CS, creative expression, and beliefs about context-specific CS abilities alongside programming mindsets and outcomes. We explain the creation of the instrument and attend to the role of mindsets as mediators of self-beliefs and how CSIBI may be adapted to other K-12 project-based learning settings. Method: We administered the instrument to 303 novice CS secondary students who largely came from historically marginalized backgrounds (gender, ethnicity, and socioeconomic status). We assessed the nine-factor structure for the 32-item instrument using confirmatory factor analysis and tested the hypothesized model of mindsets as mediators with structural equation modeling. Findings: We confirmed the nine factor structure of CSIBI and found significant positive correlations across factors. The structural model results showed that problem solving competency beliefs and CS creative expression promoted programming growth mindset, which subsequently fostered students' programming self-concept. Implications: We validated an instrument to measure secondary students' self-beliefs in CS that fills several gaps in K-12 CS measurement tools by focusing on contexts of learning by designing. CSIBI can be easily adapted to other learning by designing computing education contexts.
△ Less
Submitted 19 July, 2023;
originally announced July 2023.
-
Making Sense of Machine Learning: Integrating Youth's Conceptual, Creative, and Critical Understandings of AI
Authors:
Luis Morales-Navarro,
Yasmin B. Kafai,
Francisco Castro,
William Payne,
Kayla DesPortes,
Daniella DiPaola,
Randi Williams,
Safinah Ali,
Cynthia Breazeal,
Clifford Lee,
Elisabeth Soep,
Duri Long,
Brian Magerko,
Jaemarie Solyst,
Amy Ogan,
Cansu Tatar,
Shiyan Jiang,
Jie Chao,
Carolyn P. Rosé,
Sepehr Vakil
Abstract:
Understanding how youth make sense of machine learning and how learning about machine learning can be supported in and out of school is more relevant than ever before as young people interact with machine learning powered applications everyday; while connecting with friends, listening to music, playing games, or attending school. In this symposium, we present different perspectives on understandin…
▽ More
Understanding how youth make sense of machine learning and how learning about machine learning can be supported in and out of school is more relevant than ever before as young people interact with machine learning powered applications everyday; while connecting with friends, listening to music, playing games, or attending school. In this symposium, we present different perspectives on understanding how learners make sense of machine learning in their everyday lives, how sensemaking of machine learning can be supported in and out of school through the construction of applications, and how youth critically evaluate machine learning powered systems. We discuss how sensemaking of machine learning applications involves the development and integration of conceptual, creative, and critical understandings that are increasingly important to prepare youth to participate in the world.
△ Less
Submitted 4 May, 2023;
originally announced May 2023.
-
Designing Bugs or Doing Another Project: Effects on Secondary Students' Self-Beliefs in Computer Science
Authors:
Luis Morales-Navarro,
Deborah A. Fields,
Michael Giang,
Yasmin B Kafai
Abstract:
Debugging, finding and fixing bugs in code, is a heterogeneous process that shapes novice learners' self-beliefs and motivation in computing. Our Debugging by Design intervention (DbD) provocatively puts students in control over bugs by having them collaborate on designing creative buggy projects during an electronic textiles unit in an introductory computing course. We implemented DbD virtually i…
▽ More
Debugging, finding and fixing bugs in code, is a heterogeneous process that shapes novice learners' self-beliefs and motivation in computing. Our Debugging by Design intervention (DbD) provocatively puts students in control over bugs by having them collaborate on designing creative buggy projects during an electronic textiles unit in an introductory computing course. We implemented DbD virtually in eight classrooms with two teachers in public schools with historically marginalized populations, using a quasi-experimental design. Data from this study included post-activity results from a validated survey instrument (N=144). For all students, project completion correlated with increased computer science creative expression and e-textiles coding self-efficacy. In the comparison classes, project completion correlated with reduced programming anxiety, problem-solving competency beliefs, and programming self-concept. In DbD classes, project completion is uniquely correlated with increased fascination with design and programming growth mindset. In the discussion, we consider the relative benefits of DbD versus other open-ended projects.
△ Less
Submitted 4 May, 2023;
originally announced May 2023.
-
Conceptualizing Approaches to Critical Computing Education: Inquiry, Design and Reimagination
Authors:
Luis Morales-Navarro,
Yasmin B. Kafai
Abstract:
As several critical issues in computing such as algorithmic bias, discriminatory practices, and techno-solutionism have become more visible, numerous efforts are being proposed to integrate criticality in K-16 computing education. Yet, how exactly these efforts address criticality and translate it into classroom practice is not clear. In this conceptual paper, we first historicize how current effo…
▽ More
As several critical issues in computing such as algorithmic bias, discriminatory practices, and techno-solutionism have become more visible, numerous efforts are being proposed to integrate criticality in K-16 computing education. Yet, how exactly these efforts address criticality and translate it into classroom practice is not clear. In this conceptual paper, we first historicize how current efforts in critical computing education draw on previous work which has promoted learner empowerment through critical analysis and production. We then identify three emergent approaches: (1) inquiry, (2) design and (3) reimagination that build on and expand these critical traditions in computing education. Finally, we discuss how these approaches highlight issues to be addressed and provide directions for further computing education research.
△ Less
Submitted 21 April, 2023;
originally announced April 2023.
-
Computer-Aided Personalized Education
Authors:
Rajeev Alur,
Richard Baraniuk,
Rastislav Bodik,
Ann Drobnis,
Sumit Gulwani,
Bjoern Hartmann,
Yasmin Kafai,
Jeff Karpicke,
Ran Libeskind-Hadas,
Debra Richardson,
Armando Solar-Lezama,
Candace Thille,
Moshe Vardi
Abstract:
The shortage of people trained in STEM fields is becoming acute, and universities and colleges are straining to satisfy this demand. In the case of computer science, for instance, the number of US students taking introductory courses has grown three-fold in the past decade. Recently, massive open online courses (MOOCs) have been promoted as a way to ease this strain. This at best provides access t…
▽ More
The shortage of people trained in STEM fields is becoming acute, and universities and colleges are straining to satisfy this demand. In the case of computer science, for instance, the number of US students taking introductory courses has grown three-fold in the past decade. Recently, massive open online courses (MOOCs) have been promoted as a way to ease this strain. This at best provides access to education. The bigger challenge though is co** with heterogeneous backgrounds of different students, retention, providing feedback, and assessment. Personalized education relying on computational tools can address this challenge.
While automated tutoring has been studied at different times in different communities, recent advances in computing and education technology offer exciting opportunities to transform the manner in which students learn. In particular, at least three trends are significant. First, progress in logical reasoning, data analytics, and natural language processing has led to tutoring tools for automatic assessment, personalized instruction including targeted feedback, and adaptive content generation for a variety of subjects. Second, research in the science of learning and human-computer interaction is leading to a better understanding of how different students learn, when and what types of interventions are effective for different instructional goals, and how to measure the success of educational tools. Finally, the recent emergence of online education platforms, both in academia and industry, is leading to new opportunities for the development of a shared infrastructure. This CCC workshop brought together researchers develo** educational tools based on technologies such as logical reasoning and machine learning with researchers in education, human-computer interaction, and cognitive psychology.
△ Less
Submitted 7 July, 2020;
originally announced July 2020.
