Guide to Teaching Data Science: An Interdisciplinary Approach

Prologue
Contents
List of Figures
List of Tables
List of Exercises
1 Introduction—What is This Guide About?
	1.1 Introduction
	1.2 Motivation for Writing This Guide
	1.3 Pedagogical Principles and Guidelines for Teaching Data Science
	1.4 The Structure of the Guide to Teaching Data Science
		1.4.1 The Five Parts of the Guide
		1.4.2 The Chapters of the Guide
	1.5 How to Use This Guide?
		1.5.1 Data Science Instructors in Academia
		1.5.2 K-12 Teachers
		1.5.3 Instructors of the Methods of Teaching Data Science (MTDS) Course
	1.6 Learning Environments for Data Science
		1.6.1 Textual Programing Environments for Data Science
		1.6.2 Visual Programing Environments for Data Science
	1.7 Conclusion
	Reference
Part I Overview of Data Science and Data Science Education
2 What is Data Science?
	2.1 The Interdisciplinary Development of Data Science
		2.1.1 The Origins of Data Science in Statistics
		2.1.2 The Origins of Data Science in Computer Science
		2.1.3 The Origins of Data Science in Application Domains: The Case of Business Analytics
	2.2 Data Science as a Science
	2.3 Data Science as a Research Method
		2.3.1 Exploratory Data Analysis
		2.3.2 Machine Learning as a Research Method
	2.4 Data Science as a Discipline
	2.5 Data Science as a Workflow
	2.6 Data Science as a Profession
	2.7 Conclusion
	References
3 Data Science Thinking
	3.1 Introduction
	3.2 Data Thinking and the Thinking Skills Associated with Its Components
		3.2.1 Computational Thinking
		3.2.2 Statistical Thinking
		3.2.3 Mathematical Thinking
		3.2.4 Application Domain Thinking
		3.2.5 Data Thinking
	3.3 Thinking About Data Science Thinking
	3.4 Conclusion
	References
4 The Birth of a New Discipline: Data Science Education
	4.1 Introduction
	4.2 Undergraduate Data Science Curricula Initiatives
		4.2.1 Strengthening Data Science Education Through Collaboration, 2015
		4.2.2 Curriculum Guidelines for Undergraduate Programs in Data Science, 2016
		4.2.3 The EDISON Data Science Framework, 2017
		4.2.4 Envisioning the Data Science Discipline, 2018
		4.2.5 Computing Competencies for Undergraduate Data Science Curricula, 2017–2021
	4.3 Data Science Curriculum for K-12
	4.4 Meta-Analysis of Data Science Curricula
	4.5 Conclusion
	References
Part II Opportunities and Challenges of Data Science Education
5 Opportunities in Data Science Education
	5.1 Introduction
	5.2 Teaching STEM in a Real-World Context
	5.3 Teaching STEM with Real-World Data
	5.4 Bridging Gender Gaps in STEM Education
	5.5 Teaching Twenty-First Century Skills
	5.6 Interdisciplinary Pedagogy
	5.7 Professional Development for Teachers
	5.8 Conclusion
	References
6 The Interdisciplinarity Challenge
	6.1 Introduction
	6.2 The Interdisciplinary Structure of Data Science
	6.3 Is Data Science More About Computer Science or More About Statistics?
	6.4 Integrating the Application Domain
		6.4.1 Data Science Pedagogical Content Knowledge (PCK)
		6.4.2 Developing Interdisciplinary Programs
		6.4.3 Integrating the Application Domain into Courses in Computer Science, Mathematics, and Statistics
		6.4.4 Mentoring Interdisciplinary Projects
	6.5 Conclusion
	References
7 The Variety of Data Science Learners
	7.1 Introduction
	7.2 Data Science for K-12 Pupils
	7.3 Data Science for High School Computer Science Pupils
	7.4 Data Science for Undergraduate Students
	7.5 Data Science for Graduate Students
	7.6 Data Science for Researchers
	7.7 Data Science for Data Science Educators
	7.8 Data Science for Professional Practitioners in the Industry
	7.9 Data Science for Policy Makers
	7.10 Data Science for Users
	7.11 Data Science for the General Public
	7.12 Activities on Learning Environments for Data Science
	7.13 Conclusion
	References
8 Data Science as a Research Method
	8.1 Introduction
	8.2 Data Science as a Research Method
		8.2.1 Data Science Research as a Grounded Theory
		8.2.2 The Application Domain Knowledge in Data Science Research
	8.3 Research Skills
		8.3.1 Cognitive Skills: Awareness of the Importance of Model Assessment—Explainability and Evaluation
		8.3.2 Organizational Skills: Understanding the Field of the Organization
		8.3.3 Technological Skills: Data Visualization
	8.4 Pedagogical Challenges of Teaching Research Skills
	8.5 Conclusion
	References
9 The Pedagogical Chasm in Data Science Education
	9.1 The Diffusion of Innovation Theory
	9.2 The Crossing the Chasm Theory
	9.3 The Data Science Curriculum Case Study from the Diffusion of Innovation Perspective
		9.3.1 The Story of the New Program
		9.3.2 The Teachers’ Perspective
	9.4 The Pedagogical Chasm
	9.5 Conclusion
	References
Part III Teaching Professional Aspects of Data Science
10 The Data Science Workflow
	10.1 Data Workflow
	10.2 Data Collection
	10.3 Data Preparation
	10.4 Exploratory Data Analysis
	10.5 Modeling
		10.5.1 Data Quantity, Quality, and Coverage
		10.5.2 Feature Engineering
	10.6 Communication and Action
	10.7 Conclusion
	References
11 Professional Skills and Soft Skills in Data Science
	11.1 Introduction
	11.2 Professional Skills
		11.2.1 Cognitive Skills: Thinking on Different Levels of Abstraction
		11.2.2 Organizational Skills: Storytelling
		11.2.3 Technological Skills: Programming for Data Science
	11.3 Soft Skills
		11.3.1 Cognitive Skills: Learning
		11.3.2 Organizational Skills: Teamwork and Collaboration
		11.3.3 Technological Skills: Debugging Data and Models
	11.4 Teaching Notes
	11.5 Conclusion
	References
12 Social and Ethical Issues of Data Science
	12.1 Introduction
	12.2 Data Science Ethics
	12.3 Methods of Teaching Social Aspects of Data Science
		12.3.1 Teaching Principles
		12.3.2 Kinds of Activities
	12.4 Conclusion
	References
Part IV Machine Learning Education
13 The Pedagogical Challenge of Machine Learning Education
	13.1 Introduction
	13.2 Black Box and White Box Understandings
	13.3 Teaching ML to a Variety of Populations
		13.3.1 Machine Learning for Data Science Majors and Allied Majors
		13.3.2 Machine Learning for Non-major Students
		13.3.3 Machine Learning for ML Users
	13.4 Framework Remarks for ML Education
		13.4.1 Statistical Thinking
		13.4.2 Interdisciplinary Projects
		13.4.3 The Application Domain Knowledge
	13.5 Conclusion
	References
14 Core Concepts of Machine Learning
	14.1 Introduction
	14.2 Types of Machine Learning
	14.3 Machine Learning Parameters and Hyperparameters
	14.4 Model Training, Testing, and Validation
	14.5 Machine Learning Performance Indicators
	14.6 Bias and Variance
	14.7 Model Complexity
	14.8 Overfitting and Underfitting
	14.9 Loss Function Optimization and the Gradient Descent Algorithm
	14.10 Regularization
	14.11 Conclusion
	References
15 Machine Learning Algorithms
	15.1 Introduction
	15.2 K-nearest Neighbors
	15.3 Decision Trees
	15.4 Perceptron
	15.5 Linear Regression
	15.6 Logistic Regression
	15.7 Neural Networks
	15.8 Conclusion
	References
16 Teaching Methods for Machine Learning
	16.1 Introduction
	16.2 Visualization
	16.3 Hand-On Tasks
		16.3.1 Hands-On Task for the KNN Algorithm
		16.3.2 Hands-On Task for the Perceptron Algorithm
		16.3.3 Hands-On Task for the Gradient Descent Algorithm
		16.3.4 Hands-On Task for Neural Networks
	16.4 Programming Tasks
	16.5 Project-Based Learning
	16.6 Conclusion
	References
Part V Frameworks for Teaching Data Science
17 Data Science for Managers and Policymakers
	17.1 Introduction
	17.2 Workshop for Policymakers in National Education Systems
		17.2.1 Workshop Rationale and Content
		17.2.2 Workshop Schedule
		17.2.3 Group Work Products
		17.2.4 Workshop Wrap-Up
	17.3 Conclusion
	References
18 Data Science Teacher Preparation: The “Method for Teaching Data Science” Course
	18.1 Introduction
	18.2 The MTDS Course Environment
	18.3 The MTDS Course Design
	18.4 The Learning Targets and Structure of the MTDS Course
	18.5 Grading Policy and Submissions
	18.6 Teaching Principles of the MTDS Course
	18.7 Lesson Descriptions
		18.7.1 Lesson 6
		18.7.2 Mid-Semester Questionnaire
		18.7.3 Lesson 7
	18.8 Conclusion
	References
19 Data Science for Social Science and Digital Humanities Research
	19.1 Introduction
	19.2 Relevance of Data Science for Social Science and Digital Humanities Researchers
	19.3 Data Science Bootcamps for Researchers in Social Sciences and Digital Humanities
		19.3.1 Applicants and Participants of Two 2020 Bootcamps for Researchers in Social Sciences and Digital Humanities
		19.3.2 The Design and Curriculum of the Data Science for Social Science and Digital Humanities Researchers Bootcamp
	19.4 Data Science for Psychological Sciences
		19.4.1 The Computer Science for Psychological Science Course
		19.4.2 The Data Science for Psychology Science Course
	19.5 Data Science for Social Sciences and Digital Humanities, from a Motivation Theory Perspective
		19.5.1 The Self-determination Theory
		19.5.2 Gender Perspective
	19.6 Conclusion
	References
20 Data Science for Research on Human Aspects of Science and Engineering
	20.1 Introduction
	20.2 Examples of Research Topics Related to Human Aspects of Science and Engineering that Can Use Data Science Methods
	20.3 Workshop on Data Science Research on Human Aspects of Science and Engineering
		20.3.1 Workshop Rationale
		20.3.2 Workshop Contents
		20.3.3 Target Audience
		20.3.4 Workshop Framework (in Terms of weeks)—A Proposal
		20.3.5 Prerequisites
		20.3.6 Workshop Requirements and Assessment
		20.3.7 Workshop Schedule and Detailed Contents
		20.3.8 Literature (For the Workshop)
	20.4 Conclusion
Epilogue
Index