Data Science, Statistics and Artificial Intelligence
Free to read online. For a copy of this book, or to quote, highlight or reproduce any part of it, please request permission through the contact page beforehand.
Publication and Educational Use Note
This document is an educational manuscript-development draft prepared for review, expansion, fact-checking, and editorial development. It is not a substitute for professional advice, formal laboratory risk assessment, or institution-specific safeguarding procedures. Experiments and projects should be adapted to learners' ages, local requirements, available equipment, and appropriate supervision. Scientific claims and references should be checked against current authoritative sources before publication.
The proposed author line is reproduced from the series brief. Author credentials, permissions, third-party materials, and publication rights should be verified before commercial or public release.
About This Volume
Provide a structured educational foundation in data science, statistics and artificial intelligence, combining conceptual understanding, practical application, critical analysis and responsible decision-making.
Intended level: Intermediate to advanced. The volume is designed to be adapted for different learner ages, institutional settings and levels of prior knowledge.
How to Use This Book
Read each chapter actively. Define unfamiliar terms, answer the review questions, maintain a record of assumptions and evidence, and complete activities only in suitable, authorised and supervised settings. Teachers may adapt the depth, time and assessment criteria to local curricula.
Series-wide learning principles
- Conceptual understanding: explain ideas in your own words and connect them to prior learning.
- Evidence and reasoning: distinguish observation, inference, model and conclusion.
- Practical application: test ideas through safe activities, design challenges, data analysis or case studies.
- Ethical responsibility: consider safety, accessibility, privacy, sustainability and effects on communities.
- Communication: report methods, evidence, uncertainty and limitations clearly.
Contents
- Chapter 1. Data Science Questions and Workflows
- Chapter 2. Data Collection, Quality and Governance
- Chapter 3. Cleaning, Transformation and Documentation
- Chapter 4. Exploratory Analysis and Visualisation
- Chapter 5. Probability, Statistics and Inference
- Chapter 6. Machine Learning Fundamentals
- Chapter 7. Model Evaluation, Validation and Drift
- Chapter 8. Generative AI and Responsible Use
- Chapter 9. Bias, Privacy, Explainability and Governance
- Chapter 10. Applied Data Science Capstone
Additional sections: Learning outcomes, practical activities, assessment tasks, glossary, and references.
Data Science Questions and Workflows
Core explanation
Data science turns questions into evidence through data acquisition, preparation, analysis, modelling and communication. The quality of a conclusion depends on data provenance, assumptions, evaluation and whether uncertainty and limitations are explained.
Key concepts and connections
Core ideas
This topic connects disciplinary knowledge with methods, tools and decisions relevant to data science questions and workflows. Learners should define key terms, identify assumptions and distinguish established evidence from unresolved questions.
Methods and evidence
Select methods that fit the question. Document inputs, procedures, measurement conditions and limitations; use comparisons and validation where appropriate.
Responsible application
Consider safety, accessibility, environmental impact, privacy where relevant, lifecycle cost and the needs of affected communities.
Worked example or case study
A data project related to data science questions and workflows should document data sources, define variables, check missingness and bias, select an appropriate method and communicate the limits of any model.
Practical activity
Use a small, public or fictional dataset. Write a data dictionary, document cleaning decisions, create a suitable visualisation and explain one potential source of bias.
Review and discussion questions
- What are the central concepts in data science questions and workflows?
- Which methods would produce relevant and trustworthy evidence?
- What assumptions or limitations could affect the conclusion?
- How should safety, ethics and practical constraints influence implementation?
Data Collection, Quality and Governance
Core explanation
Data science turns questions into evidence through data acquisition, preparation, analysis, modelling and communication. The quality of a conclusion depends on data provenance, assumptions, evaluation and whether uncertainty and limitations are explained.
Key concepts and connections
Core ideas
This topic connects disciplinary knowledge with methods, tools and decisions relevant to data collection, quality and governance. Learners should define key terms, identify assumptions and distinguish established evidence from unresolved questions.
Methods and evidence
Select methods that fit the question. Document inputs, procedures, measurement conditions and limitations; use comparisons and validation where appropriate.
Responsible application
Consider safety, accessibility, environmental impact, privacy where relevant, lifecycle cost and the needs of affected communities.
Worked example or case study
A data project related to data collection, quality and governance should document data sources, define variables, check missingness and bias, select an appropriate method and communicate the limits of any model.
Practical activity
Use a small, public or fictional dataset. Write a data dictionary, document cleaning decisions, create a suitable visualisation and explain one potential source of bias.
Review and discussion questions
- What are the central concepts in data collection, quality and governance?
- Which methods would produce relevant and trustworthy evidence?
- What assumptions or limitations could affect the conclusion?
- How should safety, ethics and practical constraints influence implementation?
Cleaning, Transformation and Documentation
Core explanation
Data science turns questions into evidence through data acquisition, preparation, analysis, modelling and communication. The quality of a conclusion depends on data provenance, assumptions, evaluation and whether uncertainty and limitations are explained.
Key concepts and connections
Core ideas
This topic connects disciplinary knowledge with methods, tools and decisions relevant to cleaning, transformation and documentation. Learners should define key terms, identify assumptions and distinguish established evidence from unresolved questions.
Methods and evidence
Select methods that fit the question. Document inputs, procedures, measurement conditions and limitations; use comparisons and validation where appropriate.
Responsible application
Consider safety, accessibility, environmental impact, privacy where relevant, lifecycle cost and the needs of affected communities.
Worked example or case study
A data project related to cleaning, transformation and documentation should document data sources, define variables, check missingness and bias, select an appropriate method and communicate the limits of any model.
Practical activity
Use a small, public or fictional dataset. Write a data dictionary, document cleaning decisions, create a suitable visualisation and explain one potential source of bias.
Review and discussion questions
- What are the central concepts in cleaning, transformation and documentation?
- Which methods would produce relevant and trustworthy evidence?
- What assumptions or limitations could affect the conclusion?
- How should safety, ethics and practical constraints influence implementation?
Exploratory Analysis and Visualisation
Core explanation
Data science turns questions into evidence through data acquisition, preparation, analysis, modelling and communication. The quality of a conclusion depends on data provenance, assumptions, evaluation and whether uncertainty and limitations are explained.
Key concepts and connections
Core ideas
This topic connects disciplinary knowledge with methods, tools and decisions relevant to exploratory analysis and visualisation. Learners should define key terms, identify assumptions and distinguish established evidence from unresolved questions.
Methods and evidence
Select methods that fit the question. Document inputs, procedures, measurement conditions and limitations; use comparisons and validation where appropriate.
Responsible application
Consider safety, accessibility, environmental impact, privacy where relevant, lifecycle cost and the needs of affected communities.
Worked example or case study
A data project related to exploratory analysis and visualisation should document data sources, define variables, check missingness and bias, select an appropriate method and communicate the limits of any model.
Practical activity
Use a small, public or fictional dataset. Write a data dictionary, document cleaning decisions, create a suitable visualisation and explain one potential source of bias.
Review and discussion questions
- What are the central concepts in exploratory analysis and visualisation?
- Which methods would produce relevant and trustworthy evidence?
- What assumptions or limitations could affect the conclusion?
- How should safety, ethics and practical constraints influence implementation?
Probability, Statistics and Inference
Core explanation
Data science turns questions into evidence through data acquisition, preparation, analysis, modelling and communication. The quality of a conclusion depends on data provenance, assumptions, evaluation and whether uncertainty and limitations are explained.
Key concepts and connections
Core ideas
This topic connects disciplinary knowledge with methods, tools and decisions relevant to probability, statistics and inference. Learners should define key terms, identify assumptions and distinguish established evidence from unresolved questions.
Methods and evidence
Select methods that fit the question. Document inputs, procedures, measurement conditions and limitations; use comparisons and validation where appropriate.
Responsible application
Consider safety, accessibility, environmental impact, privacy where relevant, lifecycle cost and the needs of affected communities.
Worked example or case study
A data project related to probability, statistics and inference should document data sources, define variables, check missingness and bias, select an appropriate method and communicate the limits of any model.
Practical activity
Use a small, public or fictional dataset. Write a data dictionary, document cleaning decisions, create a suitable visualisation and explain one potential source of bias.
Review and discussion questions
- What are the central concepts in probability, statistics and inference?
- Which methods would produce relevant and trustworthy evidence?
- What assumptions or limitations could affect the conclusion?
- How should safety, ethics and practical constraints influence implementation?
Machine Learning Fundamentals
Core explanation
Data science turns questions into evidence through data acquisition, preparation, analysis, modelling and communication. The quality of a conclusion depends on data provenance, assumptions, evaluation and whether uncertainty and limitations are explained.
Key concepts and connections
Core ideas
This topic connects disciplinary knowledge with methods, tools and decisions relevant to machine learning fundamentals. Learners should define key terms, identify assumptions and distinguish established evidence from unresolved questions.
Methods and evidence
Select methods that fit the question. Document inputs, procedures, measurement conditions and limitations; use comparisons and validation where appropriate.
Responsible application
Consider safety, accessibility, environmental impact, privacy where relevant, lifecycle cost and the needs of affected communities.
Worked example or case study
A data project related to machine learning fundamentals should document data sources, define variables, check missingness and bias, select an appropriate method and communicate the limits of any model.
Practical activity
Use a small, public or fictional dataset. Write a data dictionary, document cleaning decisions, create a suitable visualisation and explain one potential source of bias.
Review and discussion questions
- What are the central concepts in machine learning fundamentals?
- Which methods would produce relevant and trustworthy evidence?
- What assumptions or limitations could affect the conclusion?
- How should safety, ethics and practical constraints influence implementation?
Model Evaluation, Validation and Drift
Core explanation
Data science turns questions into evidence through data acquisition, preparation, analysis, modelling and communication. The quality of a conclusion depends on data provenance, assumptions, evaluation and whether uncertainty and limitations are explained.
Key concepts and connections
Core ideas
This topic connects disciplinary knowledge with methods, tools and decisions relevant to model evaluation, validation and drift. Learners should define key terms, identify assumptions and distinguish established evidence from unresolved questions.
Methods and evidence
Select methods that fit the question. Document inputs, procedures, measurement conditions and limitations; use comparisons and validation where appropriate.
Responsible application
Consider safety, accessibility, environmental impact, privacy where relevant, lifecycle cost and the needs of affected communities.
Worked example or case study
A data project related to model evaluation, validation and drift should document data sources, define variables, check missingness and bias, select an appropriate method and communicate the limits of any model.
Practical activity
Use a small, public or fictional dataset. Write a data dictionary, document cleaning decisions, create a suitable visualisation and explain one potential source of bias.
Review and discussion questions
- What are the central concepts in model evaluation, validation and drift?
- Which methods would produce relevant and trustworthy evidence?
- What assumptions or limitations could affect the conclusion?
- How should safety, ethics and practical constraints influence implementation?
Generative AI and Responsible Use
Core explanation
Data science turns questions into evidence through data acquisition, preparation, analysis, modelling and communication. The quality of a conclusion depends on data provenance, assumptions, evaluation and whether uncertainty and limitations are explained.
Key concepts and connections
Core ideas
This topic connects disciplinary knowledge with methods, tools and decisions relevant to generative ai and responsible use. Learners should define key terms, identify assumptions and distinguish established evidence from unresolved questions.
Methods and evidence
Select methods that fit the question. Document inputs, procedures, measurement conditions and limitations; use comparisons and validation where appropriate.
Responsible application
Consider safety, accessibility, environmental impact, privacy where relevant, lifecycle cost and the needs of affected communities.
Worked example or case study
A data project related to generative ai and responsible use should document data sources, define variables, check missingness and bias, select an appropriate method and communicate the limits of any model.
Practical activity
Use a small, public or fictional dataset. Write a data dictionary, document cleaning decisions, create a suitable visualisation and explain one potential source of bias.
Review and discussion questions
- What are the central concepts in generative ai and responsible use?
- Which methods would produce relevant and trustworthy evidence?
- What assumptions or limitations could affect the conclusion?
- How should safety, ethics and practical constraints influence implementation?
Bias, Privacy, Explainability and Governance
Core explanation
Data science turns questions into evidence through data acquisition, preparation, analysis, modelling and communication. The quality of a conclusion depends on data provenance, assumptions, evaluation and whether uncertainty and limitations are explained.
Key concepts and connections
Core ideas
This topic connects disciplinary knowledge with methods, tools and decisions relevant to bias, privacy, explainability and governance. Learners should define key terms, identify assumptions and distinguish established evidence from unresolved questions.
Methods and evidence
Select methods that fit the question. Document inputs, procedures, measurement conditions and limitations; use comparisons and validation where appropriate.
Responsible application
Consider safety, accessibility, environmental impact, privacy where relevant, lifecycle cost and the needs of affected communities.
Worked example or case study
A data project related to bias, privacy, explainability and governance should document data sources, define variables, check missingness and bias, select an appropriate method and communicate the limits of any model.
Practical activity
Use a small, public or fictional dataset. Write a data dictionary, document cleaning decisions, create a suitable visualisation and explain one potential source of bias.
Review and discussion questions
- What are the central concepts in bias, privacy, explainability and governance?
- Which methods would produce relevant and trustworthy evidence?
- What assumptions or limitations could affect the conclusion?
- How should safety, ethics and practical constraints influence implementation?
Applied Data Science Capstone
Core explanation
Data science turns questions into evidence through data acquisition, preparation, analysis, modelling and communication. The quality of a conclusion depends on data provenance, assumptions, evaluation and whether uncertainty and limitations are explained.
Key concepts and connections
Core ideas
This topic connects disciplinary knowledge with methods, tools and decisions relevant to applied data science capstone. Learners should define key terms, identify assumptions and distinguish established evidence from unresolved questions.
Methods and evidence
Select methods that fit the question. Document inputs, procedures, measurement conditions and limitations; use comparisons and validation where appropriate.
Responsible application
Consider safety, accessibility, environmental impact, privacy where relevant, lifecycle cost and the needs of affected communities.
Worked example or case study
A data project related to applied data science capstone should document data sources, define variables, check missingness and bias, select an appropriate method and communicate the limits of any model.
Practical activity
Use a small, public or fictional dataset. Write a data dictionary, document cleaning decisions, create a suitable visualisation and explain one potential source of bias.
Review and discussion questions
- What are the central concepts in applied data science capstone?
- Which methods would produce relevant and trustworthy evidence?
- What assumptions or limitations could affect the conclusion?
- How should safety, ethics and practical constraints influence implementation?
End-of-Volume Assessment
Use the following tasks to demonstrate understanding. Instructors may adapt marks and deadlines to the course level.
- Concept explanation: explain three major concepts from the volume and connect each to a practical example.
- Evidence analysis: evaluate a claim, dataset, case study or model. State what the evidence supports and what remains uncertain.
- Applied task: complete one practical activity or design challenge and document method, results and limitations.
- Ethics and responsibility: identify relevant safety, ethical, accessibility or sustainability considerations.
- Communication: prepare a short report or presentation with clear terminology, appropriate visuals and references.
Suggested assessment rubric
Glossary
Evidence: Information used to support, challenge or refine a claim.
Model: A simplified representation of a system used to explain, analyse or predict behaviour.
Variable: A characteristic or quantity that can take different values.
Uncertainty: A description of limits in knowledge or measurement.
Validation: Assessment of whether a method, model or solution is suitable for its intended purpose.
Reproducibility: The ability to follow documented methods and obtain consistent analytical results under defined conditions.
Ethics: Principles used to evaluate responsibilities, harms, fairness and appropriate conduct.
Sustainability: Meeting needs while considering long-term environmental, social and economic consequences.
References and Further Reading
The following are starter references for editorial development. Verify editions, URLs, publication details and relevance; add chapter-specific sources and APA 7 in-text citations throughout before publication.
- National Institute of Standards and Technology. (2023). Artificial intelligence risk management framework (AI RMF 1.0). https://doi.org/10.6028/NIST.AI.100-1
- National Academies of Sciences, Engineering, and Medicine. (2019). Reproducibility and replicability in science. The National Academies Press. https://doi.org/10.17226/25303
Editorial Development Checklist
- Verify scientific and technical claims against current authoritative sources.
- Add APA 7 in-text citations to claims, frameworks, definitions and empirical findings that require support.
- Confirm that all figures, images, tables and excerpts are original, licensed or used with permission.
- Adapt safety notes and practical activities to learner age, local rules and institutional risk assessments.
- Check accessibility, inclusive language, terminology consistency and cross-references.
- Obtain subject-matter review and copy-editing before publication.