Data Science in Education Using R

How to Contribute

Reading the Book

We wrote this book for you and are excited to share it! You can read the work in progress at datascienceineducation.com right now. The print version will be released in 2020 through Routledge.

The Aims of This Book

School districts, government agencies, and education businesses are generating data at a dizzying pace and serving it to teachers, administrators, and education consultants in a mind-boggling variety of formats. Educators and educational data practitioners who want to use data to improve the educational outcomes of students often have a clear idea of the questions they want to ask of their data, but they are left to analyze data as it is presented to them, often times using high-cost proprietary systems. Educational data rarely comes in a “ready-to-analyze” format, so educators and educational data practitioners who are eager to leverage data to promote student success often feel very little connection between the analytic questions they have and the numbers on their laptop. But some educational data practitioners are adopting the tools of data science, including open source projects like R, to make better use of the data deluge. When data science meets education, the numbers previously confined to websites and PDF reports are set free. Teachers, administrators, and consultants apply programming and statistics to prepare data, transform it, visualize it, and analyze it to answer questions that are as unique as are their roles in education.

Data science represents the intersection of domain-level expertise, statistics, and computer programming, applied to data as a means to answer research questions and solve problems. Making sense of data requires, among other things, collaboration, data processing (i.e., wrangling or munging), visualization, and communication of both processes and results in transparent and reproducible ways. Our book focuses on data science in education, which we define as using data science techniques like preparing, exploring, visualizing, and modeling data, in order to support schooling at all levels. Our book advances the larger conversation about data science by introducing the idea that the application of data science in a specific field–in this case, education–requires an exploration of unique challenges and the development of unique language. We feel that discussing data science using scenarios that are familiar to education professionals at all levels of education more effectively addresses the needs of those professionals–data analysts and others–who work in that field. This concept of data science in education is separate from data science education, which seeks to teach the broader techniques of data science while not necessarily teaching the unique application of those techniques in any particular industry. Educators have different needs than data science enthusiasts who aim to self-teach using data science education materials. As educational technology transforms both the administrative and student-facing sides of education, it will become increasingly important for education professionals - not just people hired to analyze data - to be able to understand and respond to the data they gather. Our book empowers educators from elementary school to higher education to transform the educational data they are immersed in into actionable insights, in order to help them better serve the students and institutions for whom they work. It could be used as a main textbook in a graduate data science in education course. Alternatively, it could be used as a supplementary textbook for individuals looking to expand their professional toolkit and become more proficient in data science techniques.

By the end of this book the reader will understand:

The diversity of data analysis skills and applications in the education field
Special considerations that come with analyzing education data
That good data analysis has a basic workflow and how they might implement such a workflow
The wonderful opportunity we have to shape the usefulness of data science in our education jobs

And, the reader will be able to:

Reflect and determine what their role is as a data analyst within their role as an educator
Identify and apply solutions to education data’s unique challenges, such as cleaning datasets and working with private student data
Apply a basic analytic workflow through practice with education datasets
Be thoughtful, empathetic, and effective when introducing data science techniques in their education jobs

Chapters

Introduction: Data Science in Education
How to Use This Book
What is a Data Scientist in Education?
Special Considerations
Getting Started with R and R Studio
Foundational Skills
Walkthrough 1: The Education Dataset Science Pipeline With Online Science Class Data
Walkthrough 2: Approaching Gradebook Data From a Data Science Perspective
Walkthrough 3: Introduction to Aggregate Data
Walkthrough 4: Longitudinal Analysis With Federal Students With Disabilities Data
Walkthrough 5: Text Analysis With Social Media Data
Walkthrough 6: Exploring Relationships Using Social Network Analysis Models and Methods
Walkthrough 7: The Role (and Usefulness) of Multi-Level Models
Walkthrough 8: Predicting Students’ Final Grades Using Machine Learning Methods
Introducing Data Science Tools To Your Education Job
Teaching Data Science
Learning More
Additional Resources
Conclusion

Contributing

This project started in the #dataedu Slack channel. You can join the workspace here.

Community members can contribute by making changes through a pull request. We encourage community members to do their pull requests on separate branches. We're actively editing as we get closer to our manuscript's due date and this helps us keep all the changes synced up.

Git Issue Labels

To help contributors participate, we're using labels so community members can identify tasks they want to help with. When working on an issue, assign yourself to the issue. This helps us keep track of the work and lets us know who to contact for more collaboration. The labels are:

good first issue: These are requests for changes that we think would be fun and achievable if you're new to git and GitHub.
discussion: Sometimes we need help talking through a topic to help us make a good design choice for our readers. These issues won't always result in a change, but they help us clarify what's best for the final product.
test code: These issues are for running code and giving feedback about how it went. If there were problems, you can help us by letting us know what happened.
bug: The code isn't running as expected and needs fixing.
help wanted: Need help getting code to run or writing a section. We'll make sure the problem we're working on is clearly described in the issue.
writing: New content needed. At least one author will be assigned to writing issues, but we welcome collaboration! Feel free to message the author on Slack or in the issue comments to coordinate.
review draft: These are requests to read through a draft chapter and provide feedback on the experience, including reability.

Contact Us

If you have questions, comments, or ideas you can reach the authors by email at authors@datascienceineducation.com or on Twitter:

Emily @ebovee09
Isabella @ivelasq3
Jesse @kierisi
Joshua @jrosenberg6432
Ryan @RyanEs

kunken / data-science-in-education