2 A Wofkflow Apporach to Learning
There are many ways you could be learning social network analysis! In fact, there a multiple books on social network analysis that take a different approach to teaching it. We could focus on social theories of interaction and apply them to structural analyses like networks. Or, we could focus on graph theory and explore that. In this book, however, we are going to focus on a workflow of data. This approach to learning is data-forward. By workflow, I am referring to an organised approach to working with messy data all the way to analysing and communicating it. As such, each module of the book reflects a step of this process. By data-forward, I mean that I want to peel back the curtain for you and give you an insight into how you can build confidence working with an alternative data type such as network data. This book, then, is designed to help you work through the steps of building a project start to finish. As such, I draw on a project-based teaching and learning framework.
On The ADAPT Model
The pedagogical approach of this book is patterned after the ADAPT model championed by North Carolina State University’s Data Science and AI Academy (DSA). This model stands for “All-campus Data science and AI through Project-based Teaching and learning.” This model serves as a framework for the workflow of this book and it’s pedagogical ethos.
For a little context, all instructors at the Data Science and AI Academy are trained to teach using this model. The Academy’s data science courses are open to all students of any discipline and professional stage. Students include undergraduate, graduate university attendees as well as staff, faculty, and community members. With such a diverse student body, the DSA uses the ADAPT teaching model to create a consistent form of instruction across all courses. In a way, the teaching model threads all students together, weaving their courses into a cohesive experience that emboldens all students to participate. At its core, then, the model is designed to make computational skills accessible, regardless of discipline.
Students are the beating heart of the Data Science and AI Academy. As such, this model emphasises student agency as much as possible within the confines of the unique course objectives. Following a project-based pedagogy, students are encouraged to take ownership for their learning and, in line with the DSA’s interdisciplinarity, apply their learning to their interests. For example, students are encouraged to select their own dataset based on their interests and use it for their final project. Instructors are guided to introduce their students to different datasets and techniques to appeal to such students in hopes to encourage all students to see themselves in the courses.
In this spirit of this model, you’ll see that I very much lean on its principles (discussed below) and project-building nature. For example, throughout the book we will use some fun data (e.g., Harry Potter, UK Grime music, etc.) to teach network principles and R skills. I find these appeal to a very broad student body. I also teach based on concepts more than endlessly explaining code. Understanding the ‘what’ and the ‘why’ behind social network analysis more than explaining every syntactical function, I find, facilitates learning for many. As such, you will build competence in the process of working with social networks alongside exposure to code to help you execute effective analysis. For students more interested in the nature of the code or the mathematics behind the measurements, I extend more resources on an individual basis. In this book, I discuss these in the concluding chapters. So, you can think of this book as an introductory course and an application of social network data and the ADAPT teaching and learning approach.
ADAPT for Learners
Students, the biggest thing you need to know is that you will be working on a project for the duration of the course. For more details see the Final Project Instructions.
Each unit and assignment is a milestone in your project development. For this project, you may select any social network data that appeals to you. I have collated quite a few from various sources available on the Github repo. Alternatively, you may wish to explore other network data (from places like the network repository, Kaggle or others). Any data that you use, make sure you cite where you are using it from!
You do not necessarily need to know the details of the ADAPT model much beyond this. However, if you want to learn more about it, keep going!
ADAPT for Instructors
I will start by saying that the teaching model is a living document that the Data Science and AI team, headed by Rachel Levy, are constantly working on. As such, at the time of publication this is the current version of this model. You may wish to read a little more on any updates by checking with the NCSU Data Science and AI Academy’s website. That being said, here is a quick introduction to it’s ethos.
A Brief Overview of ADAPT
As previously mentioned, the students are at the core of this model as is the belief of the DSA that data science acumen can be taught to people from every disciplinary background. In fact, it behooves all to engage with data and become data literate (understanding how to interpret data) in a data-driven world! I do not go into great detail here but suffice it to say that the ADAPT model comprises three parts:
Project-based Learning
10 Common Learning Elements
Workforce Preparedness
The project-based teaching and learning component is central to this model and to this book. This component goes hand-in-hand with Workforce Preparedness. As an instructor, it is up to you to construct an environment wherein the project of your class can be suitable for students with a variety of career orientations. I encourage you to be flexible with the nature of the project by designing a rubric that is applicable various projects. For more on this model, visit https://datascienceacademy.ncsu.edu/.
ADAPT and This Book
This book was created from a course taught at the DSA under the direction of the ADAPT model. It’s goal is not to serve as an exemplar of the model, but rather to encourage instructors to engage with the “spirit” of the model. Consider teaching project-based courses that centres the interests of your students - maximising their agency. Further, consider teaching computational skills in such a way as to open its content to all interested parties.
Each of the units and modules in this book are designed with students’ project in mind. I discuss element of the ADAPT model in the Unit introduction chapters. The purpose of this book is to enact these principles and enable other instructors wishing to teach social network analysis to do the same.
All data used in this book is available using the “ADAPTSNA” R package. Alternatively, you may wish to use other data and just use this guide to inform your own syllabus and activities. If you use data from the ADAPTSNA package, please be sure to cite it. After installing and loading the package, run citation(“ADAPTSNA”) and it will provide the appropriate format.
Finally, let me point you to the assignments in this course. You are welcome to use or to change these assignments as you wish. They are designed to build a student’s skills throughout the semester as they progress towards their final project. Courses taught using the ADAPT model do not have quizzes or tests, but rather scaffold multiple assignments throughout the semester aimed at building a capstone project at the end of the course. This scaffolding approach enables the student to build their progress continually and the assignments serve as milestones in their progress. The material in this book is designed to fill a 1 credit-hour course. If yours is a longer course, you may wish to alter the pacing of the semester adding or rearranging the content of this course. Regardless here is a little insight into the assignments I use and why I think these milestones are useful to the student’s success and for you to grade the student.
Assignment 1 (10 marks) acts as an orientation to network principles and parlance. This, no-tech/low-tech assignment sees students create an ego-network, their ego-network. The aim of this assignment is to get them familiar with some of the vocabulary of network analysis.
Assignment 2 (10 marks) is the first milestone of their project. This no-tech assignment sees students present ideas for their final project encouraging them to present two separate ideas for datasets once they have either looked through the GitHub repository or discovered their own dataset. This serves as the beginning of a discussion between the instructor and student to guide their final project.
Assignment 3 (20 marks) is the second milestone of their project. This is designed to help students bring their data into R, describe it, and suggest any cleaning that they may need to do. By this point, they are encouraged to select their data that they are going to use.This is an opportunity for the instructor to give more in-depth feedback on the data they are using and suggest further transformations that they may need.
Assignment 4 (20 marks) is the third milestone of their project. This is an opportunity for instructors to provide guidance on producing useful visualisations.
The Final Project (40 marks) is the last milestone of their project. This an exploratory data analysis where students learn to use their exploratory visualisations and transformations to form and answer questions about their network.