Week 2 Basic Concepts

In this week, we will:

  1. Explore social network perspectives
  2. Become familiar with basic SNA concepts
  3. Start to draft an initial SNA project idea

As always, please read on for details.

2.1 Understanding SNA in Research

2.1.1 Three levels of considerations

In the previous video, I described three levels of considerations that educational researchers often need to be aware of. You could find more information in text such as (Niglas 2010) or some other research methodology courses. Please note that this framework was constructed to help us grapple with the complex terrain of research, and it is highly debatable.

Below, I try to re-iterate the key message of the video – but from the bottom up:

  • m, or methods/techniques: The “small m” in SNA constitutes methods or techniques we apply in SNA research. Imagine we are using SNA to investigate friendship of a network of high-schoolers (think about “Gossip Girls” if you’ve watched that TV series). A technique in this SNA research could be a questionnaire used to collect friendship data among students; it could be the force-directed layout we use to visualize this network; it could be the measure of betweenness centrality we use to characterize high-schoolers; it could also be a network modeling algorithm we apply to model the flow of gossips. In a nutshell, these techniques are more about what we concretely do in an SNA research.

  • M: When SNA is referred to as a “big M” Methodology, it is treated as a systematic approach of investigating a phenomenon. Beyond simply applying these techniques, a methodology is also concerned with why a technique gets used. In other words, understanding SNA as a methodology means learning to make informed decisions in any stage of an SNA project. For example, why using a questionnaire instead of observations or interviews? In which cases should one use a circle layout instead of a force-directed layout? Why a specific SNA measure is appropriate for addressing a research question? In a nutshell, the big M is concerned with how knowledge could be best gained by following many SNA methodologists and researchers have created so far.

  • P, worldviews, philosophical schools of thoughts, paradigms: In SNA, some scholars go further to argue SNA offers a unique way of “seeing the world.” In Carolan (2014) chapter 2 you will read about the relational perspective that represent a particular worldview that emphasizes relations instead of attributes. You will also read about relational realism that is referred to as an ontology grounding SNA. To a great extent, SNA offers a new research paradigm. As put by Barry Wellman, a guru in SNA from the University of Toronto, “It is a comprehensive paradigmatic way of taking social structure seriously by studying directly how patterns of ties allocate resources in a social system” (see, Carolan, 2014, p. 33).

In this course, we are mostly concerned with the “big M” level. We will not dive too deep into the P level, and we will not settle with specific techniques. Together, we will learn how to apply various techniques to systematically produce knowledge about a phenomenon.

2.2 Basic Concepts

Last week, you’ve identified a wide range of terms from course readings. This week, we will be immersed in a number of key SNA concepts.

In the following video, I “glide over” some basic terms that are often used in SNA. You do not need to memorize them all. As a matter of fact, I see those terms existing as a network (see the image below); as you “unlock” one term, you are also activating others. So spend time on some terms, and you’re implicitly learning about others.

Small World of Words
(Credit: Small World of Words project)

2.3 Week 2 Activities

2.3.1 Read & Annotate


  • Grunspan, Wiggins, and Goodreau (2014): link
  • Carolan, chapters 2-3: ch. 2 (starting from “The Integration of Theory and Method”) and ch. 3

Pro Tip:

  • Please use the links I provided above. Please also annotate the webpage directly instead of PDFs. Otherwise, our annotations/discussions may get scattered across different pages.

Annotate using Hypothes.is. Even though I do not require you to make a specific number of annotations, I encourage you to accomplish the following “ABC” this week:

  1. Ask a question
  2. Brag about your understanding about an SNA term, a domain-specific theory, a cool tool, etc.
  3. When you reply to one peer, Connect another peer’s ideas in your annotation

2.3.2 Start to put together an intial project idea

I see most of you already having fascinating project ideas. Even though we are only in Week 2, I encourge you to take your ideas to the next level when you engage with our readings and annotations.

Specifically, you can consider the following: (1) what is the central problem of your research/practice and why it is worth investigating, (2) what phenomenon are you examining, in which context, and what key question(s) you have about the phenomenon, and (3) why SNA is potentially fruitful based on your current understanding.

We will share out our initial project ideas by the end of Week 4.

2.3.3 Install R/RStudio and Play with Lab 1

There are many software packages for SNA, and you are certainly encouraged to explore different options. Gephi is a good option for creating network visualizations. Ucinet is broadly used in academic circles. NodeXL is an Excel add-in that allows you to generate network visualizations, compute network measures, and even to collect Twitter data. NetworkX is great if you are familiar with the Python programming language.

This course will focus on supporting your learning of network analysis in R, a free software environment for statistical computing and graphics. I understand it might be a bit intimidating for some learners to think about programming, but it’s really not that difficult to get started and the benefits are unlimited. If you have not used R (or other programming languages) before, rest assured that I will guide you through the process.

In the next two weeks, we will focus on getting started with R and network analysis using a R package named igraph. Below are three concrete steps, together with some materials, for us to get started. Please post questions in the 2021-questions channel whenever you get stuck.

Step 1: Install R and RStudio.

Step 2: Install the igraph package

Step 3: Work through Lab 1

### Lab 1: Read a social network dataset and make a graph
### Source of dataset: http://networkrepository.com/soc-karate.php

# Read data ---------------------------------------------------------------
# change the following line to point your Working Directory to the correct folder
# setwd("/tmp")

# read data from the .csv file
karate <- read.csv("soc-karate.csv")
karate <- as.matrix(karate)
# show the data in Console

# Load igraph and create a network object ---------------------------------

# load the library
# create a network/graph object
g <- graph_from_edgelist(karate, directed = FALSE)
# show the graph in console

# Plot the graph ----------------------------------------------------------

# plot using the default layout
# plot in a circle
plot(g, layout=layout_in_circle)

# Bonus: play with plotting settings
# check this link for ideas: https://www.r-graph-gallery.com/248-igraph-plotting-parameters.html
# plot in a cleaner way, esp. for larger graphs
  vertex.size = 2, # control the size of nodes
  edge.arrow.size = 0.001, # control the size of edges
  vertex.color = adjustcolor("white", alpha.f = 0),  # adjust the color of nodes
  vertex.label = NA # hide labels

A gentle reminder – we are NOT meeting synchronously on this coming Tuesday but will chat asychronously on Slack. Details will be shared on Monday, Feb 1.

Have a wonderful week!


Grunspan, Daniel Z., Benjamin L. Wiggins, and Steven M. Goodreau. 2014. “Understanding Classrooms Through Social Network Analysis: A Primer for Social Network Analysis in Education Research.” Edited by Erin Dolan. CBELife Sciences Education 13 (2): 167–78. https://doi.org/10.1187/cbe.13-08-0162.

Niglas, Katrin. 2010. “The Multidimensional Model of Research Methodology: An Integrated Set of Continua.” In SAGE Handbook of Mixed Methods in Social & Behavioral Research, edited by Abbas Tashakkori and Charles Teddlie, 215–36. Thousand Oaks: SAGE.