Social Media Data Analytics – Winter 2022
Instructor: Dr. Ashley Sanders (asandersgarcia@ucla.edu)
Meeting Times: 3-hour interactive lecture + weekly lab
Course Description
This course is a critical social media data analysis course that explores not only how to, but also why, study this type of data. This course moves beyond a business-minded functional understanding and analysis of social media data to engage with questions of power, privilege, identity, whose voices count and in what spaces, as well as how data science and DH may be used to challenge power structures. It considers how social media has been used both to undermine and to support social justice and political change movements, the ways in which social media data is currently used by corporate entities, and ethical data usage. Specifically, we will utilize computational methodologies from text analysis (qualitative data analytics), statistics, as well as data visualization to examine Twitter data to understand various communities and discourses circulating in society today and in the recent past. This interdisciplinary project-based class will help you develop social media analysis skills by employing user-friendly tools, such as Google Sheets/Excel, Tableau, Voyant Tools (text analysis), as well as Python 3.
Learning Outcomes
This course will guide you in developing fundamental digital research skills, including how to:
- Use different methods for analyzing, and exploring social media data for research and development purposes.
- Process collected data – primarily structured – using methods involving correlation, regression, and classification to derive insights about the sources and people who generated that data.
- Analyze unstructured data – primarily textual comments – for themes and trends.
- Visualize patterns in social media data
- Craft compelling, argument-driven, humanistic narratives about social media data sets.
Questions
We will be using Bruin Learn to communicate. Please post your general questions on Bruin Learn on Piazza so others can see the responses (i.e., questions regarding homework, projects, assignments, assessments, and expectations).
Piazza
This term we will be using Piazza for class discussion. The system is highly catered to getting you help fast and efficiently from classmates, the TA, and myself. Rather than emailing questions to the teaching staff, I encourage you to post your questions on Piazza. If you have any problems or feedback for the developers, email team@piazza.com.
Assignments
There is a 24-hour grace period available to everyone for every assignment. Please contact your TA to take advantage of this flexibility.
| Due Date | Assignment | Percentage |
| [periodically] | Quizzes | 15% |
| Final Team Project | 30% | |
| ~Weekly Assignments | 40% | |
| Individual reflections | 15% |
Course Schedule
| Week 1 | INTRODUCTION |
Learning Objectives: -Understand the difference between structured and unstructured data -Determine & employ appropriate analytical methods for structured and unstructured data. | Prior to 1st class Complete the Intro and Background Knowledge Survey Whole Class Activities Introduction to the course, materials, learning objectives, and projects Share final project output possibilities Lecture: What is Twitter? Structured data, taxonomy, folksonomy Example of what we can do with structured data from the survey Screening: The Social Dilemma Lab Work personality style quiz and reflection Explore Sentiments of Notre Dame project Begin to explore datasets and possible topics |
| Week 2 | |
Learning Objectives: -Describe social benefits & harm of social media-Employ data visualization techniques to explore social media data -Employ word frequency, dispersion, and KWIC methods to unstructured text | Homework Due T2.1 Submit team member names and topic Social Dilemma Quiz Update your operating system. Watch: The Social Dilemma (Netflix – $8.99 per month & you can cancel at any time) We will also screen this in class during Week 1. Review Voyant Tools overview Whole Class Activities Small Groups: Discuss Social Dilemma Mini-Lecture: Power of Social Media Data Workshop: How to scrape social media data with TWARC & TAGS Screening: Coded Bias Lab Select topic/Twitter data set Small Groups: Write down at least one question that you can ask about each column in your archive. Choose at least one to propose as the basis for a Twitter research project |
| Week 3 | Ethical Data Use and Reuse |
Learning Objectives: -Define what ethical data use/reuse means -Employ TF-IDF, collocation analysis, and a combination of tools to explore social textual data. -Understand and demonstrate how to split texts for analysis. | Homework Due Read: Restricted Uses of the Twitter APIs Watch Coded Bias (also on Netflix. We will screen this during Week 2) T3.1 Team Assignment 3.2 Individual Assignment 3.3 Coded Bias Quiz Whole Class Activities Q&A: Ethical data use Small Groups: Discuss Coded Bias Share out: Findings and visualizations from data set exploration Workshop: How to split texts Workshop: Collocations, TF-IDF, and combining tools Individual Reflection on findings, thoughts on the chosen data set so far, remaining questions, and changes they would like to make to their project direction Lab Upload your Twitter texts into Voyant and answer the following questions: What are the most common terms that appear in your tweets? Who do you retweet most often? How many of your tweets are retweeted? Why do you think this is the case? What do retweets indicate? Use each of the methods you’ve learned so far to continue exploring your research questions and data set. Post-Class: Quiz on Restricted uses of Twitter APIs |
| Week 4 | Subverting Social Media’s Divisive Tendencies |
Learning Objectives: -Describe social media’s use for connection and activism -Explain what topic modeling is and does -Employ topic modeling with mallet to identify themes and trends | Homework Due Watch: Cambridge Analytica whistleblower Brittany Kaiser: “The law doesn’t protect you” Listen (16min): Sheilah Kast & Jonna McKone, “The Politics of Archiving #BlackLivesMatter,” Maryland Morning with Sheilah Kast (15 May 2015). Read: Whitelaw Reid, “Black Twitter 101: What is it? Where did it originate? Where is it headed?” UVa Today (28 November 2018). Content Quiz Discussion Post: Explain one of the two methods presented in class and provide an example either from your own research or from another project. Then upvote the most instructive explanations of each of the methods. Whole Class Activities Discuss readings Mini-Lecture: Hashtag activism Workshop: Topic Modeling Lab Begin text preparation and topic modeling with data set |
| Week 5 | Twitter Sentiment Analysis |
Learning Objectives: -Define computational sentiment analysis -Describe possibilities and limitations of sentiment analysis -Employ computational sentiment analysis to study social media data | Homework Due Read: Jessica Uwoghiren, “The Year 2020: Analyzing Twitter Users’ Reflections using NLP: A Sentiment Analysis Project using Python and Tableau,” Towards Data Science (30 December 2020). Watch: Jake Tompkins, presentation on social media analysis. Content Quiz Individual Assignment: Use Voyant Tools, Python, R, or MALLET (topic modeling) to begin your qualitative research. Share 1 visualization or table that yields a significant insight into your dataset and 1 paragraph (> 4 sentences) of analysis that summarizes the findings from your chosen qualitative analysis method, as well as what they mean in the context of your research. Whole Class Activities Discuss two examples from Jessica Uwoghiren and Jake Tompkins Lecture: Opportunities and limitations of sentiment analysis Workshop: Sentiment analysis Lab In project teams, draft a question related to sentiment Apply sentiment analysis Python script to project data set and interpret the results |
| Week 6 | Refining questions and methods |
Learning Objectives: -Explain the links between social media and political conflict -Identify exemplary projects as models -Explain why and when network analysis is an effect method to answer questions related to social media | Homework Due Twitter Project checkpoint Whole Class Activities Team check-in survey Scavenger hunt for other Twitter projects that describe their methods and share out via Google Doc Mini-Lecture: How social media works and its relationship with conflict and democracy Workshop: Using the Twitter plugin for Gephi to scrape and visualize network data Lab Lightning round project checkpoint and feedback |
| Week 7 | Social Network Analysis |
Learning Objectives: -Describe the connections between social media and political conflict -Employ network analysis to answer questions of social media data | Homework Due 1st group of lightning talks on specific country case studies of social media, conflict, and democracy Whole Class Activities Lightning talks Individual and whole-group reflection on the talks Workshop: Computational network analysis Lab Project work time |
| Week 8 | Mapping Social Media Data |
Learning Objectives: -Pose and answer location-based questions of social media data -Be able to clean and structure geographical data for analysis -Describe the connections between social media and political conflict, as well as options for addressing this issue. | Homework Due Create a network visualization of your data and choose two quantitative measures to explore and write up an analysis of what each of the measures tells you about your dataset. 2nd group of lightning talks on specific country case studies of social media, conflict, and democracy. Whole Class Activities Lightning talks Workshop: Preparing social media data for location-based analysis Application: Clean, structure, and map at least a selection of project data set Lab Q&A: Final Projects Discussion + Q&A: lightning talk reflection question prompt Project work time |
| Week 9 | Polishing Presentations |
Learning Objectives: -Explain how social media analytics is used in business and provide at least one example -Describe a social media analytics report, its purpose, and how it can transform marketing strategies -Prepare a social media report based on final project | Homework Due Read at least one article on business applications of social media analytics and prepare notes to share with the class Discussion Post: reflection on the connections between social media analytics and social media’s role in political conflict based on the lighting presentations from weeks 7 & 8 (recordings will be available for review) (Before lab) Prepare final project presentation draft Whole Class Activities Crowdsource: Social media analytics: meaning, what’s new for 2022/2023, tools available to run analytics, tips to prepare reports, how-to articles, and examples Workshop: Preparing a social media analytics report Application: Prepare a quick report based on a template using the final project findings Small Groups: Create mindmaps of connections between business and marketing analytics, the social and political issues we’ve discussed, power, silences, and ethics. Whole class: share snapshots of mindmaps from groups Lab Practice presentations and feedback |
| Week 10 | Presenting Results |
Learning Objectives: -Polish visualizations and analysis to communicate key findings clearly -Craft a presentation of insights that is appropriate for a specified target audience (client, boss, politicians, general public, etc.) | Homework Due Final team project Whole Class Activities Workshop: polishing visualizations and analysis for presentation Peer Review: Groups swap analysis sections of final projects and provide feedback for each other based on the final project and writing rubrics Class closing activities Lab Final project presentations |
| FINALS WEEK | |
| Individual Reflections Due |