CLOSED
Using Social Media Data for Research: Potentials and Pitfalls
Workshop Details
Date: November 9-10, 2020 (2-day workshop)
Time: 12 p.m. Turkey local time (10 a.m. Germany local time)
Venue: Online via Zoom
Workshop Language: English
Instructors: Dr. Markus Gamper, Dr. Raphael H. Heiberger
Schedule
09.11.2020, Monday | |
10:00 – 13:00 | Session 1: Introduction and Data Collection |
13:00 – 14:00 | Lunch |
14:00 – 16:45 | Session 2: Data Processing |
16:45 – 17:30 | Session 3: Identifying Errors |
10.11.2020, Tuesday | |
10:00 – 12:30 | Session 4: In depth look at Error Identification and Characterization |
12:30 – 13:30 | Lunch |
13:30 – 15:45 | Session 5: Practical examples of identifying and mitigating error sources in social media research design |
15:45 – 16:30 | Session 6: Case studies 2: Identifying and Mitigating errors |
16:30 – 17:00 | Conclusions & Workshop Closing
Summary of lessons learned, pointers to additional material, open questions.
|
Workshop Description
The activities and interactions of hundreds of millions of people worldwide are recorded as digital traces including social media data. These data offer increasingly comprehensive pictures of both individuals and groups on different platforms, but also allow inferences about broader target populations beyond those platforms. Notwithstanding the many advantages, studying the errors that can occur when digital traces are used to learn about humans and social phenomena is essential. Incidentally, many similar errors also affect survey estimates, which survey designers have been addressing for decades using error conceptualization frameworks, most notably the Total Survey Error Framework (TSE).
In this tutorial, we will introduce the audience to the concepts and guidelines of the TSE and how they are applied by survey practitioners in the social sciences, guided by our interdisciplinary background and experience. Having understood the ‘total error’ perspective towards surveys, we will introduce our own conceptual framework to diagnose, understand, and avoid errors that may occur in studies that are based on digital traces of humans.
To help understand the utility of the error framework for digital traces, we apply it to diagnose and document errors in existing computational social science studies such as Understanding Political Opinion using Twitter and Using Search Queries for Inferring Health Statistics.
We also ask participants to apply the error framework to hypothetical scenarios utilizing novel forms of digital traces like mobility data as well as their own area of research, using social media datasets openly available on the web.
Keywords
quantitative methods, research design, data collection and analysis, complete networks, R
Learning objectives
Prerequisites
Participants should possess a general affinity toward social science theory. Participants should also be open to acquainting themselves with quantitative methods. However, no advanced statistical knowledge is required.