Analysis and Visualization Considerations for Quantitative Social Science Research Using Social Media Data

Abstract

This is the fifth in a series of white papers providing a summary of the discussions and future directions that are derived from these topical meetings. This paper focuses on issues related to analysis and visual analytics. While these two topics are distinct, there are clear overlaps between the two. It is common to use different visualizations during analysis and given the sheer volume of social media data, visual analytic tools can be important during analysis, as well as during other parts of the research lifecycle. Choices about analysis may be informed by visualization plans and vice versa - both are key in communicating about a data set and what it means. We also recognized that each field of research has different analysis techniques and different levels of familiarity with visual analytics. Putting these two topics into the same meeting provided us with the opportunity to think about analysis and visual analytics/visualization in new, synergistic ways.

My Data Are Ready, How Do I Analyze Them: Navigating Data Analysis in Social Science Research

“…Data visualization and communication Data visualization and communication are crucial in social science research, particularly in analyzing social media data (Jensen et al, 2021). Reyes (2022) emphasizes the importance of combining creativity and technique in producing compelling visualizations, providing a practical guide for researchers.…”

Social Media Data for Firearms Research: Promise and Perils

“…Recent years have witnessed increasing exploration of the intersection between social media and public health. Following Jensen et al (2021), we describe two broad ways that social media data have been used in public health contexts: informing and understanding responses to unplanned and planned public health events and assessing health-related conditions and behaviors.…”

“…Unexpected, impactful events can be thought of as “system shocks” (Jensen et al 2021); while planned events, which may be likewise impactful, are, by definition, known in advance. In the public health context, the former includes events such as pandemics, natural disasters, mass shootings, or wars—all of which have consequences for human health even if the events themselves are not precipitated by health-related factors (Nelson et al 2007).…”

“…Measurement construction is another important consideration in research applications of social media data. While survey data represent sample participants’ responses to intentionally designed, carefully worded, and deliberately ordered questions, social media data, by contrast, represent less structured, generally unsolicited data that can take many forms ranging from a written (narrative) post that shares a feeling, belief, opinion, or perspective to shared pictures or videos, a link to external content, or a reaction or resharing of another’s posted content (Jensen et al 2021). Much (though not all) of the work to develop valid and reliable measures in surveys happens in the questionnaire design process before data are collected; with social media data, the measure development work happens after individuals have posted to social media, that is, after the data generation process.…”