ABSTRACT
Chapter 6: Inferring Individual-Level Characteristics from Digital Trace Data: Issues and Recommendations, Deen Freelon
The purpose of this chapter is to contribute to the development of a framework for assessing the construct validity of theoretical inferences drawn from digital traces. Most high-quality trace-based empirical research does this to some extent, but what is missing is an abstract set of standards and heuristics by which the quality of its inferences may be assessed. This will help ensure the rigor of such research, which is especially important given that it is still in its infancy. This chapter defines four nested, platform-independent domains that researchers should bear in mind when choosing traces for analysis: technical design, terms of service, social context, and the potential for misrepresentation. The chapter continues by demonstrating the value of this framework in discussions of three general categories of techniques for trace inference: direct indication, proper names, and speech patterns. The chapter concludes with an application of the framework to these techniques by drawing examples from three individual-level characteristics of great interest to political communication researchers: gender, race/ethnicity, and geographical location.
