Coding Streams of Language: Techniques for the Systematic Coding of Text, Talk, and Other Verbal Data

Untangling Methodological Commitments in Writing Research: Using Collaborative Secondary Data Analysis to Maximize Interpretive Potentials of Qualitative Data

“…Taking a postpositivist approach, Madden collected survey data to look for patterns across a larger sample of doctoral student writers, while Tarabochia took a constructivist approach, conducting interviews with nine faculty writers to learn more about the experiences of particular people in context. This seeming epistemological misalignment meant that we could not make sense of our findings the way Geisler (2004) and Geisler and Swarts (2019) suggest. That is, we could not code segments, calculate frequencies of each code, and argue that the most frequent concerns were generalizable to the experiences of all doctoral student or faculty writers represented in these studies.…”

“…Given ethical and research efficacy questions that arise in SDA, it is important to detail the original or parent studies, methodological approaches, and data collection procedures (Heaton, 2008; Thorne, 1994). In that view, we explain the two parent studies and how we combined them by critically adapting and integrating established procedures for qualitative research offered by Geisler (2004), Geisler and Swarts (2019), Smagorinsky (2008), and Saldaña (2016) before going on to explain how our work responds to ethical and practical issues that arise in SDA.…”

“…Once we received IRB approval and grant funding, we combined our data sets, which comprised responses from 195 doctoral student writers to five qualitative survey questions and transcripts from hour-long interviews with nine faculty writers. To prepare the overall data set so that different data types (qualitative survey questions and interview transcripts) could be analyzed together, we turned to Geisler’s (2004) process of verbal data analysis because both data sets were types of verbal data—what Geisler (2004) defines as streams of language embedded and made meaningful in particular contexts (Geisler & Swarts, 2019). Similar to Geisler (2004) and Geisler and Swarts (2019), we assume that streams of language, what our participants said or failed to say, would tell us something about underlying phenomena, in this case their writing-related concerns.…”

“…To prepare the overall data set so that different data types (qualitative survey questions and interview transcripts) could be analyzed together, we turned to Geisler’s (2004) process of verbal data analysis because both data sets were types of verbal data—what Geisler (2004) defines as streams of language embedded and made meaningful in particular contexts (Geisler & Swarts, 2019). Similar to Geisler (2004) and Geisler and Swarts (2019), we assume that streams of language, what our participants said or failed to say, would tell us something about underlying phenomena, in this case their writing-related concerns. Following Geisler’s method, we segmented our data by topical chain using markers such as pronouns, demonstratives, and definite articles to determine the boundaries of a topic within larger streams of language.…”

“…We coded for two dimensions. For each dimension, we used what Geisler (2004) and Geisler and Swarts (2019) call “exclusive coding,” assigning one code to each verbal data segment (p. 12). The first dimension we coded for was “concern,” which we defined broadly as areas of interest or focus for writers (Tarabochia & Madden, 2018).…”