Antaki, Charles, Michael Billig, Derek Edwards, and Jonathan Potter. 2003. “Discourse Analysis Means Doing Analysis: A Critique of Six Analytic Shortcomings.” Discourse Analysis Online 1(1):1-24.
Discourse analysis is falling short because people are doing talk and textual analysis with too much summary, opinion, over/isolated quotation, circular constructs, false survey, or simply reviewing featured concepts. Similarities to conversation analysis, critical discourse analysis, post-structural and Foucauldian concepts. So, to analyze data, you can’t just present data, but you have to break it down, adding information, instead of simply compounding or “reading” information.
“The analyst in the summary might be drawing attention to certain themes, pointing to some things that the participant(s) said, and not to other things. However, this pointing out is not discourse analysis. It might prepare the way for analysis, but it does not provide it. It can impede analysis, if it distorts the original by presenting the speaker as being more consistent, smoother and briefer than they might have been. And it will distort if it is freighted with heavy implication: if the summary attributes beliefs, policies and so on to the speaker as a short-hand, then it risks changing the object of analysis even before the analysis starts in earnest” (9). Offering the addition of personal, moral, or political stances on the information reviewed is not discourse analysis – though quotation can work to offer voice to marginalized groups, this is not the same as analyzing their statements. Too many quotes (or too few) can create interruption (or lack of cohesion) in discursive context. One really shouldn’t allow a statement to “speak for itself” as self-evident, despite the “illumination” the segment is thought to offer. The author’s presumptions on the formation of discourse that may or may not be well-represented by offered texts – that is, would someone else interpreting your data come up with similar results? To better support these claims, look to additional sites of discourse outside of the primary interpretation, as well as focusing on the contextual origins and development of these modes. Speakers’ “thinking” and “feeling” statements should not be presumed to disclose internal ideas or processes – these may be malleable based upon contexts. Be careful of the generalizeability of speakers’ statements, lest you cast “false survey” of other related groups – attributing discourses to the wider population. Don’t just describe the topography of your data – take a look at both the mountain range, its rocks, and how they seemingly fit together. “Good analysis always moves convincingly back and forth between the general and the specific […] analysis means a close engagement with one’s text or transcripts, and the illumination of their meaning and significance through insightful and technically sophisticated work.” (17).