An argument in research is a unique kind of argument. However, what keeps it similar to other types of arguments is the need for some sort of conflict. When structuring an argument in research, it's helpful to think of a conversation. A researcher starts with a claim, then searches for reasons behind those claims, asks for evidence, looks for responses, and explores warrants. I like the concrete organizational pattern the book expressed in the results of research discussion—i.e., the results of questions are answers, the results of problems are solutions, and the results of arguments are claims. Language in research of any kind is important and not to be overlooked. Without language, lucid theory and conversation are stunted. For example, when analyzing the relationship between claims, reasons, and evidence, it's apt to say:
I have this claim BECAUSE OF these reasons BASED ON this evidence.
Those connecting phrases clarify the definition of reasons compared to evidence. To organize reasons, storyboarding adds logical flow to the argument. A piece of evidence must support each reason. Specificity is key when crafting a claim. In practice, I think this also helps focus the researcher from going down too many rabbit holes of data.
Not to be forgotten at the end of the research argument conversation, a warrant is the logical backing to your reasons. A claim and its associated reasons can be disputed on the basis of relevance. Warrants give the arguer a chance to explain the relevance between a claim and its reasons, often using specific instances that speak to a larger phenomenon.
The discussion of ethos "thickening" arguments is important because it forces the arguer to think critically about the way they're presenting their argument, rather than just the argument itself. I think a key component of this is to not misrepresent your evidence. As a researcher, we're often including reports of evidence, not primary evidence itself. Because we bridge the gap between the primary evidence and the report, we need to be extra careful in how we are framing the reports in our argument.
I feel as if it depends on what field a researcher is practicing data visualization in to determine what type of problem they're going to more frequently pose. I think, in our class's case, conceptual problems, like researchers in general, are more often used. However, I think in data journalism, practical problems with the solution of of action are more often used since the umbrella discipline (journalism) is quite different than academic research.