This has been my second time (we end in June). Here is how I would do it in the future.
1) Spend 2-3 class sessions on an accessible book that covers much of a (traditional) subfield. Use class to discuss the books the way my high school English teachers did.
2) Have each student choose a theme and write a paragraph on it. Call this the “commitment paragraph.”
3) Group students by some reasonable subfield division. These become “workshops.”
4) Proceed through my existing scaffolding. Workshop that work 1x per week.
5) Decide whether students should write a full paper, just the first 2/3, or just a literature review. This will depend on student energy and whether I have 10 or 14 weeks.
6) Students may final papers solo or in groups of not more than three. That decision is made in the middle of the term, and students sign contracts.
This differs in three ways from what I am doing now. First, it uses a ‘classic text’ to familiarize a basic vocabulary. Now I have each student choose a top-three journal article. That is too broad.
Then it replaces ad hoc groups with standing “workshops.” This might build rapport and lead to cross-fertilization.
Third, it introduces optional coauthorship.
Research design thus taught will reflect the instructor’s specialization(s). This is an asset. The course can rotate.