While the good genealogist does that reasonably exhaustive search in reliable sources, noting full and accurate source citations, he begins the analysis and correlation of the information collected to assess its quality as evidence. There are numerous ways to teach this I suppose, but I prefer to approach this stage by asking my students to study
intensely my favorite example of this, and come prepared to discuss the article. I would like to badger them into reading the article four times before we met. Only the learners do.
Here is the article: Thomas W. Jones, "Organizing Meager Evidence to Reveal Lineages: An Irish Example– Geddes of Tyrone", National Genealogical Society Quarterly Vol.89 (June 2001) pp. 98-112
What excites me about this article as a model is that it uses a spreadsheet to put time-lines of the people found side by side for comparison, then with this visual framework reconstitute them into families. The textual account explains his reasoning for linking each fragment into the whole. As they say in England, "Brilliant".
The discussion process using articles has been mentioned several times previously in postings, so I'll leave you to dig that out.
You can probably suggest several learning/teaching activities to follow up on this to reinforce the points.