When "Biased" Becomes a Dead End: Sourcing in the Ratification Debate
Eighth-grade U.S. history, 26 students, five days before the unit test on ratification. Students read paired excerpts from Federalist No. 10 and Brutus I. During seminar, 19 of 26 write some version of "these sources are biased, so we can't use them as evidence," and two students who did cite the sources were told by peers they were "taking sides." The team-written unit test asks students to "describe two arguments for ratification and two against," scored with a 4-point rubric for accuracy of description. You have one 50-minute period before the test, and the team has agreed not to change the test's format or items before administration.
1. Name the misconception operating here and describe a 20-minute activity that converts it into a reasoning skill rather than a vocabulary correction.
Read the worked answer
The misconception is epistemic, not factual: students treat a source's perspective as a disqualifier instead of as information about what the source can prove. Telling them "every source is biased" will not move them, because they will simply conclude that all evidence is unusable.
A 20-minute sequence:
- Predict (5 min). Post two claims - "A large republic endangers liberty" and "A bill of rights is unnecessary" - and ask pairs to predict which author would endorse each and which would reject it, with one quotation as support. This forces students to use perspective as a tool.
- Two-column chart (8 min). "What this writer wants readers to believe" | "What evidence he would accept as proof." The second column is the key move: it asks what would count as evidence for this author, which cannot be answered by dismissing him.
- Trade and extend (7 min). Pairs swap charts and must add one sentence naming what the author would have to concede for the other side to win.
Close by naming the transferable move: a source's bias tells you what it is evidence of - a position held, an argument in circulation - not whether it is evidence at all. Peer policing of "taking sides" weakens once using a source's position is the required move rather than a choice.
2. The test format is fixed. How do you still make student reasoning visible and countable without violating the team agreement?
Read the worked answer
The constraint is on the test, not on what you collect. Three moves, in order of value:
- Add a five-minute pre-test prompt outside the test itself, e.g., "Here are two one-sentence quotations. Which ratification argument does each support, and what would strengthen it?" Score it on a separate reasoning scale and do not fold it into the test grade. This gives you data on the reasoning objective without touching the graded instrument.
- Read the multiple-choice/description results diagnostically. After scoring, sort students into those who can name arguments and those who can weigh them. The rubric rewards accurate description, so a student can score 4/4 while still holding the "bias equals unusable" belief; the sort tells you which students need the next cycle of work.
- Make the rubric's own limit explicit to students. If only accuracy is scored, students will infer that accuracy is all that matters, and the peer pressure to avoid "sides" will return. Say plainly: the test measures recall of arguments; the seminar measures judgment about evidence; both are the target, only one is on the test.
Do not silently re-purpose the test items into an evidence-evaluation task. That breaks the team agreement, and it corrupts the data the team is using to compare classes.